Compare commits

..

350 Commits

Author SHA1 Message Date
Nadav Har'El
afa2c1b0bf materialized_views: propagate "view virtual columns" between nodes
db::schema_tables::ALL and db::schema_tables::all_tables() are both supposed
to list the same schema tables - the former is the list of their names, and
the latter is the list of their schemas. This code duplication makes it easy
to forget to update one of them, and indeed recently the new
"view_virtual_columns" was added to all_tables() but not to ALL.

What this patch does is to make ALL a function instead of constant vector.
The newly named all_table_names() function uses all_tables() so the list
of schema tables only appears once.

So that nobody worries about the performance impact, all_table_names()
caches the list in a per-thread vector that is only prepared once per thread.

Because after this patch all_table_names() has the "view_virtual_columns"
that was previously missing, this patch also fixes #4339, which was about
virtual columns in materialized views not being propagated to other nodes.

Unfortunately, to test the fix for #4339 we need a test with multiple
nodes, so we cannot test it here in a unit test, and will instead use
the dtest framework, in a separate patch.

Fixes #4339

Branches: 3.0
Tests: all unit tests (release and debug mode), new dtest for #4339. The unit test mutation_reader_test failed in debug mode but not in release mode, but this probably has nothing to do with this patch (?).

Signed-off-by: Nadav Har'El <nyh@scylladb.com>

Message-Id: <20190320063437.32731-1-nyh@scylladb.com>
(cherry picked from commit 7c874057f5)
2020-01-06 00:37:59 +02:00
Tomasz Grabiec
ad70fe8503 cql: alter type: Format field name as text instead of hex
Fixes #4841

Message-Id: <1565702635-26214-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit 64ff1b6405)
2020-01-05 18:55:40 +02:00
Gleb Natapov
3cd9c78056 cache_hitrate_calculator: do not ignore a future returned from gossiper::add_local_application_state
We should wait for a future returned from add_local_application_state() to
resolve before issuing new calculation, otherwise two
add_local_application_state() may run simultaneously for the same state.

Fixes #4838.

Message-Id: <20190812082158.GE17984@scylladb.com>
(cherry picked from commit 00c4078af3)
2020-01-05 18:50:27 +02:00
Benny Halevy
c5e5ed2775 tracing: one_session_records: keep local tracing ptr
Similar to trace_state keep shared_ptr<tracing> _local_tracing_ptr
in one_session_records when constructed so it can be used
during shutdown.

Fixes #5243

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
(cherry picked from commit 7aef39e400)
2019-12-24 18:42:33 +02:00
Tomasz Grabiec
666266c3cf types: Fix abort on type alter which affects a compact storage table with no regular columns
Fixes #4837

Message-Id: <1565702247-23800-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit 34cff6ed6b)
2019-12-24 17:44:40 +02:00
Dejan Mircevski
19b5d70338 tests: Add cquery_nofail() utility
Most tests await the result of cql_test_env::execute_cql().  Most
would also benefit from reporting errors with top-level location
included.

Ref #4837 (a prerequisite for backporting)

Signed-off-by: Dejan Mircevski <dejan@scylladb.com>
(cherry picked from commit a9849ecba7)
2019-12-24 17:44:40 +02:00
Amnon Heiman
b3cdee7e27 init: do not allow replace-address for seeds
If a node is a seed node, it can not be started with
replace-address-first-boot or the replace-address flag.

The issue is that as a seed node it will generate new tokens instead of
replacing the existing one the user expect it to replaec when supplying
the flags.

This patch will throw a bad_configuration_error exception
in this case.

Fixes #3889

Signed-off-by: Amnon Heiman <amnon@scylladb.com>
(cherry picked from commit 399d79fc6f)
2019-12-23 17:24:52 +02:00
Rafael Ávila de Espíndola
4c42f18d82 cql: Fix use of UDT in reversed columns
We were missing calls to underlying_type in a few locations and so the
insert would think the given literal was invalid and the select would
refuse to fetch a UDT field.

Fixes #4672

Signed-off-by: Rafael Ávila de Espíndola <espindola@scylladb.com>
Message-Id: <20190708200516.59841-1-espindola@scylladb.com>
(cherry picked from commit 4e7ffb80c0)
2019-12-23 15:57:47 +02:00
Benny Halevy
ea8f8ab7a3 sstables: mc: prevent signed integer overflow
Fix runtime error: signed integer overflow
introduced by 2dc3776407

Delta-encoded values may wrap around if the encoded value is
less than the base value.  This could happen in two places:
In the mc-format serialization header itself, where the base values are implicit
Cassandra epoch time, and in the sstables data files, where the base values
are taken from the encoding_stats (later written to the serialization_header).

In these cases, when the calculation is done using signed integer/long we may see
"runtime error: signed integer overflow" messages in debug mode
(with -fsanitize=undefined / -fsanitize=signed-integer-overflow).

Overflow here is expected and harmless since we do not gurantee that
neither the base values in the serialization header are greater than
or equal to Cassandra's epoch now that the delta-encoded values are
always greater than or equal to the respective base values in
the serialization header.

To prevent these warnings, the subtraction/addition should be done with unsigned
(two's complement) arithmetic and the result converted to the signed type.

Note that to keep the code simple where possible, when also rely on implicit
conversion of signed integers to unsigned when either one of added value is unsigned
and the other is signed.

Fixes: #4098

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20190120142950.15776-1-bhalevy@scylladb.com>
(cherry picked from commit 844a2de263)
2019-12-15 15:52:35 +02:00
Piotr Sarna
db6821ce8f table: Reduce read amplification in view update generation
This commit makes sure that single-partition readers for
read-before-write do not have fast-forwarding enabled,
as it may lead to huge read amplification. The observed case was:
1. Creating an index.
  CREATE INDEX index1  ON myks2.standard1 ("C1");
2. Running cassandra-stress in order to generate view updates.
cassandra-stress write no-warmup n=1000000 cl=ONE -schema \
  'replication(factor=2) compaction(strategy=LeveledCompactionStrategy)' \
  keyspace=myks2 -pop seq=4000000..8000000 -rate threads=100 -errors
  skip-read-validation -node 127.0.0.1;

Without disabling fast-forwarding, single-partition readers
were turned into scanning readers in cache, which resulted
in reading 36GB (sic!) on a workload which generates less
than 1GB of view updates. After applying the fix, the number
dropped down to less than 1GB, as expected.

Refs #5409
Fixes #4615
Fixes #5418

(cherry picked from commit 79c3a508f4)
2019-12-05 22:36:41 +02:00
Rafael Ávila de Espíndola
3c91bad0dc commitlog: make sure a file is closed
If allocate or truncate throws, we have to close the file.

Fixes #4877

Signed-off-by: Rafael Ávila de Espíndola <espindola@scylladb.com>
Message-Id: <20191114174810.49004-1-espindola@scylladb.com>
(cherry picked from commit 6160b9017d)
2019-11-24 17:50:06 +02:00
Tomasz Grabiec
bbe41a82be row_cache: Fix abort on bad_alloc during cache update
Since 90d6c0b, cache will abort when trying to detach partition
entries while they're updated. This should never happen. It can happen
though, when the update fails on bad_alloc, because the cleanup guard
invalidates the cache before it releases partition snapshots (held by
"update" coroutine).

Fix by destroying the coroutine first.

Fixes #5327.

Tests:
  - row_cache_test (dev)

Message-Id: <1574360259-10132-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit e3d025d014)
2019-11-24 17:44:30 +02:00
Nadav Har'El
6fb42269e9 merge: row_marker: correct row expiry condition
Merged patch set by Piotr Dulikowski:

This change corrects condition on which a row was considered expired by its
TTL.

The logic that decides when a row becomes expired was inconsistent with the
logic that decides if a single cell is expired. A single cell becomes expired
when expiry_timestamp <= now, while a row became expired when
expiry_timestamp < now (notice the strict inequality). For rows inserted
with TTL, this caused non-key cells to expire (change their values to null)
one second before the row disappeared. Now, row expiry logic uses non-strict
inequality.

Fixes #4263,
Fixes #5290.

Tests:

    unit(dev)
    python test described in issue #5290

(cherry picked from commit 9b9609c65b)
(cherry picked from commit 95acf71680)
2019-11-20 21:40:40 +02:00
Asias He
ee2255a189 gossip: Fix max generation drift measure
Assume n1 and n2 in a cluster with generation number g1, g2. The
cluster runs for more than 1 year (MAX_GENERATION_DIFFERENCE). When n1
reboots with generation g1' which is time based, n2 will see
g1' > g2 + MAX_GENERATION_DIFFERENCE and reject n1's gossip update.

To fix, check the generation drift with generation value this node would
get if this node were restarted.

This is a backport of CASSANDRA-10969.

Fixes #5164

(cherry picked from commit 0a52ecb6df)
2019-11-20 11:39:37 +02:00
Kamil Braun
3218e6cd4c view: fix bug in virtual columns.
When creating a virtual column of non-frozen map type,
the wrong type was used for the map's keys.

Fixes #5165.

(cherry picked from commit ef9d5750c8)
2019-11-19 11:17:54 +02:00
Rafael Ávila de Espíndola
1d94aac551 sstable: close file_writer if an exception in thrown
The previous code was not exception safe and would eventually cause a
file to be destroyed without being closed, causing an assert failure.

Unfortunately it doesn't seem to be possible to test this without
error injection, since using an invalid directory fails before this
code is executed.

Fixes #4948

Signed-off-by: Rafael Ávila de Espíndola <espindola@scylladb.com>
Message-Id: <20190904002314.79591-1-espindola@scylladb.com>
(cherry picked from commit 000514e7cc)
2019-11-19 11:17:54 +02:00
Avi Kivity
2e5110d063 reconcilable_result: use chunked_vector to hold partitions
Usually, a reconcilable_result holds very few partitions (1 is common),
since the page size is limited by 1MB. But if we have paging disabled or
if we are reconciling a range full of tombstones, we may see many more.
This can cause large allocations.

Change to chunked_vector to prevent those large allocations, as they
can be quite expensive.

Fixes #4780.

(cherry picked from commit 093d2cd7e5)
2019-11-19 11:17:54 +02:00
Avi Kivity
e4bb7ce73c utils::chunked_vector: add rbegin() and related iterators
Needed as an std::vector replacement.

(cherry picked from commit eaa9a5b0d7)

Prerequisite for #4780.
2019-11-19 11:17:54 +02:00
Avi Kivity
ecc54c1a68 utils: chunked_vector: make begin()/end() const correct
begin() of a const vector should return a const_iterator, to avoid
giving the caller the ability to mutate it.

This slipped through since iterator's constructor does a const_cast.

Noticed by code inspection.

(cherry picked from commit df6faae980)

Prerequisite for #4780.
2019-11-19 11:17:54 +02:00
Glauber Costa
71cfd108c6 do not crash in user-defined operations if the controller is disabled
Scylla currently crashes if we run manual operations like nodetool
compact with the controller disabled. While we neither like nor
recommend running with the controller disabled, due to some corner cases
in the controller algorithm we are not yet at the point in which we can
deprecate this and are sometimes forced to disable it.

The reason for the crash is that manual operations will invoke
_backlog_of_shares, which returns what is the backlog needed to
create a certain number of shares. That scan the existing control
points, but when we run without the controller there are no control
points and we crash.

Backlog doesn't matter if the controller is disabled, and the return
value of this function will be immaterial in this case. So to avoid the
crash, we return something right away if the controller is disabled.

Fixes #5016

Signed-off-by: Glauber Costa <glauber@scylladb.com>
(cherry picked from commit c9f2d1d105)
2019-11-19 11:17:54 +02:00
Avi Kivity
d40a7a5e9e Merge "Add proper aggregation for paged indexing" from Piotr
"
Fixes #4540

This series adds proper handling of aggregation for paged indexed queries.
Before this series returned results were presented to the user in per-page
partial manner, while they should have been returned as a single aggregated
value.

Tests: unit(dev)
"

* 'add_proper_aggregation_for_paged_indexing_for_3.0' of https://github.com/psarna/scylla:
  test: add 'eventually' block to index paging test
  tests: add indexing+paging test case for clustering keys
  tests: add indexing + paging + aggregation test case
  cql3: make DEFAULT_COUNT_PAGE_SIZE constant public
  cql3: add proper aggregation to paged indexing
  cql3: add a query options constructor with explicit page size
  cql3: enable explicit copying of query_options
  cql3: split execute_base_query implementation
2019-11-19 11:17:54 +02:00
Takuya ASADA
a163d245ec dist/common/scripts/scylla_setup: don't proceed with empty NIC name
Currently NIC selection prompt on scylla_setup just proceed setup when
user just pressed Enter key on the prompt.
The prompt should ask NIC name again until user input correct NIC name.

Fixes #4517
Message-Id: <20190617124925.11559-1-syuu@scylladb.com>

(cherry picked from commit 7320c966bc)
2019-11-19 11:17:54 +02:00
Piotr Sarna
045831b706 test: add 'eventually' block to index paging test
Without 'eventually', the test is flaky because the index can still
be not up to date while checking its conditions.

Fixes #4670

(cherry picked from commit ebbe038d19)
2019-11-15 09:15:29 +01:00
Piotr Sarna
148245ab6a tests: add indexing+paging test case for clustering keys
Indexing a non-prefix part of the clustering key has a separate
code path (see issue #3405), so it deserves a separate test case.
2019-11-14 12:32:08 +01:00
Piotr Sarna
bbe5de1403 tests: add indexing + paging + aggregation test case
Indexed queries used to erroneously return partial per-page results
for aggregation queries. This test case used to reproduce the problem
and now ensures that there would be no regressions.

Refs #4540
2019-11-14 12:32:07 +01:00
Piotr Sarna
ca0df416c0 cql3: make DEFAULT_COUNT_PAGE_SIZE constant public
The constant will be later used in test scenarios.
2019-11-14 12:25:37 +01:00
Piotr Sarna
37ed60374e cql3: add proper aggregation to paged indexing
Aggregated and paged filtering needs to aggregate the results
from all pages in order to avoid returning partial per-page
results. It's a little bit more complicated than regular aggregation,
because each paging state needs to be translated between the base
table and the underlying view. The routine keeps fetching pages
from the underlying view, which are then used to fetch base rows,
which go straight to the result set builder.

Fixes #4540
2019-11-14 12:25:37 +01:00
Piotr Sarna
7c991a276b cql3: add a query options constructor with explicit page size
For internal use, there already exists a query_options constructor
that copies data from another query_options with overwritten paging
state. This commit adds an option to overwrite page size as well.
2019-11-14 10:49:28 +01:00
Piotr Sarna
72e039be85 cql3: enable explicit copying of query_options 2019-11-14 10:49:28 +01:00
Piotr Sarna
a28ecc4714 cql3: split execute_base_query implementation
In order to handle aggregation queries correctly, the function that
returns base query results is split into two, so it's possible to
access raw query results, before they're converted into end-user
CQL message.
2019-11-14 10:49:28 +01:00
Avi Kivity
584c555698 Update seastar submodule
* seastar 3920dcb3f8...083dc0875e (2):
  > core: fix a race in execution stages
  > execution_stage: prevent unbounded growth

Fixes #4749.
Fixes #4856.
2019-11-13 13:15:54 +02:00
null
e772f11ee0 release: prepare for3.0.11 by yaronkaikov 2019-10-30 11:01:40 +02:00
Botond Dénes
d79b6a7481 repair: repair_cf_range(): extract result of local checksum calculation only once
The loop that collects the result of the checksum calculations and logs
any errors. The error logging includes `checksums[0]` which corresponds
to the checksum calculation on the local node. This violates the
assumption of the code following the loop, which assumes that the future
of `checksums[0]` is intact after the loop terminates. However this is
only true when the checksum calculation is successful and is false when
it fails, as in this case the loop extracts the error and logs it. When
the code after the loop checks again whether said calculation failed, it
will get a false negative and will go ahead and attempt to extract the
value, triggering an assert failure.
Fix by making sure that even in the case of failed checksum calculation,
the result of `checksum[0]` is extracted only once.

Fixes: #5238
Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <20191029151709.90986-1-bdenes@scylladb.com>
(cherry picked from commit e48f301e95)
2019-10-29 20:43:50 +02:00
Avi Kivity
85168c500c Merge "Fix handling of schema alters and eviction in cache" from Tomasz
"
Fixes #5134, Eviction concurrent with preempted partition entry update after
  memtable flush may allow stale data to be populated into cache.

Fixes #5135, Cache reads may miss some writes if schema alter followed by a
  read happened concurrently with preempted partition entry update.

Fixes #5127, Cache populating read concurrent with schema alter may use the
  wrong schema version to interpret sstable data.

Fixes #5128, Reads of multi-row partitions concurrent with memtable flush may
  fail or cause a node crash after schema alter.
"

* tag 'fix-cache-issues-with-schema-alter-and-eviction-v2' of github.com:tgrabiec/scylla:
  tests: row_cache: Introduce test_alter_then_preempted_update_then_memtable_read
  tests: row_cache_stress_test: Verify all entries are evictable at the end
  tests: row_cache_stress_test: Exercise single-partition reads
  tests: row_cache_stress_test: Add periodic schema alters
  tests: memtable_snapshot_source: Allow changing the schema
  tests: simple_schema: Prepare for schema altering
  row_cache: Record upgraded schema in memtable entries during update
  memtable: Extract memtable_entry::upgrade_schema()
  row_cache, mvcc: Prevent locked snapshots from being evicted
  row_cache: Make evict() not use invalidate_unwrapped()
  mvcc: Introduce partition_snapshot::touch()
  row_cache, mvcc: Do not upgrade schema of entries which are being updated
  row_cache: Use the correct schema version to populate the partition entry
  delegating_reader: Optimize fill_buffer()
  row_cache, memtable: Use upgrade_schema()
  flat_mutation_reader: Introduce upgrade_schema()

(cherry picked from commit 8ed6f94a16)
(cherry picked from commit 3f4d9f210f)
2019-10-22 19:47:02 +02:00
Botond Dénes
5b9e2cd6e6 querier_cache: correctly account entries evicted on insertion in the population
Currently, the population stat is not increased for entries that are
evicted immediately on insert, however the code that does the eviction
still decreases the population stat, leading to an imbalance and in some
cases the underflow of the population stat. To fix, unconditionally
increase the population stat upon inserting an entry, regardless of
whether it is immediately evicted or not.

Fixes: #5123

Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <20191001153215.82997-1-bdenes@scylladb.com>
(cherry picked from commit 00b432b61d)
2019-10-05 12:36:34 +03:00
Avi Kivity
77f33ca106 Merge " hinted handoff: fix races during shutdown and draining" from Vlad
"
Fix races that may lead to use-after-free events and file system level exceptions
during shutdown and drain.

The root cause of use-after-free events in question is that space_watchdog blocks on
end_point_hints_manager::file_update_mutex() and we need to make sure this mutex is alive as long as
it's accessed even if the corresponding end_point_hints_manager instance
is destroyed in the context of manager::drain_for().

File system exceptions may occur when space_watchdog attempts to scan a
directory while it's being deleted from the drain_for() context.
In case of such an exception new hints generation is going to be blocked
- including for materialized views, till the next space_watchdog round (in 1s).

Issues that are fixed are #4685 and #4836.

Tested as follows:
 1) Patched the code in order to trigger the race with (a lot) higher
    probability and running slightly modified hinted handoff replace
    dtest with a debug binary for 100 times. Side effect of this
    testing was discovering of #4836.
 2) Using the same patch as above tested that there are no crashes and
    nodes survive stop/start sequences (they were not without this series)
    in the context of all hinted handoff dtests. Ran the whole set of
    tests with dev binary for 10 times.
"

Fixes #4685
Fixes #4836.

* 'hinted_handoff_race_between_drain_for_and_space_watchdog_no_global_lock-v2' of https://github.com/vladzcloudius/scylla:
  hinted handoff: fix a race on a directory removal between space_watchdog and drain_for()
  hinted handoff: make taking file_update_mutex safe
  db::hints::manager::drain_for(): fix alignment
  db::hints::manager: serialize calls to drain_for()
  db::hints: cosmetics: identation and missing method qualifier

(cherry picked from commit 3cb081eb84)
2019-10-05 12:25:51 +03:00
Gleb Natapov
93760f13ee messaging_service: enable reuseaddr on messaging service rpc
Fixes #4943

Message-Id: <20190918152405.GV21540@scylladb.com>
(cherry picked from commit 73e3d0a283)
2019-10-03 15:24:53 +03:00
Avi Kivity
e597ae1176 Update seastar submodule
* seastar af3fc691b9...3920dcb3f8 (2):
  > net: socket::{set,get}_reuseaddr() should not be virtual
  > Merge "fix some tcp connection bugs and add reuseaddr option to a client socket" from Gleb

Prerequisite for #4943.
2019-10-03 15:23:35 +03:00
Tomasz Grabiec
79c7015cce Merge "hinted handoff: don't reuse_segments and discard corrupted segments" from Vlad
This series addresses two issues in the hinted handoff that should
complete fixing the infamous #4231.

In particular the second patch removes the requirement to manually
delete hints files after upgrading to 3.0.4.

Tested with manual unit testing.

* https://github.com/vladzcloudius/scylla.git hinted_handoff_drop_broken_segments-v3:
  hinted handoff: disable "reuse_segments"
  commitlog: introduce a segment_error
  hinted handoff: discard corrupted segments

(cherry picked from commit ac0d435c3e)
2019-09-28 19:52:57 +03:00
Asias He
00a14000cd storage_service: Replicate and advertise tokens early in the boot up process
When a node is restarted, there is a race between gossip starts (other
nodes will mark this node up again and send requests) and the tokens are
replicated to other shards. Here is an example:

- n1, n2
- n2 is down, n1 think n2 is down
- n2 starts again, n2 starts gossip service, n1 thinks n2 is up and sends
  reads/writes to n2, but n2 hasn't replicated the token_metadata to all
  the shards.
- n2 complains:
  token_metadata - sorted_tokens is empty in first_token_index!
  token_metadata - sorted_tokens is empty in first_token_index!
  token_metadata - sorted_tokens is empty in first_token_index!
  token_metadata - sorted_tokens is empty in first_token_index!
  token_metadata - sorted_tokens is empty in first_token_index!
  token_metadata - sorted_tokens is empty in first_token_index!
  storage_proxy - Failed to apply mutation from $ip#4: std::runtime_error
  (sorted_tokens is empty in first_token_index!)

The code path looks like below:

0 stoarge_service::init_server
1    prepare_to_join()
2          add gossip application state of NET_VERSION, SCHEMA and so on.
3         _gossiper.start_gossiping().get()
4    join_token_ring()
5           _token_metadata.update_normal_tokens(tokens, get_broadcast_address());
6           replicate_to_all_cores().get()
7           storage_service::set_gossip_tokens() which adds the gossip application state of TOKENS and STATUS

The race talked above is at line 3 and line 6.

To fix, we can replicate the token_metadata early after it is filled
with the tokens read from system table before gossip starts. So that
when other nodes think this restarting node is up, the tokens are
already replicated to all the shards.

In addition, this patch also fixes the issue that other nodes might see
a node miss the TOKENS and STATUS application state in gossip if that
node failed in the middle of a restarting process, i.e., it is killed
after line 3 and before line 7. As a result we could not replace the
node.

Tests: update_cluster_layout_tests.py
Fixes: #4709
Fixes: #4723
(cherry picked from commit 3b39a59135)
2019-09-22 12:46:36 +03:00
Avi Kivity
1c40a0fcd2 Update seastar submodule
* seastar ea859b5840...af3fc691b9 (1):
  > iotune: fix exception handling in case test file creation fails

Fixes #5001.
2019-09-18 18:37:23 +03:00
Gleb Natapov
e10735852b messaging_service: configure different streaming domain for each rpc server
A streaming domain identifies a server across shards. Each server should
have different one.

Fixes: #4953

Message-Id: <20190908085327.GR21540@scylladb.com>
(cherry picked from commit 9e9f64d90e)
2019-09-09 20:37:40 +03:00
Avi Kivity
42433a25a8 Update seastar submodule
* seastar 445b5126c2...ea859b5840 (1):
  > perftune: fix missing import for logging

Fixes #4958.
2019-09-04 13:50:29 +03:00
Paweł Dziepak
d04d3fa653 mutation_partition: verify row::append_cell() precondition
row::append_cell() has a precondition that the new cell column id needs
to be larger than that of any other already existing cell. If this
precondition is violated the row will end up in an invalid state. This
patch adds assertion to make sure we fail early in such cases.

(cherry picked from commit 060e3f8ac2)
2019-08-23 15:06:18 +02:00
Avi Kivity
1bcc5a1b5c Merge "database: assign proper io priority for streaming view updates" from Piotr
"
Streamed view updates parasitized on writing io priority, which is
reserved for user writes - it's now properly bound to streaming
write priority.

Verified manually by checking appropriate io metrics: scylla_io_queue_total_bytes{class="streaming_write" ...} vs scylla_io_queue_total_bytes{class="query" ...}

Tests: unit(dev)
"

Fixes #4615.

* 'assign_proper_io_priority_to_streaming_view_updates' of https://github.com/psarna/scylla:
  db,view: wrap view update generation in stream scheduling group
  database: assign proper io priority for streaming view updates

(cherry picked from commit 2c7435418a)
2019-08-22 16:21:42 +03:00
Botond Dénes
450b9ac9bf multishard_combining_reader: shard reader: don't stop on non-full prefixes
This patch is a backport of the fix for #4733 (merged to master as
0cf4fab). As the shard reader code has been substantially refactored
post the 3.0 branch cut time, that fix cannot be backported at all,
instead this is a separate fix developed specially for 3.0.

To quickly reiterate, the problem at hand is that when recreating a
previously evicted shard reader of a multishard reader, the position of
the last fragment seen by that reader is used as the position after
which the read resumes. For this we just created a clustering range
starting from *after* the key (open bound). This works well in most
cases but when that last key is a non-full prefix this will also ignore
any still unread clustering rows that falls into that prefix.

This patch doesn't attempt to fix the problem in a systematic way like
the fix in master does, making sure reader recreation works properly
with prefixes as well, instead, for the sake of minimizing the impact,
we simply avoid ending the buffer on a prefix key. This fix is more
naive and can cause over-read when the stream contains lots of
successive range tombstones with prefix positions. On the other hand,
this leads to a *much* simpler fix, and anyway, as reader eviction is
much rarer in 3.0 this should have a lesser impact.

A unit test is also added to make sure the problem is fixed.

Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <20190819120748.28168-1-bdenes@scylladb.com>
2019-08-19 15:09:47 +03:00
Jenkins
b3bfd8c08d release: prepare for 3.0.10 by hagitsegev 2019-08-14 14:58:50 -04:00
Tomasz Grabiec
53c10b72dc Merge "Fix the system.size_estimates table" from Kamil
Fixes a segfault when querying for an empty keyspace.

Also, fixes an infinite loop on smp > 1. Queries to
system.size_estimates table which are not single-partition queries
caused Scylla to go into an infinite loop inside
multishard_combining_reader::fill_buffer. This happened because
multishard_combinind_reader assumes that shards return rows belonging
to separate partitions, which was not the case for
size_estimates_mutation_reader.

Fixes #4689
2019-08-14 15:31:54 +02:00
Kamil Braun
a690e20966 Fix infinite looping when performing a range query on system.size_estimates.
Queries to system.size_estimates table which are not single parition queries
caused Scylla to go into an infinite loop inside multishard_combining_reader::fill_buffer.
This happened because multishard_combinind_reader assumes that shards return rows belonging
to separate partitions, which was not the case for size_estimates_mutation_reader.
This commit fixes the issue and closes #4689.
2019-08-14 12:51:33 +02:00
Kamil Braun
7172009a0d Fix segmentation fault when querying system.size_estimates for an empty keyspace. 2019-08-14 12:51:33 +02:00
Kamil Braun
cb688ef62e Refactor size_estimates_virtual_reader
Move the implementation of size_estimates_mutation_reader
to a separate compilation unit to speed up compilation times
and increase readability.

Refactor tests to use seastar::thread.
2019-08-14 12:51:27 +02:00
Kamil Braun
ff8265dd66 Fix command line argument parsing in main.
Command line arguments are parsed twice in Scylla: once in main and once in Seastar's app_template::run.
The first parse is there to check if the "--version" flag is present --- in this case the version is printed
and the program exists.  The second parsing is correct; however, most of the arguments were improperly treated
as positional arguments during the first parsing (e.g., "--network host" would treat "host" as a positional argument).
This happened because the arguments weren't known to the command line parser.
This commit fixes the issue by moving the parsing code until after the arguments are registered.
Resolves #4141.

Signed-off-by: Kamil Braun <kbraun@scylladb.com>
(cherry picked from commit f155a2d334)
2019-08-13 20:13:24 +03:00
Avi Kivity
a198db31dc Merge "Fix disable_sstable_write synchronization with on_compaction_completion" from Benny
"
disable_sstable_write needs to acquire _sstable_deletion_sem to properly synchronize
with background deletions done by on_compaction_completion to ensure no sstables will
be created or deleted during reshuffle_sstables after
storage_service::load_new_sstables disables sstable writes.

Fixes #4622

Test: unit(dev), nodetool_additional_test.py migration_test.py
"

* 'fix-disable-sstable-write-for-3.0' of https://github.com/bhalevy/scylla:
  table: document _sstables_lock/_sstable_deletion_sem locking order
  table: disable_sstable_write: acquire _sstable_deletion_sem
  table: uninline enable_sstable_write
2019-08-12 16:53:47 +03:00
Avi Kivity
094a2a4263 Merge "Catch unclosed partition sstable write #4794" from Tomasz
"
Not emitting partition_end for a partition is incorrect. SStable
writer assumes that it is emitted. If it's not, the sstable will not
be written correctly. The partition index entry for the last partition
will be left partially written, which will result in errors during
reads. Also, statistics and sstable key ranges will not include the
last partition.

It's better to catch this problem at the time of writing, and not
generate bad sstables.

Another way of handling this would be to implicitly generate a
partition_end, but I don't think that we should do this. We cannot
trust the mutation stream when invariants are violated, we don't know
if this was really the last partition which was supposed to be
written. So it's safer to fail the write.

Enabled for both mc and la/ka.

Passing --abort-on-internal-error on the command line will switch to
aborting instead of throwing an exception.

The reason we don't abort by default is that it may bring the whole
cluster down and cause unavailability, while it may not be necessary
to do so. It's safer to fail just the affected operation,
e.g. repair. However, failing the operation with an exception leaves
little information for debugging the root cause. So the idea is that the
user would enable aborts on only one of the nodes in the cluster to
get a core dump and not bring the whole cluster down.
"

* 'catch-unclosed-partition-sstable-write' of https://github.com/tgrabiec/scylla:
  sstables: writer: Validate that partition is closed when the input mutation stream ends
  config, exceptions: Add helper for handling internal errors
  utils: config_file: Introduce named_value::observe()

(cherry picked from commit 95c0804731)
(cherry picked from commit cf4c238b28)
2019-08-08 16:47:26 +03:00
Asias He
cc0b4d249b streaming: Send error code from the sender to receiver
In case of error on the sender side, the sender does not propagate the
error to the receiver. The sender will close the stream. As a result,
the receiver will get nullopt from the source in
get_next_mutation_fragment and pass mutation_fragment_opt with no value
to the generating_reader. In turn, the generating_reader generates end
of stream. However, the last element that the generating_reader has
generated can be any type of mutation_fragment. This makes the sstable
that consumes the generating_reader violates the mutation_fragment
stream rule.

To fix, we need to propagate the error. However RPC streaming does not
support propagate the error in the framework. User has to send an error
code explicitly.

Fixes: #4789
(cherry picked from commit bac987e32a)

streaming: Move stream_mutation_fragments_cmd to a new file

Avoid including the stream_session.hh in messaging_service.hh.

More importantly, fix the build because currently messaging_service.cc
and messaging_service.hh does not include stream_mutation_fragments_cmd.
I am not sure why it builds on my machine. Spotted this when backporting
the change to 3.0 branch.

Refs: #4789
(cherry picked from commit 49a73aa2fc)

streaming: Do not call rpc stream flush in send_mutation_fragments

The stream close() guarantees the data sent will be flushed. No need to
call the stream flush() since the stream is not reused.

Follow up fix for commit bac987e32a (streaming: Send error code from
the sender to receiver).

Fixes: #4789
(cherry picked from commit 288371ce75)
Message-Id: <87058e290ae3f59f874b860121786b22f24957c7.1565189319.git.asias@scylladb.com>
2019-08-08 11:41:25 +02:00
Asias He
e10afc7f50 messaging_service: Check if messaging_service is stopped before get_rpc_client
get_rpc_client assumes the messaging_service is not stopped. We should check
is_stopping() before we call get_rpc_client.

We do such check in existing code, e.g., send_message and friends. Do
the same check in the newly introduced
make_sink_and_source_for_stream_mutation_fragments() and friends for row
level repair.

Fixes: #4767
(cherry picked from commit 5d3e4d7b73)

Note: only the change for make_sink_and_source_for_stream_mutation_fragments is backported.
Message-Id: <06079d4e48ea81ba567a2f45be2ab3a51f042e28.1565189319.git.asias@scylladb.com>
2019-08-08 11:40:49 +02:00
Tomasz Grabiec
407dfe0d68 lsa: Fix spurios abort with --enable-abort-on-lsa-bad-alloc
allocate_segment() can fail even though we're not out of memory, when
it's invoked inside an allocating section with the cache region
locked. That section may later succeed after retried after memory
reclamation.

We should ignore bad_alloc thrown inside allocating section body and
fail only when the whole section fails.

Fixes #2924

Message-Id: <1550597493-22500-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit dafe22dd83)
2019-08-08 11:39:39 +02:00
Raphael S. Carvalho
9370996a18 table: do not rely on undefined behavior in cleanup_sstables
It shouldn't rely on argument evaluation order, which is ub.

Fixes #4718.

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
(cherry-picked from commit 0e732ed1cf)
2019-08-07 21:53:12 +03:00
Rafael Ávila de Espíndola
ac105dd2a7 mc writer: Fix exception safety when closing _index_writer
This fixes a possible cause of #4614.

From the backtrace in that issue, it looks like a file is being closed
twice. The first point in the backtrace where that seems likely is in
the MC writer.

My first idea was to add a writer::close and make it the responsibility
of the code using the writer to call it. That way we would move work
out of the destructor.

That is a bit hard since the writer is destroyed from
flat_mutation_reader::impl::~consumer_adapter and that would need to
get a close function too.

This patch instead just fixes an exception safety issue. If
_index_writer->close() throws, _index_writer is still valid and
~writer will try to close it again.

If the exception was thrown after _completed.set_value(), that would
explain the assert about _completed.set_value() being called twice.

With this patch the path outside of the destructor now moves the
writer to a local variable before trying to close it.

Fixes #4614
Message-Id: <20190710171747.27337-1-espindola@scylladb.com>

(cherry picked from commit 281f3a69f8)
2019-08-07 21:43:44 +03:00
Benny Halevy
1e62fc8aac table: document _sstables_lock/_sstable_deletion_sem locking order
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
(cherry picked from commit 0e4567c881)
2019-08-07 17:09:47 +03:00
Benny Halevy
c724eee649 table: disable_sstable_write: acquire _sstable_deletion_sem
`disable_sstable_write` needs to acquire `_sstable_deletion_sem`
to properly synchronize with background deletions done by
`on_compaction_completion` to ensure no sstables will be created
or deleted during `reshuffle_sstables` after
`storage_service::load_new_sstables` disables sstable writes.

Fixes #4622

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
(cherry picked from commit 6dad9baa1c)
2019-08-07 17:06:38 +03:00
Benny Halevy
ebb14d93c9 table: uninline enable_sstable_write
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
(cherry picked from commit bbbd749f70)
2019-08-07 17:04:08 +03:00
Tomasz Grabiec
d77aaada86 sstables: ka/la: reader: Make sure push_ready_fragments() does not miss to emit partition_end
Currently, if there is a fragment in _ready and _out_of_range was set
after row end was consumer, push_ready_fragments() would return
without emitting partition_end.

This is problematic once we make consume_row_start() emit
partiton_start directly, because we will want to assume that all
fragments for the previous partition are emitted by then. If they're
not, then we'd emit partition_start before partition_end for the
previous partition. The fix is to make sure that
push_ready_fragments() emits everything.

Fixes #4786

(cherry picked from commit 9b8ac5ecbc)
Signed-off-by: Piotr Jastrzebski <piotr@scylladb.com>
2019-08-01 13:06:56 +03:00
Avi Kivity
acd05e089f Update seastar submodule
* seastar 16641efb15...445b5126c2 (1):
  > reactor: fix deadlock of stall detector vs dlopen

Fixes #4759.
2019-07-31 18:33:28 +03:00
Avi Kivity
f591c9c710 sstable: index_reader: close index_reader::reader more robustly
If we had an error while reading, then we would have failed to close
the reader, which in turn can cause memory corruption. Make the
closing more robust by using then_wrapped (that doesn't skip on
exception) and log the error for analysis.

Fixes #4761.

(cherry picked from commit b272db368f)
2019-07-27 18:20:17 +03:00
Jenkins
dea4489078 release: prepare for 3.0.9 by hagitsegev 2019-07-24 12:09:49 +03:00
Raphael S. Carvalho
3172cc6bac sstables/compaction: Fix segfault when replacing expired sstable in incremental compaction
Fully expired sstable is not added to compacting set, meaning it's not actually
compacted, but it's kept in a list of sstables which incremental compaction
uses to check if any sstable can be replaced.
Incremental compaction was unconditionally removing expired sstable from compacting
set, which led to segfault because end iterator was given.

The fix is about changing sstable_set::erase() behavior to follow standard one
for erase functions which will works if the target element is not present.

Fixes #4085.

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Message-Id: <20190130163100.5824-1-raphaelsc@scylladb.com>
(cherry picked from commit 930f8caff9)
2019-07-22 15:07:00 +03:00
Asias He
840d466c4d streaming: Do not open rpc stream connection if ranges are not relevant to a shard
Given a list of ranges to stream, stream_transfer_task will create an
reader with the ranges and create a rpc stream connection on all the shards.

When user provides ranges to repair with -st -et options, e.g.,
using scylla-manger, such ranges can belong to only one shard, repair
will pass such ranges to streaming.

As a result, only one shard will have data to send while the rpc stream
connections are created on all the shards, which can cause the kernel
run out of ports in some systems.

To mitigate the problem, do not open the connection if the ranges do not
belong to the shard at all.

Refs: #4708
(cherry picked from commit 64a4c0ede2)
2019-07-21 10:24:21 +03:00
Kamil Braun
e30c289835 Fix timestamp_type_impl::timestamp_from_string.
Now it accepts the 'z' or 'Z' timezone, denoting UTC+00:00.
Fixes #4641.

Signed-off-by: Kamil Braun <kbraun@scylladb.com>
(cherry picked from commit 4417e78125)
2019-07-17 21:56:03 +03:00
Eliran Sinvani
f769828a68 auth: Prevent race between role_manager and pasword_authenticator
When scylla is started for the first time with PasswordAuthenticator
enabled, it can be that a record of the default superuser
will be created in the table with the can_login and is_superuser
set to null. It happens because the module in charge of creating
the row is the role manger and the module in charge of setting the
default password salted hash value is the password authenticator.
Those two modules are started together, it the case when the
password authenticator finish the initialization first, in the
period until the role manager completes it initialization, the row
contains those null columns and any loging attempt in this period
will cause a memory access violation since those columns are not
expected to ever be null. This patch removes the race by starting
the password authenticator and autorizer only after the role manger
finished its initialization.

Tests:
  1. Unit tests (release)
  2. Auth and cqlsh auth related dtests.

Fixes #4226

Signed-off-by: Eliran Sinvani <eliransin@scylladb.com>
Message-Id: <20190714124839.8392-1-eliransin@scylladb.com>
(cherry picked from commit 997a146c7f)
2019-07-15 21:18:24 +03:00
kbr-
7d743563bf Implement tuple_type_impl::to_string_impl. (#4645)
Resolves #4633.

Signed-off-by: Kamil Braun <kbraun@scylladb.com>
(cherry picked from commit 8995945052)
2019-07-08 11:11:30 +03:00
Jenkins
23da53c4f3 release: prepare for 3.0.8 by hagitsegev 2019-06-27 11:12:21 +03:00
Piotr Sarna
d4df119735 main: stop view builder conditionally
The view builder is started only if it's enabled in config,
via the view_building=true variable. Unfortunately, stopping
the builder was unconditional, which may result in failed
assertions during shutdown. To remedy this, view building
is stopped only if it was previously started.

Fixes #4589

(cherry picked from commit efa7951ea5)
2019-06-26 11:05:13 +03:00
Avi Kivity
bdcbf4aa4e Merge "Backport fixing infinite paging for indexed queries" from Piotr
"
This series backports fixing infinite paging for indexed queries
to branch-3.0.

Tests: unit(dev)
"

Fixes #4569

* 'fix_infinite_paging_for_indexed_queries_for_3.0' of https://github.com/psarna/scylla:
  tests: add test case for finishing index paging
  cql3: fix infinite paging for indexed queries
2019-06-25 11:56:11 +03:00
Avi Kivity
e80cd9dfed Merge "Backport fixing ignoring ck restrictions in filtering" from Piotr
"
Tests: unit(dev)
Refs #4541
"

* 'fix_ignoring_ck_restrictions_in_filtering_for_3.0_2' of https://github.com/psarna/scylla:
  tests: add a test case for filtering clustering key
  cql3: fix qualifying clustering key restrictions for filtering
  cql3: fix fetching clustering key columns for filtering
2019-06-25 11:56:11 +03:00
Piotr Sarna
87fd298a6e tests: add a test case for filtering clustering key
The test cases makes sure that clustering key restriction
columns are fetched for filtering if they form a clustering key prefix,
but not a primary key prefix (partition key columns are missing).

Ref #4541
2019-06-25 10:05:34 +02:00
Piotr Sarna
7dce5484c2 cql3: fix qualifying clustering key restrictions for filtering
Clustering key restrictions can sometimes avoid filtering if they form
a prefix, but that can happen only if the whole partition key is
restricted as well.

Ref #4541
2019-06-25 10:05:34 +02:00
Piotr Sarna
23df964b96 cql3: fix fetching clustering key columns for filtering
When a column is not present in the select clause, but used for
filtering, it usually needs to be fetched from replicas.
Sometimes it can be avoided, e.g. if primary key columns form a valid
prefix - then, they will be optimized out before filtering itself.
However, clustering key prefix can only be qualified for this
optimization if the whole partition key is restricted - otherwise
the clustering columns still need to be present for filtering.

This commit also fixes tests in cql_query_test suite, because they now
expect more values - columns fetched for filtering will be present as
well (only internally, the clients receive only data they asked for).

Fixes #4541
2019-06-25 10:05:27 +02:00
Piotr Sarna
fcab0d1392 tests: add test case for finishing index paging
The test case makes sure that paging indexes does not result
in an infinite loop.

Refs #4569

(cherry picked from commit b8cadc928c)
2019-06-24 10:14:35 +02:00
Piotr Sarna
a0c4a8501e cql3: fix infinite paging for indexed queries
Indexed queries need to translate between view table paging state
and base table paging state, in order to be able to page the results
correctly. One of the stages of this translation is overwriting
the paging state obtained from the base query, in order to return
view paging state to the user, so it can be used for fetching next
pages. Unfortunately, in the original implementation the paging
state was overwritten only if more pages were available,
while if 'remaining' pages were equal to 0, nothing was done.
This is not enough, because the paging state of the base query
needs to be overwritten unconditionally - otherwise a guard paging state
value of 'remaining == 0' is returned back to the client along with
'has_more_pages = true', which will result in an infinite loop.
This patch correctly overwrites the base paging state unconditionally.

Fixes #4569

(cherry picked from commit 88f3ade16f)
2019-06-24 09:37:06 +02:00
Nadav Har'El
b6fa715f7b storage_proxy: fix race and crash in case of MV and other node shutdown
Recently, in merge commit 2718c90448,
we added the ability to cancel pending view-update requests when we detect
that the target node went down. This is important for view updates because
these have a very long timeout (5 minutes), and we wanted to make this
timeout even longer.

However, the implementation caused a race: Between *creating* the update's
request handler (create_write_response_handler()) and actually starting
the request with this handler (mutate_begin()), there is a preemption point
and we may end up deleting the request handler before starting the request.
So mutate_begin() must gracefully handle the case of a missing request
handler, and not crash with a segmentation fault as it did before this patch.

Eventually the lifetime management of request handlers could be refactored
to avoid this delicate fix (which requires more comments to explain than
code), or even better, it would be more correct to cancel individual writes
when a node goes down, not drop the entire handler (see issue #4523).
However, for now, let's not do such invasive changes and just fix bug that
we set out to fix.

Fixes #4386.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20190620123949.22123-1-nyh@scylladb.com>
(cherry picked from commit 6e87bca65d)
2019-06-23 21:13:10 +03:00
Avi Kivity
9b3ca26d7f Merge "Fix deciding whether a query uses indexing" from Piotr
"
This series backports fixing deciding whether a query uses indexing
for branch-3.0

Fixes #4539
Branches: 3.0
"

* 'fix_deciding_whether_a_query_uses_indexing_for_3.0' of https://github.com/psarna/scylla:
  tests: add case for partition key index and filtering
  cql3: fix deciding if a query uses indexing
2019-06-18 14:41:47 +03:00
Piotr Sarna
7b8e570e6c tests: add case for partition key index and filtering
The test ensures that partition key index does not influence
filtering decisions for regular columns.

Ref #4539
2019-06-18 13:35:32 +02:00
Piotr Sarna
a947f2cd84 cql3: fix deciding if a query uses indexing
The code that decides whether a query should used indexing
was buggy - a partition key index might have influenced the decision
even if the whole partition key was passed in the query (which
effectively means that indexing it is not necessary).

Fixes #4539
2019-06-18 13:19:31 +02:00
Avi Kivity
5ce5f61b08 Update seastar submodule
* seastar f541231...16641ef (1):
  > build: add libatomic to install-depenencies.sh

Fixes #4562.
2019-06-17 13:52:04 +03:00
Piotr Jastrzebski
7b65ec866b sstables: distinguish empty and missing cellpath
Before this patch mc sstables writer was ignoring
empty cellpaths. This is a wrong behaviour because
it is possible to have empty key in a map. In such case,
our writer creats a wrong sstable that we can't read back.
This is becaus a complex cell expects cellpath for each
simple cell it has. When writer ignores empty cellpath
it writes nothing and instead it should write a length
of zero to the file so that we know there's an empty cellpath.

Fixes #4533

Tests: unit(release)

Signed-off-by: Piotr Jastrzebski <piotr@scylladb.com>
Message-Id: <46242906c691a56a915ca5994b36baf87ee633b7.1560532790.git.piotr@scylladb.com>
(cherry picked from commit a41c9763a9)
2019-06-16 09:06:37 +03:00
Jenkins
4c16c1fe1b release: prepare for 3.0.7 by hagitsegev 2019-05-26 22:30:19 +03:00
Paweł Dziepak
f2d2a9f5b8 Merge "Fix empty counters handling in MC" from Piotr
"
Before this patchset empty counters were incorrectly persisted for
MC format. No value was written to disk for them. The correct way
is to still write a header that informs the counter is empty.

We also need to make sure that reading wrongly persisted empty
counters works because customers may have sstables with wrongly
persisted empty counters.

Fixes #4363
"

* 'haaawk/4363/v3' of github.com:scylladb/seastar-dev:
  sstables: add test for empty counters
  docs: add CorrectEmptyCounters to sstable-scylla-format
  sstables: Add a feature for empty counters in Scylla.db.
  sstables: Write header for empty counters
  sstables: Remove unused variables in make_counter_cell
  sstables: Handle empty counter value in read path

(cherry picked from commit 899ebe483a)
2019-05-24 06:23:38 +03:00
Gleb Natapov
cb3b687492 cache_hitrate_calculator: make cache hitrate calculation preemptable
The calculation is done in a non preemptable loop over all tables, so if
numbers of tables is very large it may take a while since we also build
a string for gossiper state. Make the loop preemtable and also make
the string calculation more efficient by preallocating memory for it.
Message-Id: <20190516132748.6469-3-gleb@scylladb.com>

(cherry picked from commit 31bf4cfb5e)
2019-05-17 14:38:38 +02:00
Gleb Natapov
1bb84cdbcf cache_hitrate_calculator: do not copy stats map for each cpu
invoke_on_all() copies provided function for each shard it is executed
on, so by moving stats map into the capture we copy it for each shard
too. Avoid it by putting it into the top level object which is already
captured by reference.
Message-Id: <20190516132748.6469-2-gleb@scylladb.com>

(cherry picked from commit 4517c56a57)
2019-05-17 12:40:45 +02:00
Gleb Natapov
b6307d54be cache_hitrate_calculator: wait for ongoing calculation to complete during stop
Currently stop returns ready future immediately. This is not a problem
since calculation loop holds a shared pointer to the local service, so
it will not be destroyed until calculation completes and global database
object db, that also used by the calculation, is never destroyed. But the
later is just a workaround for a shutdown sequence that cannot handle
it and will be changed one day. Make cache hitrate calculation service
ready for it.

Message-Id: <20190422113538.GR21208@scylladb.com>
(cherry picked from commit c6b3b9ff13)
2019-05-17 12:40:41 +02:00
Avi Kivity
a20000c1a2 Merge "multishard reader: fix handling of non strictly monotonous positions" from Botond
"
The shard readers of the multishard reader assumed that the positions in
the data stream are strictly monotonous. This assumption is invalid.
Range tombstones can have positions that they can share with other range
tombstones and/or a clustering row. The effect of this false assumption
was that when the shard reader was evicted such that the last seen
fragment was a range tombstone, when recreated it would skip any unseen
fragments that have the same position as that of the last seen range
tombstone.

This series contains some additional fixes for the
`flat_mutation_reader_from_mutations()` reader, to make the backported unit
tests pass.

Fixes: #4418

Tests: unit(release: network_topology_strategy_test times out - don't
think it is related to these changes)
"

* 'multishard_reader_handle_non_strictly_monotonous_positions-branch-3.0/v1' of https://github.com/denesb/scylla:
  tests: add unit test for multishard reader correctly handling non-strictly monotonous positions
  flat_mutation_reader: add make_flat_mutation_reader_from_fragments() overload with range and slice
  flat_mutation_reader: add flat_mutation_reader_from_mutations() overload with range and slice
  flat_mutation_reader_from_mutations: destroy all remaining mutations
  flat_mutation_reader_from_mutations: fix empty range case
  multishard_combining_reader: fix handling of non-strictly monotonous positions
  position_in_partition_view: add region() accessor
2019-05-06 11:42:29 +03:00
Botond Dénes
b3cbc2e58a tests: add unit test for multishard reader correctly handling non-strictly monotonous positions
(cherry picked from commit aa18bb33b9)
2019-05-06 11:19:04 +03:00
Botond Dénes
e4c1c4f052 flat_mutation_reader: add make_flat_mutation_reader_from_fragments() overload with range and slice
To be able to support this new overload, the reader is made
partition-range aware. It will now correctly only return fragments that
fall into the partition-range it was created with. For completeness'
sake and to be able to test it, also implement
`fast_forward_to(const dht::partition_range)`. Slicing is done by
filtering out non-overlapping fragments from the initial list of
fragments. Also add a unit test that runs it through the mutation_source
test suite.

(cherry picked from commit 51e81cf027)
2019-05-06 11:19:04 +03:00
Benny Halevy
bfe3b4cc59 time_window_backlog_tracker: fix use after free
Fixes #4465

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20190430094209.13958-1-bhalevy@scylladb.com>
(cherry picked from commit 3a2fa82d6e)
2019-05-06 09:38:31 +03:00
Botond Dénes
6a4bc5bd71 flat_mutation_reader: add flat_mutation_reader_from_mutations() overload with range and slice
To be able to run the mutation-source test suite with this reader. In
the next patch, this reader will be used in testing another reader, so
it is important to make sure it works correctly first.

(cherry picked from commit bc08f8fd07)
2019-05-06 09:17:48 +03:00
Paweł Dziepak
6c818bcec0 flat_mutation_reader_from_mutations: destroy all remaining mutations
If the reader is fast-forwarded to another partition range mutation_ may
be left with some partial mutations. Make sure that those are properly
destroyed.

(cherry picked from commit 048ed2e3d3)
2019-05-06 09:17:17 +03:00
Paweł Dziepak
1598d358f0 flat_mutation_reader_from_mutations: fix empty range case
An iterator shall not be dereferenced until it is verified that it is
dereferencable.

(cherry picked from commit d50cd31eee)
2019-05-06 09:17:17 +03:00
Botond Dénes
7252715c69 multishard_combining_reader: fix handling of non-strictly monotonous positions
The shard readers under a multishard reader are paused every time the
read moves to another ahrd. When paused they can be evicted at any time.
When this happens, they will be re-created lazily on the next
operation, with a start position such that they continue reading from
where the evicted reader left off. This start position is determined
from the last fragment seen by the previous reader. When this position
is clustering position, the reader will be recreated such that it reads
the clustering range (from the half-read partition): (last-ckey, +inf).
This can cause problems if the last fragment seen by the evicted reader
was a range-tombstone. Range tombstones can share the same clustering
position with other range tombstones and potentially one clustering row.
This means that when the reader is recreated, it will start from the
next clustering position, ignoring any unread fragments that share the
same position as the last seen range tombstone.
To fix, ensure that when pausing the reader, we extract all fragments
for the last position. To this end, when the last extracted fragment
is a range tombstone (with pos x), we continue reading until we see a
fragment with a position y that is greater. This way it is ensured that
we have seen all fragments for pos x and it is safe to resume the read,
starting from after position x.

(cherry picked from commit eba310163d)
2019-05-06 09:17:17 +03:00
Botond Dénes
37e143cba5 position_in_partition_view: add region() accessor
(cherry picked from commit b30af48c83)
2019-05-06 09:17:17 +03:00
Jenkins
bf68fae01b release: prepare for 3.0.6 by penberg 2019-05-03 14:14:31 +03:00
Gleb Natapov
d566466fca batchlog_manager: fix array out of bound access
endpoint_filter() function assumes that each bucket of
std::unordered_multimap contains elements with the same key only, so
its size can be used to know how many elements with a particular key
are there.  But this is not the case, elements with multiple keys may
share a bucket. Fix it by counting keys in other way.

Fixes #3229

Message-Id: <20190501133127.GE21208@scylladb.com>
(cherry picked from commit 95c6d19f6c)
2019-05-03 11:59:29 +03:00
Avi Kivity
e32e682911 Merge "SI: Add virtual columns to underlying MV" from Duarte
"
Virtual columns are MV-specific columns that contribute to the
liveness of view rows. However, we were not adding those columns when
creating an index's underlying MV, causing indexes to miss base rows.

Fixes #4144
Branches: master, branch-3.0
"

Reviewed-by: Nadav Har'El <nyh@scylladb.com>

* 'sec-index/virtual-columns/v1' of https://github.com/duarten/scylla:
  tests/secondary_index_test: Add reproducer for #4144
  index/secondary_index_manager: Add virtual columns to MV

(cherry picked from commit ebf179318c)
2019-05-01 12:58:35 +01:00
Tomasz Grabiec
3c46bbf244 lsa: Fix compact_and_evict() being called with a too low step
compact_and_evict gets memory_to_release in bytes while
reclamation step is in segments.

Broken in f092decd90.

It doesn't make much difference with the current default step of 1
segment since we cannot reclaim less than that, so shouldn't cause
problems in practice.

Ref #4445

Message-Id: <1556013920-29676-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit 21fbf59fa8)
2019-04-23 23:10:38 +03:00
Gleb Natapov
5567cf4b1b cache_hitrate_calculator: fix use after free in non_system_filter lambda
non_system_filter lambda is defined static which means it is initialized
only once, so the 'this' that is will capture will belong to a shard
where the function runs first. During service destruction the function
may run on different shard and access already other's shard service that
may be already freed.

Fixed #4425

Message-Id: <20190421152139.GN21208@scylladb.com>
(cherry picked from commit 306f5b99b5)
2019-04-22 09:52:42 +03:00
Tomasz Grabiec
733c04ad50 lsa: Fix potential bad_alloc even though evictable memory exists
When we start the LSA reclamation it can be that
segment_pool::_free_segments is 0 under some conditions and
segment_pool::_current_emergency_reserve_goal is set to 1. The
reclamation step is 1 segment, and compact_and_evict_locked() frees 1
segment back into the segment_pool. However,
segment_pool::reclaim_segments() doesn't free anything to the standard
allocator because the condition _free_segments >
_current_emergency_reserve_goal is false. As a result,
tracker::impl::reclaim() returns 0 as the amount of released memory,
tracker::reclaim() returns
memory::reclaiming_result::reclaimed_nothing and the seastar allocator
thinks it's a real OOM and throws std::bad_alloc.

The fix is to change compact_and_evict() to make sure that reserves
are met, by releasing more if they're not met at entry.

This change also allows us to drop the variant of allocate_segment()
which accepts the reclamation step as a means to refill reserves
faster. This is now not needed, because compact_and_evict() will look
at the reserve deficit to increase the amount of memory to reclaim.

Fixes #4445

Message-Id: <1555671713-16530-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit f092decd90)
2019-04-20 16:44:49 +03:00
Raphael S. Carvalho
05913b6f58 database: fix 2x increase in disk usage during cleanup compaction
Don't hold reference to sstables cleaned up, so that file descriptors
for their index and data files will be closed and consequently disk
space released.

Fixes #3735.

Backport note:
To reduce risk considerably, we'll not backport a mechanism to release
sstable introduced in incremental compaction work.
Instead, only one sstable is passed to table::cleanup_sstables() at a
time (it won't affect performance because the operation is serialized
anyway), to make it easy to release reference to cleaned sstable held
by compaction manager.

tests: release mode; manually checked cleanup's disk space issue is gone.

(cherry picked from commit 5bc028f78b)

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Message-Id: <20190417155024.27359-1-raphaelsc@scylladb.com>
2019-04-17 18:01:48 +01:00
Duarte Nunes
79cf277ea2 db/schema_tables: Diff tables using ID instead of name
Currently we diff schemas based on table/view name, and if the names
match, then we detect altered schemas by comparing the schema
mutations. This fails to detect transitions which involve dropping and
recreating a schema with the same name, if a node receives these
notifications simultaneously (for example, if the node was temporarily
down or partitioned).

Note that because the ID is persisted and created when executing a
create_table_statement, then even if a schema is re-created with the
exact same structure as before, we will still considered it altered
because the mutations will differ.

This also stops schema pulling from working, since it relies on schema
merging.

The solution is to diff schemas using their ID, and not their name.

Keyspaces and user types are also susceptible to this, but in their
case it's fine: these are values with no identity, and are just
metadata. Dropping and recreating a keyspace can be views as dropping
all tables from the keyspace, altering it, and eventually adding new
tables to the keyspace.

Note that this solution doesn't apply to tables dropped and created
with the same ID (using the `WITH ID = {}` syntax). For that, we would
need to detect deltas instead of applying changes and then reading the
new state to find differences. However, this solution is enough,
because tables are usually created with ID = {} for very specific,
peculiar reasons. The original motivation meant for the new table to
be treated exactly as the old, so the current behavior is in fact the
desired one.

Tests: unit(release), dtests(schema_test, schema_management_test)

Fixes #3797

Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20181001230932.47153-2-duarte@scylladb.com>
(cherry picked from commit 40a30d4129)
2019-04-17 18:01:48 +01:00
Duarte Nunes
03ada48b40 db/schema_tables: Drop tables before creating new ones
Doing it by the inverse order doesn't support dropping and creating a
schema with the same name.

Refs #3797

Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20181001230932.47153-1-duarte@scylladb.com>
(cherry picked from commit e404f09a23)
2019-04-17 18:01:48 +01:00
Duarte Nunes
394afae3a8 service/migration_manager: Validate duplicate ID in time
We allow tables to be created with the ID property, mostly for
advanced recovery cases. However, we need to validate that the ID
doesn't match an existing one. We currently do this in
database::add_column_family(), but this is already too late in the
normal workflow: if we allow the schema change to go through, then
it is applied to the system tables and loaded the next time the node
boots, regardless of us throwing from database::add_column_family().

To fix this, we perform this validation when announcing a new table.

Note that the check wasn't removed from database::add_column_family();
it's there since 2015 and there might have been other reasons to add
it that are not related to the ID property.

Refs #2059

Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20181001230142.46743-1-duarte@scylladb.com>
(cherry picked from commit 7ba944a243)
2019-04-17 18:01:48 +01:00
Tomasz Grabiec
69d0b1e15c schema_tables: Serialize schema merges fairly
All schema changes made to the node locally are serialized on a
semaphore which lives on shard 0. For historical reasons, they don't
queue but rather try to take the lock without blocking and retry on
failure with a random delay from the range [0, 100 us]. Contenders
which do not originate on shard 0 will have an extra disadvantage as
each lock attempt will be longer by the across-shard round trip
latency. If there is constant contention on shard 0, contenders
originating from other shards may keep loosing to take the lock.

Schema merge executed on behalf of a DDL statement may originate on
any shard. Same for the schema merge which is coming from a push
notification. Schema merge executed as part of the background schema
pull will originate on shard 0 only, where the application state
change listeners run. So if there are constant schema pulls, DDL
statements may take a long time to get through.

The fix is to serialize merge requests fairly, by using the blocking
semaphore::wait(), which is fair.

We don't have to back-off any more, since submit_to() no longer has a
global concurrency limit.

Fixes #4436.

Message-Id: <1555349915-27703-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit 3fd82021b1)
2019-04-16 10:19:45 +03:00
Avi Kivity
403f66ecad Revert "Revert "compaction: fix use-after-free when calculating backlog after schema change""
This reverts commit 841ceac4f9. It was reverted because the test
failed; this turned out due to a miscompile of the test.

With this additional fix, the test compiles correctly:

--- a/tests/sstable_datafile_test.cc
+++ b/tests/sstable_datafile_test.cc
@@ -4785,11 +4785,11 @@ SEASTAR_TEST_CASE(backlog_tracker_correctness_after_stop_tracking_compaction) {

             auto fut = sstables::compact_sstables(sstables::compaction_descriptor(ssts), *cf, sst_gen);

             bool stopped_tracking = false;
             for (auto& info : cf._data->cm.get_compactions()) {
-                if (info->cf == &*cf) {
+                if (info->cf == cf->schema()->cf_name()) {
                     info->stop_tracking();
                     stopped_tracking = true;
                 }
             }
             BOOST_REQUIRE(stopped_tracking);

info->cf is an sstring, and &*cf is a table*. It's not clear how the compiler
was able to compare an sstring and a pointer.
2019-04-12 21:45:59 +03:00
Shlomi Livne
841ceac4f9 Revert "compaction: fix use-after-free when calculating backlog after schema change"
This reverts commit 2b326fc7fa.
2019-04-12 09:55:06 +03:00
Avi Kivity
0fce4b228e Update seastar submodule
* seastar e9bb565...f541231 (1):
  > Merge "perftune.py: log NVMe IRQ SMP" from Vlad

Fixes #4057.
2019-04-12 00:19:58 +03:00
Botond Dénes
2336c092a0 types: fix date_type_impl::less() (timestamp cql type)
date_type_impl::less() invokes `compare_unsigned()` to compare the
underlying raw byte values. `compared_unsigned()` is a tri comparator,
however `date_type_impl::less()` implicitely converted the returned
value to bool. In effect, `date_type_impl::less()` would *always* return
`true` when the two compared values were not equal.

Found while working on a unit test which empoly a randomly generated
schema to test a component.


Fixes #4419.

Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <8a17c81bad586b3772bf3d1d1dae0e3dc3524e2d.1554907100.git.bdenes@scylladb.com>
(cherry picked from commit f201f8abab)
2019-04-12 00:19:58 +03:00
Raphael S. Carvalho
2b326fc7fa compaction: fix use-after-free when calculating backlog after schema change
The problem happens after a schema change because we fail to properly
remove ongoing compaction, which stopped being tracked, from list that
is used to calculate backlog, so it may happen that a compaction read
monitor (ceases to exist after compaction ends) is used after freed.

Fixes #4410.

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Message-Id: <20190409024936.23775-1-raphaelsc@scylladb.com>
(cherry picked from commit 8a117c338a)
2019-04-12 00:19:58 +03:00
Asias He
a62edaf7a9 streaming: Reject stream if the _sys_dist_ks or _view_update_generator are not ready
They are of type db::system_distributed_keyspace and
db::view::view_update_generator.

n1 is in normal status
n2 boots up and _sys_dist_ks or _view_update_generator are not
initialized
n1 runs stream, n2 is the follower.
n2 uses the _sys_dist_ks or _view_update_generator
"Assertion `local_is_initialized()' failed" is observed

Fixes #4360

Message-Id: <4ae13e1640ac8707a9ba0503a2744f6faf89ecf4.1554330030.git.asias@scylladb.com>
(cherry picked from commit f212dfb887)
2019-04-11 07:38:53 +03:00
Takuya ASADA
d527ef19f7 dist/docker/redhat: switch to python36
Since EPEL switched python3 default version to 3.6, we need to follow
the change.

Signed-off-by: Takuya ASADA <syuu@scylladb.com>
Message-Id: <20190409094139.4797-2-syuu@scylladb.com>
2019-04-10 16:10:30 +03:00
Takuya ASADA
8568dc94f4 dist/ami: switch to python36
Since EPEL switched python3 default version to 3.6, we need to follow
the change.

Signed-off-by: Takuya ASADA <syuu@scylladb.com>
Message-Id: <20190409094139.4797-1-syuu@scylladb.com>
2019-04-10 16:10:29 +03:00
Takuya ASADA
6e51a95668 dist/redhat: switch to python36
Since EPEL switched python3 default version to 3.6, we need to follow
the change.

Signed-off-by: Takuya ASADA <syuu@scylladb.com>
Message-Id: <20190408160934.32701-1-syuu@scylladb.com>
2019-04-08 19:12:59 +03:00
Avi Kivity
071191b967 Update seastar submodule
> fstream: remove default extent allocation hint
  > deleter: prevent early memory free caused by deleter append.
  > iotune: throw exception if no data was collected
  > perftune.py: fix irqbalance tuning on Ubuntu 18
  > fix use after free in rpc server handler

Fixes #4336
Fixes #4400
Fixes #4401
Fixes #4402
Fixes #4404
Ref #3867
2019-04-03 16:32:45 +03:00
Shlomi Livne
c6c841c34f Revert "Merge 'Add canceling long-standing view update requests' from Piotr"
This series introduces regression in dtests
materialized_views_test/TestMaterializedViews/interrupt_build_process_with_resharding_*.

This reverts commit b2227c7a5e.

Ref #3826
Ref #3966
Ref #4028

Signed-off-by: Shlomi Livne <shlomi@scylladb.com>
Message-Id: <a3aea137bfde956241acc6d57e1c387a8202486c.1554116404.git.shlomi@scylladb.com>
2019-04-01 14:07:47 +03:00
Nadav Har'El
83a8f779bb view_complex_test: fix another ttl
In a previous patch I fixed most TTLs in the view_complex_test.cc tests
from low numbers to 100 seconds. I missed one. This one never caused
problems in practice, but for good form, let's fix it too.

Ref #3918.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20181115160234.26478-1-nyh@scylladb.com>
(cherry picked from commit 45f05b06d2)
2019-03-31 12:28:37 +03:00
Nadav Har'El
6066968e33 view_complex_test: increase low ttl which may fail test on busy machine
Several of the tests in tests/view_complex_test.cc set a cell with a
TTL, and then skip time ahead artificially with forward_jump_clocks(),
to go past the TTL time and check the cell disappeared as expected.

The TTLs chosen for these tests were arbitrary numbers - some had 3 seconds,
some 5 seconds, and some 60 seconds. The actual number doesn't matter - it
is completely artificial (we move the clock with forward_jump_clocks() and
never really wait for that amount of time) and could very well be a million
seconds. But *low* numbers, like the 3 seconds, present a problem on extremely
overcomitted test machines. Our eventually() function already allows for
the possibility that things can hang for up to 8 seconds, but with a 3 second
TTL, we can find ourselves with data being expired and the test failing just
after 3 seconds of wall time have passed - while the test intended that the
dataq will expire only when we explicitly call forward_jump_clocks().

So this patch changes all the TTLs in this test to be the same high number -
100 seconds. This hopefully fixes #3918.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20181115125607.22647-1-nyh@scylladb.com>
(cherry picked from commit 4108458b8e)
2019-03-31 12:27:48 +03:00
Tomasz Grabiec
d3d877b9db Merge "db/view: Apply tracked tombstones for new updates" from Duarte
When generating view updates for base mutations when no pre-existing
data exists, we were forgetting to apply the tracked tombstones.

Fixes #4321
Tests: unit(dev)

* https://github.com/duarten/scylla materialized-views/4321/v1.1:
  db/view: Apply tracked tombstones for new updates
  tests/view_schema_test: Add reproducer for #4321

(cherry picked from commit 2b8bf0dbf8)
2019-03-27 21:56:21 +00:00
Piotr Sarna
5ec646cb4e types: fix varint and decimal serialization
Varint and decimal types serialization did not update the output
iterator after generating a value, which may lead to corrupted
sstables - variable-length integers were properly serialized,
but if anything followed them directly in the buffer (e.g. in a tuple),
their value will be overwritten.

Fixes #4348

Tests: unit (dev)
dtest: json_test.FromJsonUpdateTests.complex_data_types_test
       json_test.FromJsonInsertTests.complex_data_types_test
       json_test.ToJsonSelectTests.complex_data_types_test

Note that dtests still do not succeed 100% due to formatting differences
in compared results (e.g. 1.0e+07 vs 1.0E7, but it's no longer a query
correctness issue.

(cherry picked from commit 287a02dc05)
2019-03-26 16:38:37 +02:00
Jenkins
68b54b2e52 release: prepare for 3.0.5 by hagitsegev 2019-03-26 10:52:59 +02:00
Duarte Nunes
66a48746b8 service/storage_proxy: Don't consider view hints for MV backpressure
When a view replica becomes unavailable, updates to it are stored as
hints at the paired based replica. This on-disk queue of pending view
updates grows as long as there are view updated and the view replica
remains unavailable. Currently, we take that relative queue size into
account when calculating the delay for new base writes, in the context
of the backpressure algorithm for materialized views.

However, the way we're calculating that on-disk backlog is wrong,
since we calculate it per-device and then feed it to all the hints
managers for that device. This means that normal hints will show up as
backlog for the view hints manager, which in turn introduces delays.
This can make the view backpressure mechanism kick-in even if the
cluster uses no materialized views.

There's yet another way in which considering the view hints backlog is
wrong: a view replica that is unavailable for some period of time can
cause the backlog to grow to a point where all base writes are applied
the maximum delay of 1 second. This turns a single-node failure into
cluster unavailability.

The fix to both issues is to simply not take this on-disk backlog into
account for the backpressure algorithm.

Fixes #4351
Fixes #4352

Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Reviewed-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20190321170418.25953-1-duarte@scylladb.com>
(cherry picked from commit 93a1c27b31)
2019-03-25 15:02:06 +02:00
Duarte Nunes
b2227c7a5e Merge 'Add canceling long-standing view update requests' from Piotr
"
This series allows canceling view update requests when a node is
discovered DOWN. View updates are sent in the background with long
timeout (5 minutes), and in case we discover that the node is
unavailable, there's no point in waiting that long for the request
to finish. What's more, waiting for these requests occurs on shutdown,
which may result in waiting 5 minutes until Scylla properly shuts down,
which is bad for both users and dtests.

This series implements storage_proxy as a lifecycle subscriber,
so it can react to membership changes. It also keeps track of all
"interruptible" writes per endpoint, so once a node is detected as DOWN,
an artificial timeout can be triggered for all aforementioned write
requests.

Fixes #3826
Fixes #3966
Fixes #4028
"

* 'write_hints_for_view_updates_on_shutdown_4' of https://github.com/psarna/scylla:
  service: remove unused stop_hints_manager
  storage_proxy: add drain_on_shutdown implementation
  main: register storage proxy as lifecycle subscriber
  storage_proxy: add endpoint_lifecycle_subscriber interface
  storage_proxy: register view update handlers for view write type
  storage_proxy: add intrusive list of view write handlers
  storage_proxy: add view_update_write_response_handler

(cherry picked from commit 2718c90448)
2019-03-24 17:27:18 +02:00
Avi Kivity
97357a7321 Merge "sstables: mc: Write and read static compact tables the same way as Cassandra" from Tomasz
"
Static compact tables are tables with compact storage and no
clustering columns.

Before this patch, Scylla was writing rows of static compact tables as
clustered rows instead of as static rows. That's because in our in-memory
model such tables have regular rows and no static row. In Cassandra's
schema (since 3.x), those tables have columns which are marked as
static and there are no regular columns.

This worked fine as long as Scylla was writing and reading those
sstables. But when importing sstables from Cassandra, our reader was
skipping the static row, since it's not present in our schema, and
returning no rows as a result. Also, Cassandra, and Scylla tools,
would have problems reading those sstables.

Fix this by writing rows for such tables the same way as Cassandra
does. In order to support rolling downgrade, we do that only when all
nodes are upgraded.

Fixes #4139.

Tests:

  - unit (dev)
"

* tag 'static-compact-mc-fix-v3.1' of github.com:tgrabiec/scylla:
  tests: sstables: Test reading of static compact sstable generated by Cassandra
  tests: sstables: Add test for writing and reading of static compact tables
  sstables: mc: Write static compact tables the same way as Cassandra
  sstable: mc: writer: Set _static_row_written inside write_static_row()
  sstables: Add sstable::features()
  sstables: mc: writer: Prepare write_static_row() for working with any column_kind
  storage_service: Introduce the CORRECT_STATIC_COMPACT feature flag
  sstables: mc: writer: Build indexed_columns together with serialization_header
  sstables: mc: writer: De-optimize make_serialization_header()
  sstable: mc: writer: Move attaching of mc-specific components out of generic code

(cherry picked from commit eddb98e8c6)
2019-03-24 16:34:42 +02:00
Tomasz Grabiec
089e41999a tests: sstables: Extract make_sstable_mutation_source()
Message-Id: <1540459849-27612-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit 46d0c157ae)
2019-03-24 13:48:41 +02:00
Asias He
c537b3dd8e storage_service: Wait for gossip to settle only if do_bind is set
In commit 71bf757b2c, we call
wait_for_gossip_to_settle() which takes some time to complete in
storage_service::prepare_to_join().

In tests/cql_query_test calls init_server with do_bind == false which in
turn calls storage_service::prepare_to_join(). Since in the test, there
is only one node, there is no point to wait for gossip to settle.

To make the cql_query_test fast again, do not call
wait_for_gossip_to_settle if do_bind is false.

Before this patch, cql_query_test takes forever to complete.
After it takes 10s.

Ref #4289.

Tests: tests/cql_query_test
Message-Id: <3ae509e0a011ae30eef3f383c6a107e194e0e243.1553147332.git.asias@scylladb.com>
(cherry picked from commit c0f744b407)
2019-03-23 15:13:40 +02:00
Gleb Natapov
75a737c958 messaging_service: keep shared pointer to an rpc connection while opening mutation fragment stream
Current code captures a reference to rpc::client in a continuation, but
there is no guaranty that the reference will be valid when continuation runs.
Capture shared pointer to rpc::client instead.

Fixes #4350.

Message-Id: <20190314135538.GC21521@scylladb.com>
(cherry picked from commit bb93d990ad)
2019-03-23 15:11:28 +02:00
Tomasz Grabiec
ea0f1c039d row_cache: Fix abort in cache populating read concurrent with memtable flush
When we're populating a partition range and the population range ends
with a partition key (not a token) which is present in sstables and
there was a concurrent memtable flush, we would abort on the following
assert in cache::autoupdating_underlying_reader:

     utils::phased_barrier::phase_type creation_phase() const {
         assert(_reader);
         return _reader_creation_phase;
     }

That's because autoupdating_underlying_reader::move_to_next_partition()
clears the _reader field when it tries to recreate a reader but it finds
the new range to be empty:

         if (!_reader || _reader_creation_phase != phase) {
            if (_last_key) {
                auto cmp = dht::ring_position_comparator(*_cache._schema);
                auto&& new_range = _range.split_after(*_last_key, cmp);
                if (!new_range) {
                    _reader = {};
                    return make_ready_future<mutation_fragment_opt>();
                }

Fix by not asserting on _reader. creation_phase() will now be
meaningful even after we clear the _reader. The meaning of
creation_phase() is now "the phase in which the reader was last
created or 0", which makes it valid in more cases than before.

If the reader was never created we will return 0, which is smaller
than any phase returned by cache::phase_of(), since cache starts from
phase 1. This shouldn't affect current behavior, since we'd abort() if
called for this case, it just makes the value more appropriate for the
new semantics.

Tests:

  - unit.row_cache_test (debug)

Fixes #4236
Message-Id: <1553107389-16214-1-git-send-email-tgrabiec@scylladb.com>

(cherry picked from commit 69775c5721)
2019-03-22 09:31:59 -03:00
Asias He
27cf758f12 streaming: Get rid of the keep alive timer in streaming
There is no guarantee that rpc streaming makes progress in some time
period. Remove the keep alive timer in streaming to avoid killing the
session when the rpc streaming is just slow.

The keep alive timer is used to close the session in the following case:

n2 (the rpc streaming sender) streams to n1 (the rpc streaming receiver)
kill -9 n2

We need this because we do not kill the session when gossip think a node
is down, because we think the node down might only be temporary
and it is a waste to drop the previous work that has done especially
when the stream session takes long time.

Since in range_streamer, we do not stream all data in a single stream
session, we stream 10% of the data per time, and we have retry logic.
I think it is fine to kill a stream session when gossip thinks a node is
down. This patch changes to close all stream session with the node that
gossip think it is down.
Message-Id: <bdbb9486a533eee25fcaf4a23a946629ba946537.1551773823.git.asias@scylladb.com>

(cherry picked from commit b8158dd65d)
2019-03-20 19:47:11 +01:00
Avi Kivity
76fd69244a Merge "Fix empty remote common_features in check_knows_remote_features" from Asias
Three nodes in the cluster node1, node2, node3

Shutdown the whole cluster

Start node1

Start node2, node2 sees empty remote common_features.

   gossip - Feature check passed.  Local node 127.0.0.2 features =
   {CORRECT_COUNTER_ORDER, CORRECT_NON_COMPOUND_RANGE_TOMBSTONES, COUNTERS,
   DIGEST_MULTIPARTITION_READ, INDEXES, LARGE_PARTITIONS,
   LA_SSTABLE_FORMAT, MATERIALIZED_VIEWS, MC_SSTABLE_FORMAT,
   RANGE_TOMBSTONES, ROLES, ROW_LEVEL_REPAIR, SCHEMA_TABLES_V3,
   STREAM_WITH_RPC_STREAM, WRITE_FAILURE_REPLY, XXHASH},
   Remote common_features = {}

The problem is node3 hasn't started yet, node1 sees node3 has empty
features. In get_supported_features(), an empty common features will be
returned if an empty features of a node is seen. To fix, we should
fallback to use the features saved in system table.

Start node3, node3 sees empty remote common_features.

   gossip - Feature check passed. Local node 127.0.0.3 features =
   {CORRECT_COUNTER_ORDER, CORRECT_NON_COMPOUND_RANGE_TOMBSTONES, COUNTERS,
   DIGEST_MULTIPARTITION_READ, INDEXES, LARGE_PARTITIONS,
   LA_SSTABLE_FORMAT, MATERIALIZED_VIEWS, MC_SSTABLE_FORMAT,
   RANGE_TOMBSTONES, ROLES, ROW_LEVEL_REPAIR, SCHEMA_TABLES_V3,
   STREAM_WITH_RPC_STREAM, WRITE_FAILURE_REPLY, XXHASH},
   Remote common_features = {}

The problem is node3 hasn't inserted its own features into gossip
endpoint_state_map. get_supported_features() returns the common features
of all nodes in endpoint_state_map. To fix, we should fallback to use
the features stored in the system table for such node in this case.

Fixes #4225
Fixes #4341

* tag 'backport.fix.empty.remote.common.features.for.3.0.v1' of github.com:scylladb/seastar-dev:
  gossiper: Enable features only after gossip is settled
  Merge "Fix empty remote common_features in check_knows_remote_features" from Asias
  Merge "Fix window during init where waiting for a feature can be ignored" from Avi
2019-03-20 12:40:45 +02:00
Asias He
0eb2ea8f00 gossiper: Enable features only after gossip is settled
n1, n2, n3 in the cluster,

shutdown n1, n2, n3

start n1, n2

start n3, we saw features are enabled using the system table while n1 and n2 are already up and running in the cluster.

INFO  2019-02-27 09:24:41,023 [shard 0] gossip - Feature check passed. Local node 127.0.0.3 features = {CORRECT_COUNTER_ORDER, CORRECT_NON_COMPOUND_RANGE_TOMBSTONES, COUNTERS, DIGEST_MULTIPARTITION_READ, INDEXES, LARGE_PARTITIONS, LA_SSTABLE_FORMAT, MATERIALIZED_VIEWS, MC_SSTABLE_FORMAT, RANGE_TOMBSTONES, ROLES, ROW_LEVEL_REPAIR, SCHEMA_TABLES_V3, STREAM_WITH_RPC_STREAM, TRUNCATION_TABLE, WRITE_FAILURE_REPLY, XXHASH}, Remote common_features = {CORRECT_COUNTER_ORDER, CORRECT_NON_COMPOUND_RANGE_TOMBSTONES, COUNTERS, DIGEST_MULTIPARTITION_READ, INDEXES, LARGE_PARTITIONS, LA_SSTABLE_FORMAT, MATERIALIZED_VIEWS, MC_SSTABLE_FORMAT, RANGE_TOMBSTONES, ROLES, ROW_LEVEL_REPAIR, SCHEMA_TABLES_V3, STREAM_WITH_RPC_STREAM, TRUNCATION_TABLE, WRITE_FAILURE_REPLY, XXHASH}
INFO  2019-02-27 09:24:41,025 [shard 0] storage_service - Starting up server gossip
INFO  2019-02-27 09:24:41,063 [shard 0] gossip - Node 127.0.0.1 does not contain SUPPORTED_FEATURES in gossip, using features saved in system table, features={CORRECT_COUNTER_ORDER, CORRECT_NON_COMPOUND_RANGE_TOMBSTONES, COUNTERS, DIGEST_MULTIPARTITION_READ, INDEXES, LARGE_PARTITIONS, LA_SSTABLE_FORMAT, MATERIALIZED_VIEWS, MC_SSTABLE_FORMAT, RANGE_TOMBSTONES, ROLES, ROW_LEVEL_REPAIR, SCHEMA_TABLES_V3, STREAM_WITH_RPC_STREAM, TRUNCATION_TABLE, WRITE_FAILURE_REPLY, XXHASH}
INFO  2019-02-27 09:24:41,063 [shard 0] gossip - Node 127.0.0.2 does not contain SUPPORTED_FEATURES in gossip, using features saved in system table, features={CORRECT_COUNTER_ORDER, CORRECT_NON_COMPOUND_RANGE_TOMBSTONES, COUNTERS, DIGEST_MULTIPARTITION_READ, INDEXES, LARGE_PARTITIONS, LA_SSTABLE_FORMAT, MATERIALIZED_VIEWS, MC_SSTABLE_FORMAT, RANGE_TOMBSTONES, ROLES, ROW_LEVEL_REPAIR, SCHEMA_TABLES_V3, STREAM_WITH_RPC_STREAM, TRUNCATION_TABLE, WRITE_FAILURE_REPLY, XXHASH}

The problem is we enable the features too early in the start up process.
We should enable features after gossip is settled.

Fixes #4289
Message-Id: <04f2edb25457806bd9e8450dfdcccc9f466ae832.1551406991.git.asias@scylladb.com>

(cherry picked from commit 71bf757b2c)
2019-03-20 17:28:20 +08:00
Tomasz Grabiec
20eaf0b85f Merge "Fix empty remote common_features in check_knows_remote_features" from Asias
Three nodes in the cluster node1, node2, node3

Shutdown the whole cluster

Start node1

Start node2, node2 sees empty remote common_features.

   gossip - Feature check passed.  Local node 127.0.0.2 features =
   {CORRECT_COUNTER_ORDER, CORRECT_NON_COMPOUND_RANGE_TOMBSTONES, COUNTERS,
   DIGEST_MULTIPARTITION_READ, INDEXES, LARGE_PARTITIONS,
   LA_SSTABLE_FORMAT, MATERIALIZED_VIEWS, MC_SSTABLE_FORMAT,
   RANGE_TOMBSTONES, ROLES, ROW_LEVEL_REPAIR, SCHEMA_TABLES_V3,
   STREAM_WITH_RPC_STREAM, WRITE_FAILURE_REPLY, XXHASH},
   Remote common_features = {}

The problem is node3 hasn't started yet, node1 sees node3 has empty
features. In get_supported_features(), an empty common features will be
returned if an empty features of a node is seen. To fix, we should
fallback to use the features saved in system table.

Start node3, node3 sees empty remote common_features.

   gossip - Feature check passed. Local node 127.0.0.3 features =
   {CORRECT_COUNTER_ORDER, CORRECT_NON_COMPOUND_RANGE_TOMBSTONES, COUNTERS,
   DIGEST_MULTIPARTITION_READ, INDEXES, LARGE_PARTITIONS,
   LA_SSTABLE_FORMAT, MATERIALIZED_VIEWS, MC_SSTABLE_FORMAT,
   RANGE_TOMBSTONES, ROLES, ROW_LEVEL_REPAIR, SCHEMA_TABLES_V3,
   STREAM_WITH_RPC_STREAM, WRITE_FAILURE_REPLY, XXHASH},
   Remote common_features = {}

The problem is node3 hasn't inserted its own features into gossip
endpoint_state_map. get_supported_features() returns the common features
of all nodes in endpoint_state_map. To fix, we should fallback to use
the features stored in the system table for such node in this case.

Fixes #4225
Fixes #4341

* dev asias/fix_check_knows_remote_features.upstream.v4.1:
  gossiper: Remove unused register_feature and unregister_feature
  gossiper: Remove unused wait_for_feature_on_all_node and
    wait_for_feature_on_node
  gossiper: Log feature is enabled only if the feature is not enabled
    previously
  gossiper: Fix empty remote common_features in
    check_knows_remote_features

(cherry picked from commit b0e6f17a22)
2019-03-20 17:27:30 +08:00
Tomasz Grabiec
751fdc9f6c Merge "Fix window during init where waiting for a feature can be ignored" from Avi
storage_service keeps a bunch of "feature" variables, indicating cluster-wide
supported features, and has the ability to wait until the entire cluster supports
a given feature.

The propagation of features depends on gossip, but gossip is initialized after
storage_service, so the current code late-initializes the features. However, that
means that whoever waits on a feature between storage_service initialization and
gossip initialization loses their wait entry. In #3952, we have proof that this
in fact happens.

Fix this by removing the circular dependency. We now store features in a new
service, feature_service, that is started before both gossip and storage_service.
Gossip updates feature_service while storage_service reads for it.

Fixes #3953.

* https://github.com/avikivity/3953/v4.1:
  storage_service: deinline enable_all_features()
  gossiper: keep features registered
  tests/gossip: switch to seastar::thread
  storage_service: deinline init/deinit functions
  gossiper: split feature storage into a new feature_service
  gossiper: maybe enable features after start_gossiping()
  storage_service: fix gap when feature::when_enabled() doesn't work

(cherry picked from commit 6012a63660)
2019-03-20 17:25:35 +08:00
Tomasz Grabiec
5e3a52024e sstable/compaction: Use correct schema in the writing consumer
Introduced in 2a437ab427.

regular_compaction::select_sstable_writer() creates the sstable writer
when the first partition is consumed from the combined mutation
fragment stream. It gets the schema directly from the table
object. That may be a different schema than the one used by the
readers if there was a concurrent schema alter duringthat small time
window. As a result, the writing consumer attached to readers will
interpret fragments using the wrong version of the schema.

One effect of this is storing values of some columns under a different
column.

This patch replaces all column_family::schema() accesses with accesses
to the _schema memeber which is obtained once per compaction and is
the same schema which readers use.

Fixes #4304.

Tests:

  - manual tests with hard-coded schema change injection to reproduce the bug
  - build/dev/scylla boot
  - tests/sstable_mutation_test

Message-Id: <1551698056-23386-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit 58e7ad20eb)
2019-03-04 18:16:43 +02:00
Avi Kivity
3869b5ab51 Merge "Fix commitlog chunks overwriting each other" from Paweł
"
This series fixes a problem in the commitlog cycle() function that
confused in-memory and on-disk size of chunks it wrote to disk. The
former was used to decide how much data needs to be actually written,
and the latter was used to compute the offset of the next chunk. If two
chunk writes happened concurrently one the one positioned earlier in
the file could corrupt the header of the next one.

Fixes #4231.

Tests: unit(dev), dtest(commitlog_test.py:TestCommitLog.test_commitlog_replay_on_startup,test_commitlog_replay_with_alter_table)
"

* tag 'fix-commitlog-cycle/v1' of https://github.com/pdziepak/scylla:
  commitlog: write the correct buffer size
  utils/fragmented_temporary_buffer_view: add remove suffix

(cherry picked from commit d95dec22d9)
2019-03-04 17:58:46 +02:00
Jenkins
3cca6f5384 release: prepare for 3.0.4 by hagitsegev 2019-03-04 10:14:33 +02:00
Nadav Har'El
15188b5ea5 Materialized views: fix accidental zeroing of flow-control delay
The materialized-views flow control carefully calculates an amount of
microseconds to delay a client to slow it down to the desired rate -
but then a typo (std::min instead of std::max) causes this delay to
be zeroed, which in effect completely nullifies the flow control
algorithm.

Before this fix, experiments suggested that view flow control was
not having any effect and view backlog not bounded at all. After this
fix, we can see the flow control having its desired effect, and the
view backlog converging.

Fixes #4143.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20190226161452.498-1-nyh@scylladb.com>
(cherry picked from commit da54d0fc7d)
2019-02-27 22:17:44 +02:00
Avi Kivity
96d9ebb67e Update seastar submodule
* seastar b79be02...4a06029 (1):
  > net: fix tcp load balancer accounting leak while moving socket to other shard

Fixes #4269.
2019-02-25 23:22:09 +02:00
Avi Kivity
f18b370198 Merge "sstables: mc: writer: Avoid large allocations for keeping promoted index entries" from Tomasz
"
Currently we keep the entries in a circular_buffer, which uses
a contiguous storage. For large partitions with many promoted index
entries this can cause OOM and sstable compaction failure.

A similar problem exists for the offset vector built
in write_promoted_index().

This change solves the problem by serializing promoted index entries
and the offset vector on the fly directly into a bytes_ostream, which
uses fragmented storage.

The serialization of the first entry is deferred, so that
serialization is avoided if there will be less than 2
entries. Promoted index is not added for such partitions.

There still remains a problem that large-enough promoted index can cause OOM.

Refs #4217

Tests:
  - unit (release)
  - scylla-bench write

Branches: 3.0
"

* tag 'fix-large-alloc-for-promoted-index-v3' of github.com:tgrabiec/scylla:
  sstables: mc: writer: Avoid large allocations for maintaining promoted index
  sstables: mc: writer: Avoid double-serialization of the promoted index

(cherry picked from commit fdefee696e)
2019-02-24 15:45:32 +02:00
Avi Kivity
8e657e5685 Merge " Fix INSERT JSON with null values" from Piotr
"
Fixes #4256

This miniseries fixes a problem with inserting NULL values through
INSERT JSON interface.

Tests: unit (dev)
"

* 'fix_insert_json_with_null' of https://github.com/psarna/scylla:
  tests: add test for INSERT JSON with null values
  cql3: add missing value erasing to json parser

(cherry picked from commit 5520fc37ba)
2019-02-22 15:52:46 +02:00
Avi Kivity
4fde670abf Merge "Add DEFAULT UNSET support to JSON" from Piotr
"
This series adds DEFAULT UNSET and DEFAULT NULL keyword support
to INSERT JSON statement, as stated in #3909.

Tests: unit (release)
"

* 'add_json_default_unset_2' of https://github.com/psarna/scylla:
  tests: add DEFAULT UNSET case to JSON cql tests
  tests: split JSON part of cql query test
  cql3: add DEFAULT UNSET to INSERT JSON

(cherry picked from commit 447f953a2c)
2019-02-22 15:52:16 +02:00
Avi Kivity
923318e636 sstables: checksummed_file_writer: fix dma alignment
checksummed_file_writer does not override allocate_buffer(), so it inherits
data_source_impl's default allocate_buffer, which does not care about alignment.
The buffer is then passed to the real file_data_sink_impl, and thence to the file
itself, which cannot complete the write since it is not properly aligned.

This doesn't fail in release mode, since the Seastar allocator will supply a
properly aligned buffer even if not asked to do so. The ASAN allocator usually
does supply an aligned buffer, but not always, which causes the test to fail.

Fix by forwarding the allocate_buffer() function to the underlying data_source.

Fixes #4262.
Branches: branch-3.0
Message-Id: <20190221184115.6695-1-avi@scylladb.com>

(cherry picked from commit 34b254381f)
2019-02-22 09:28:55 +02:00
Amnon Heiman
35cc09b150 scylla-housekeeping: Read JSON as UTF-8 string for older Python 3 compatibility
Python 3.6 is the first version to accept bytes to the json.loads(),
which causes the following error on older Python 3 versions:

  Traceback (most recent call last):
    File "/usr/lib/scylla/scylla-housekeeping", line 175, in <module>
      args.func(args)
    File "/usr/lib/scylla/scylla-housekeeping", line 121, in check_version
      raise e
    File "/usr/lib/scylla/scylla-housekeeping", line 116, in check_version
      versions = get_json_from_url(version_url + params)
    File "/usr/lib/scylla/scylla-housekeeping", line 55, in get_json_from_url
      return json.loads(data)
    File "/usr/lib64/python3.4/json/__init__.py", line 312, in loads
      s.__class__.__name__))
  TypeError: the JSON object must be str, not 'bytes'

To support those older Python versions, convert the bytes read to utf8
strings before calling the json.loads().

Fixes #4239
Branches: master, 3.0

Signed-off-by: Amnon Heiman <amnon@scylladb.com>
Message-Id: <20190218112312.24455-1-amnon@scylladb.com>
(cherry picked from commit 750b76b1de)
2019-02-20 11:03:12 +02:00
Asias He
22f41f04ba range_streamer: Futurize add_ranges
It might take long time for get_all_ranges_with_sources_for and
get_all_ranges_with_strict_sources_for to calculate which cause reactor
stall. To fix, run them in a thread and yield. Those functions are used in
the slow path, it is ok to yield more than needed.

Fixes #3639

Message-Id: <63aa7794906ac020c9d9b2984e1351a8298a249b.1536135617.git.asias@scylladb.com>
(cherry picked from commit 8edf3defdf)
2019-02-20 11:03:12 +02:00
Avi Kivity
fae11c0d6b fragmented_temporary_buffer: fix read_exactly() during premature end-of-stream
read_exactly(), when given a stream that does not contain the amount of data
requested, will loop endlessly, allocating more and more memory as it does, until
it fails with an exception (at which point it will release the memory).

Fix by returning an empty result, like input_stream::read_exactly() (which it
replaces). Add a test case that fails without a fix.

Affected callers are the native transport, commitlog replay, and internal
deserialization.

Fixes #4233.

Branches: master, branch-3.0
Tests: unit(dev)
Message-Id: <20190216150825.14841-1-avi@scylladb.com>
(cherry picked from commit 03531c2443)
2019-02-20 11:03:11 +02:00
Nadav Har'El
82016c07f2 Materialized views: limit size of row batching during bulk view building
The bulk materialized-view building processes (when adding a materialized
view to a table with existing data) currently reads the base table in
batches of 128 (view_builder::batch_size) rows. This is clearly better
than reading entire partitions (which may be huge), but still, 128 rows
may grow pretty large when we have rows with large strings or blobs,
and there is no real reason to buffer 128 rows when they are large.

Instead, when the rows we read so far exceed some size threshold (in this
patch, 1MB), we can operate on them immediately instead of waiting for
128.

As a side-effect, this patch also solves another bug: At worst case, all
the base rows of one batch may be written into one output view partition,
in one mutation. But there is a hard limit on the size of one mutation
(commitlog_segment_size_in_mb, by default 32MB), so we cannot allow the
batch size to exceed this limit. By not batching further after 1MB,
we avoid reaching this limit when individual rows do not reach it but
128 of them did.

Fixes #4213.

This patch also includes a unit test reproducing #4213, and demonstrating
that it is now solved.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20190214093424.7172-1-nyh@scylladb.com>
(cherry picked from commit fec562ec8f)
2019-02-16 21:54:41 +02:00
Botond Dénes
282ccbb072 service/storage_service: fix pre-bootstrap wait for schema agreement
When bootstrapping, a node should to wait to have a schema agreement
with its peers, before it can join the ring. This is to ensure it can
immediately accept writes. Failing to reach schema agreement before
joining is not fatal, as the node can pull unknown schemas on writes
on-demand. However, if such a schema contains references to UDFs, the
node will reject writes using it, due to #3760.

To ensure that schema agreement is reached before joining the ring,
`storage_service::join_token_ring()` has to checks. First it checks that
at least one peer was connected previously. For this it compares
`database::get_version()` with `database::empty_version`. The (implied)
assumption is that this will become something other than
`database::empty_version` only after having connected (and pulled
schemas from) at least one peer. This assumption doesn't hold anymore,
as we now set the version earlier in the boot process.
The second check verifies that we have the same schema version as all
known, live peers. This check assumes (since 3e415e2) that we have
already "met" all (or at least some) of our peers and if there is just
one known node (us) it concludes that this is a single-node cluster,
which automatically has schema agreement.
It's easy to see how these two checks will fail. The first fails to
ensure that we have met our peers, and the second wrongfully concludes
that we are a one-node cluster, and hence have schema agreement.

To fix this, modify the first check. Instead of relying on the presence
of a non-empty database version, supposedly implying that we already
talked to our peers, explicitely make sure that we have really talked to
*at least* one other node, before proceeding to the second check, which
will now do the correct thing, actually checking the schema versions.

Fixes: #4196

Branches: 3.0, 2.3

Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <40b95b18e09c787e31ba6c5519fb64d68b4ca32e.1550228389.git.bdenes@scylladb.com>
(cherry picked from commit 2125e99531)
2019-02-16 19:04:08 +02:00
Avi Kivity
873e0f0e14 auth: password_authenticator: protect against NULL salted_hash
In case salted_hash was NULL, we'd access uninitialized memory when dereferencing
the optional in get_as<>().

Protect against that by using get_opt() and failing authentication if we see a NULL.

Fixes #4168.

Tests: unit (release)
Branches: 3.0, 2.3
Message-Id: <20190211173820.8053-1-avi@scylladb.com>
(cherry picked from commit da9628c6dc)
2019-02-11 23:55:53 +02:00
Jenkins
c36c58c64e release: prepare for 3.0.3 by hagitsegev 2019-02-11 18:58:02 +02:00
Duarte Nunes
0685c8f5bc Merge 'Fix misdetection of remote counter shards' from Paweł
"
The code reading counter cells form sstables verifies that there are no
unsupported local or remote shards. The latter are detected by checking
if all shards are present in the counter cell header (only remote shards
do not have entries there). However, the logic responsible for doing
that was incorrectly computing the total number of counter shards in a
cell if the header was larger than a single counter shard. This resulted
in incorrect complaints that remote shards are present.

Fixes #4206

Tests: unit(release)
"

* tag 'counter-header-fix/v1' of https://github.com/pdziepak/scylla:
  tests/sstables: test counter cell header with large number of shards
  sstables/counters: fix remote counter shard detection

(cherry picked from commit d2d885fb93)
2019-02-11 14:09:55 +02:00
Calle Wilund
92cf2934c6 tls: Use a default prio string disabling TLS1.0 forcing min 128bits
Fixes #4010

Unless user sets this explicitly, we should try explicitly avoid
deprecated protocol versions. While gnutls should do this for
connections initiated thusly, clients such as drivers etc might
use obsolete versions.

Message-Id: <20190107131513.30197-1-calle@scylladb.com>
(cherry picked from commit ba6a8ef35b)
2019-02-05 19:45:13 +02:00
Calle Wilund
ed2fb65732 commitlog_replayer: Bugfix: finding truncation positions uses local var ref
"uuid" was ref:ed in a continuation. Works 99.9% of the time because
the continuation is not actually delayed (and assuming we begin the
checks with non-truncated (system) cf:s it works).
But if we do delay continuation, the resulting cf map will be
borked.

Fixes #4187.

Message-Id: <20190204141831.3387-1-calle@scylladb.com>
(cherry picked from commit 9cadbaa96f)
2019-02-04 20:25:17 +02:00
Gleb Natapov
ce2957d106 messaging_service: do not forget to close stream when sending it to another side failed
Fixes #4124

Message-Id: <20190131091857.GC3172@scylladb.com>
(cherry picked from commit a70374d982)
2019-02-03 13:00:20 +02:00
Avi Kivity
b31d94e317 Update seastar submodule
* seastar 5226277...b79be02 (1):
  > rpc: support closing streaming when only sink or source was created

Ref #4124.
2019-02-03 12:59:43 +02:00
Asias He
da80f27f44 migration_manager: Fix nullptr dereference in maybe_schedule_schema_pull
Commit 976324bbb8 changed to use
get_application_state_ptr to get a pointer of the application_state. It
may return nullptr that is dereferenced unconditionally.

In resharding_test.py:ReshardingTest_nodes4_with_SizeTieredCompactionStrategy.resharding_by_smp_increase_test, we saw:

   4 nodes in the tests

   n1, n2, n3, n4 are started

   n1 is stopped

   n1 is changed to use different shard config

   n1 is restarted ( 2019-01-27 04:56:00,377 )

The backtrace happened on n2 right fater n1 restarts:

   0 INFO 2019-01-27 04:56:05,175 [shard 0] gossip - Feature STREAM_WITH_RPC_STREAM is enabled
   1 INFO 2019-01-27 04:56:05,175 [shard 0] gossip - Feature WRITE_FAILURE_REPLY is enabled
   2 INFO 2019-01-27 04:56:05,175 [shard 0] gossip - Feature XXHASH is enabled
   3 WARN 2019-01-27 04:56:05,177 [shard 0] gossip - Fail to send EchoMessage to 127.0.58.1: seastar::rpc::closed_error (connection is closed)
   4 INFO 2019-01-27 04:56:05,205 [shard 0] gossip - InetAddress 127.0.58.1 is now UP, status =
   5 Segmentation fault on shard 0.
   6 Backtrace:
   7 0x00000000041c0782
   8 0x00000000040d9a8c
   9 0x00000000040d9d35
   10 0x00000000040d9d83
   11 /lib64/libpthread.so.0+0x00000000000121af
   12 0x0000000001a8ac0e
   13 0x00000000040ba39e
   14 0x00000000040ba561
   15 0x000000000418c247
   16 0x0000000004265437
   17 0x000000000054766e
   18 /lib64/libc.so.6+0x0000000000020f29
   19 0x00000000005b17d9

We do not know when this backtrace happened, but according to log from n3 an n4:

   INFO 2019-01-27 04:56:22,154 [shard 0] gossip - InetAddress 127.0.58.2 is now DOWN, status = NORMAL
   INFO 2019-01-27 04:56:21,594 [shard 0] gossip - InetAddress 127.0.58.2 is now DOWN, status = NORMAL

We can be sure the backtrace on n2 happened before 04:56:21 - 19 seconds (the
delay the gossip notice a peer is down), so the abort time is around 04:56:0X.
The migration_manager::maybe_schedule_schema_pull that triggers the backtrace
must be scheduled before n1 is restarted, because it dereference
application_state pointer after it sleeps 60 seconds, so the time
maybe_schedule_schema_pull is called is around 04:55:0X which is before n1 is
restarted.

So my theory is: migration_manager::maybe_schedule_schema_pull is scheduled, at this time
n1 has SCHEMA application_state, when n1 restarts, n2 gets new application
state from n1 which does not have SCHEMA yet, when migration_manager::maybe_schedule
wakes up from the 60 sleep, n1 has non-empty endpoint_state but empty
application_state for SCHEMA. We dereference the nullptr
application_state and abort.

Fixes: #4148
Tests: resharding_test.py:ReshardingTest_nodes4_with_SizeTieredCompactionStrategy.resharding_by_smp_increase_test
Message-Id: <9ef33277483ae193a49c5f441486ee6e045d766b.1548896554.git.asias@scylladb.com>
(cherry picked from commit 28d6d117d2)
2019-02-01 13:00:38 +02:00
Jenkins
5174b1cd13 release: prepare for 3.0.2 by slivne 2019-01-30 16:14:34 +02:00
Nadav Har'El
9ba608cae4 cql3: really ensure retrieval of columns for filtering
Commit fd422c954e aimed to fix
issue #3803. In that issue, if a query SELECTed only certain columns but
did filtering (ALLOW FILTERING) over other unselected columns, the filtering
didn't work. The fix involved adding the columns being filtered to the set
of columns we read from disk, so they can be filtered.

But that commit included an optimization: If you have clustering keys
c1 and c2, and the query asks for a specific partition key and c1 < 3 and
c2 > 3, the "c1 < 3" part does NOT need to be filtered because it is already
done as a slice (a contiguous read from disk). The committed code erroneously
concluded that both c1 and c2 don't need to be filtered, which was wrong
(c2 *does* need to be read and filtered).

In this patch, we fix this optimization. Previously, we used the "prefix
length", which in the above example was 2 (both c1 and c2 were filtered)
but we need a new and more elaborate function,
num_prefix_columns_that_need_not_be_filtered(), to determine we can only
skip filtering of 1 (c1) and cannot skip the second.

Fixes #4121. This patch also adds a unit test to confirm this.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>

Message-Id: <20190123131212.6269-1-nyh@scylladb.com>
(cherry picked from commit 76f1fcc346)
2019-01-23 21:11:05 +02:00
Avi Kivity
f7c5cbc645 build: fix libdeflate object file corruption during parallel build
libdeflate's build places some object files in the source directory, which is
shared between the debug and release build. If the same object file (for the two
modes) is written concurrently, or if one more reads it while the other writes it,
it will be corrupted.

Fix by not building the executables at all. They aren't needed, and we already
placed the libraries' objects in the build directory (which is unshared). We only
need the libraries anyway.

Fixes #4130.
Branches: master, branch-3.0
Message-Id: <20190123145435.19049-1-avi@scylladb.com>

(cherry picked from commit c83ae62aed)
2019-01-23 21:11:05 +02:00
Duarte Nunes
cf4b4d4878 Merge 'hinted handoff: cache cf mappings' from Vlad
"
Cache cf mappings when breaking in the middle of a segment sending so
that the sender has them the next time it wants to send this segment
for where it left off before.

Also add the "discard" metric so that we can track hints that are being
discarded in the send flow.
"

Fixes #4122

* 'hinted_handoff_cache_cf_mappings-v1' of https://github.com/vladzcloudius/scylla:
  hinted handoff: cache column family mappings for segments that were not sent out in full
  hinted handoff: add a "discarded" metric

(cherry picked from commit 88c7c1e851)
2019-01-23 17:14:29 +02:00
Asias He
45bb1ba1b7 streaming: Futurize estimate_partitions
The loop can take a long time if the number of sstables and/or ranges
are large. To fix, futurize the loop.

Fixes: #4005

Message-Id: <3b05cb84f3f57cc566702142c6365a04b075018e.1545290730.git.asias@scylladb.com>
(cherry picked from commit bcba6b4f4d)
2019-01-22 18:20:21 +02:00
Botond Dénes
28294ed42e auth/service: unregister migration listener on stop()
Otherwise any event that triggers notification to this listener would
trigger a heap-use-after-free.

Refs: #4107

Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <b6bbd609371a2312aed7571b05119d59c7d103d7.1548067626.git.bdenes@scylladb.com>
(cherry picked from commit f229dff210)
2019-01-22 17:54:36 +02:00
Jenkins
3c4f8cf6ed release: prepare for 3.0.1 by hagitsegev 2019-01-20 12:42:00 +02:00
Botond Dénes
7b94264ae5 mutlishard_mutation_query(): use correct reader concurrency semaphore
The multishard mutation query used the semaphore obtained from
`database::user_read_concurrency_sem()` to pause-resume shard readers.
This presented a problem when `multishard_mutation_query()` was reading
from system tables. In this case the readers themselves would obtain
their permits from the system read concurrency semaphore. Since the
pausing of shard readers used the user read semaphore, pausing failed to
fulfill its objective of alleviating pressure on the semaphore the reads
obtained their permits from. In some cases this lead to a deadlock
during system reads.
To ensure the correct semaphore is used for pausing-resuming readers,
obtain the semaphore from the `table` object. To avoid looking up the
table on every pause or resume call, cache the semaphores when readers
are created.

Fixes: #4096

Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <c784a3cd525ce29642d7216fbe92638fa7884e88.1547729119.git.bdenes@scylladb.com>
(cherry picked from commit 4537ec7426)
2019-01-17 18:08:01 +02:00
Duarte Nunes
22a085fbd3 Merge 'Fix filtering with LIMIT and paging' from Piotr
"
Before this series the limit was applied per page instead
of globally, which might have resulted in returning too many
rows.

To fix that:
 1. restrictions filter now has a 'remaining' parameter
    in order to stop accepting rows after enough of them
    have already been accepted
 2. pager passes its row limit to restrictions filter,
    so no more rows than necessary will be served to the client
 3. results no longer need to be trimmed on select_statement
    level

Tests: unit (release)
"

Fixes #4100

* 'fix_filtering_limit_with_paging_3' of https://github.com/psarna/scylla:
  tests: add filtering+limit+paging test case
  tests: allow null paging state in filtering tests
  cql3: fix filtering with LIMIT with regard to paging

(cherry picked from commit 7505815013)
2019-01-17 18:07:41 +02:00
Tomasz Grabiec
2d181da656 row_cache: Fix crash on memtable flush with LCS
Presence checker is constructed and destroyed in the standard
allocator context, but the presence check was invoked in the LSA
context. If the presence checker allocates and caches some managed
objects, there will be alloc-dealloc mismatch.

That is the case with LeveledCompactionStrategy, which uses
incremental_selector.

Fix by invoking the presence check in the standard allocator context.

Fixes #4063.

Message-Id: <1547547700-16599-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit 32f711ce56)
2019-01-15 21:16:13 +02:00
Nadav Har'El
d427a23d42 scylla_util.py: make view_hints_directory setting optional
It is optional to set "view_hints_directory", so we shouldn't insist that
it is defined in scylla.yaml on upgrade.

Fixes #4091.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20190114125225.10794-1-nyh@scylladb.com>
(cherry picked from commit 9062750089)
2019-01-14 16:59:40 +02:00
Shlomi Livne
37ab553f02 release: prepare for 3.0.0
Signed-off-by: Shlomi Livne <shlomi@scylladb.com>
2019-01-12 22:24:08 +02:00
Raphael S. Carvalho
6a3f4fb3f9 database: Fix race condition in sstable snapshot
Race condition takes place when one of the sstables selected by snapshot
is deleted by compaction. Snapshot fails because it tries to link a
sstable that was previously unlinked by compaction's sstable deletion.

Refs #4051.

(master commit 1b7cad3531)
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Message-Id: <20190110194048.26051-1-raphaelsc@scylladb.com>
2019-01-11 13:48:12 +02:00
Avi Kivity
8168d13887 Merge "Fix UDTs representation in serialization header" from Piotr
"
Tests: unit(release)
"

Fixes #4073.

* commit 'FETCH_HEAD~1':
  Add test for serialization header with UDT
  Fix UDT names in serialization header

(cherry picked from commit 4a6aeced59)
2019-01-11 07:48:23 +02:00
Benny Halevy
13bdec6eb4 sstables: mc: sign-extend serialization_header min_local_deletion_time_base and min_ttl_base
Refs #4074
Refs #3353

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20190110141439.1324-1-bhalevy@scylladb.com>
(cherry picked from commit 2dc3776407)
2019-01-11 07:47:45 +02:00
Benny Halevy
57e7081d86 sstables: mc: sign-extend delta local_deletion_time and delta ttl
Follow Cassandra's encoding so that values that are less than the
baseline encoding_stats will wrap-around in 64-bits rather tham 32.

Fixes #4074
Refs #3353

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20190109192703.18371-1-bhalevy@scylladb.com>
(cherry picked from commit 60323b79d1)
2019-01-09 23:16:00 +02:00
Avi Kivity
2fcae36d96 tests: mutation_source_test: generate valid utf-8 data
test_fast_forwarding_across_partitions_to_empty_range uses an uninitialized
string to populate an sstable, but this can be invalid utf-8 so that sstable
cannot be sstabledumped.

Make it valid by using make_random_string().

Fixes #4040.
Message-Id: <20190107193240.14409-1-avi@scylladb.com>

(cherry picked from commit d8adbeda11)
2019-01-08 14:53:55 +02:00
Avi Kivity
ba62dcd5c7 Update seastar submodule
* seastar 618bc23...5226277 (1):
  > iotune: Initialize io_rates member variables

Fixes #4064.
2019-01-08 11:39:50 +02:00
Nadav Har'El
515399ce17 materialized views: move hints to top-level directory
While we keep ordinary hints in a directory parallel to the data directory,
we decided to keep the materialized view hints in a subdirectory of the data
directory, named "view_pending_updates". But during boot, we expect all
subdirectories of data/ to be keyspace names, and when we notice this one,
we print a warning:

   WARN: database - Skipping undefined keyspace: view_pending_updates

This spurious warning annoyed users. But moreover, we could have bigger
problems if the user actually tries to create a keyspace with that name.

So in this patch, we move the view hints to a separate top-level directory,
which defaults to /var/lib/scylla/view_hints, but as usual can be configured.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20190107142257.16342-1-nyh@scylladb.com>
(cherry picked from commit da090a5458)
2019-01-07 22:01:56 +02:00
Benny Halevy
772c4b5fdc sstables: mc: expired_liveness_ttl should be max int32_t rather than max uint32_t
Corresponding to Cassandra's EXPIRED_LIVENESS_TTL = Integer.MAX_VALUE;

Fixes #4060

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20190107172457.20430-1-bhalevy@scylladb.com>
(cherry picked from commit 40410465d7)
2019-01-07 21:59:59 +02:00
Avi Kivity
874d88c98d Update seastar submodule
* seastar 08f1258...618bc23 (1):
  > perftune.py: tune only active NVMe HW queues on i3 AWS instances

Ref #3831.
2019-01-06 12:59:22 +02:00
Avi Kivity
5a178ff635 compaction_controller: increase minimum shares to 50 (~5%) for small-data workloads
The workload in #3844 has these characteristics:
 - very small data set size (a few gigabytes per shard)
 - large working set size (all the data, enough for high cache miss rate)
 - high overwrite rate (so a compaction results in 12X data reduction)

As a result, the compaction backlog controller assigns very few shares to
compaction (low data set size -> low backlog), so compaction proceeds very slowly.
Meanwhile, we have tons of cache misses, and each cache miss needs to read from a
large number of sstables (since compaction isn't progressing). The end result is
a high read amplification, and in this test, timeouts.

While we could declare that the scenario is very artificial, there are other
real-world scenarios that could trigger it. Consider a 100% write load
(population phase) followed by 100% read. Towards the end of the last compaction,
the backlog will drop more and more until compaction slows to a crawl, and until
it completes, all the data (for that compaction) will have to be read from its
input sstables, resulting in read amplification.

We should probably have read amplification affect the backlog, but for now the
simpler solution is to increase the minimum shares to 50 so that compaction
always makes forward progress. This will result in higher-than-needed compaction
bandwidth in some low write rate scenarios so we will see fluctuations in request
rate (what the controller was designed to avoid), but these fluctioations will be
limited to 5%.

Since the base class backlog_controller has a fixed (0, 0) point, remove it
and add it to derived classes (setting it to (0, 50) for compaction).

Fixes #3844 (or at least improves it).
Message-Id: <20181231162710.29410-1-avi@scylladb.com>

(cherry picked from commit b0980ba7c6)
2019-01-04 13:28:43 +02:00
Avi Kivity
d67439b910 Revert "release: prepare for 3.0-rc4"
This reverts commit 21a5a4c76a. we were already
at rc4, and the commit only changes the syntax (from the incorrect one to
the correct one).
2019-01-04 12:34:38 +02:00
Hagit Segev
21a5a4c76a release: prepare for 3.0-rc4 2019-01-03 23:53:47 +02:00
Tomasz Grabiec
f818d6ee3f tests: cql_test_env: Start the compaction manager
Broken in fee4d2e

Not doing this results in compaction requests being ignored.

One effect of this is that perf_fast_forward produces many sstables instead of one.

Refs #3984
Refs #3983

Message-Id: <1544719540-10178-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit 245a0d953a)
2019-01-03 14:56:42 +01:00
Tomasz Grabiec
20c2745592 Merge "Improve times to start / stop the nodes" from Glauber
If the compaction manager is started, compactions may start (this is
regardless of whether or not we trigger them). The problem with that is
that they start at a time in which we are flushing the commitlog and the
initialization procedure waits for the commitlog to be fully flushed and
the resulting memtables flushed before we move on.

Because there are no incoming writes, the amount of shares in memtable
flushes decrease as memory used decreases and that can cause the startup
procedure to take a long time.

We have recently started to bump the shares manually for manual flushes.
While that guarantees that we will not drive the shares to zero, I will
make the argument that we can do better by making sure that those things
are, at this point, running alone: user experience is affected by
startup times and the bump we give to user-triggered operations will
only do so much. Even if we increase the shares a lot flushes will still
be fighting for resources with compactions and startup will take longer
than it could.

By making sure that flushes are this point running alone we improve the
user experience by making sure the startup is as fast as it can be.

There is a similar problem at the drain level, which is also fixed in this
series.

Fixes #3958

* git@github.com:glommer/scylla.git faster-restart
  compaction_manager: delay initialization of the compaction manager.
  drain: stop compactions early

(cherry picked from commit 3e70ae1d06)
2019-01-03 14:56:16 +01:00
Avi Kivity
cf5c72561c release: prepare for scylla-3.0-rc4 2019-01-03 13:15:58 +02:00
Botond Dénes
53b85e5d32 querier_cache: unregister queriers evicted due to expired TTL
Currently queriers evicted due to their TTL expiring are not
unregistered from the `reader_concurrency_semaphore`. This can cause a
use-after-free when the semaphore tries to evict the same querier at
some later point in time, as the querier entry it has a pointer to is
now invalid.

Fix by unregistering the querier from the semaphore before destroying
the entry.

Refs: #4018
Refs: #4031

Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <4adfd09f5af8a12d73c29d59407a789324cd3d01.1546504034.git.bdenes@scylladb.com>
(cherry picked from commit e5a0ea390a)
2019-01-03 13:14:02 +02:00
Avi Kivity
2456cf63f2 querier_cache: unregister querier from reader_concurrency_semaphore during eviction
In insert_querier(), we may evict older queriers to make room for the new one.
However, we forgot to unregister the evicted queriers from
reader_concurrency_semaphore. As a result, when reader_concurrency_semaphore
eventually wanted to evict something, it saw an inactive_read_handle that was
not connected to a querier_cache::entry, and crashed on use-after-free.

Fix by evicting through the inactive_read_handle associated with the querier
to be evicted. This removes traces of the querier from both
reader_concurrency_semaphore and querier_cache. We also have to massage the
statistics since querier_inactive_read::evict() updates different counters.

Fixes #4018.

Tests: unit(release)
Reviewed-by: Botond Denes <bdenes@scylladb.com>
Message-Id: <20190102175023.26093-1-avi@scylladb.com>
(cherry picked from commit 918d255168)
2019-01-03 13:14:00 +02:00
Pekka Enberg
c1f6ce4251 Merge 'Fixes for the view_update_from_staging_generator' from Duarte
"This series contains a couple of fixes to the
view_update_from_staging_generator, the object responsible for
generating view updates from sstables written through streaming.

Fixes #4021"
* 'materialized-views/staging-generator-fixes/v2' of https://github.com/duarten/scylla:
  db/view/view_update_from_staging_generator: Break semaphore on stop()
  db/view/view_update_from_staging_generator: Restore formatting
  db/view/view_update_from_staging_generator: Avoid creating more than one fiber

(cherry picked from commit 96172b7bca)
2018-12-29 20:22:54 +02:00
Duarte Nunes
fc82eb5586 streaming/stream_session: Only stage sstables for tables with views
When streaming, sstables for which we need to generate view updates
are placed in a special staging directory. However, we only need to do
this for tables that actually have views.

Refs #4021
Message-Id: <20181227215412.5632-1-duarte@scylladb.com>

(cherry picked from commit bab7e6877b)
2018-12-28 20:52:15 +02:00
Avi Kivity
f58e592345 Merge "Fix use-after-free when destroying partition_snapshots in the background"from Tomasz
"
partition_snapshots created in the memtable will keep a reference to
the memtable (as region*) and to memtable::_cleaner. As long as the
reader is alive, the memtable will be kept alive by
partition_snapshot_flat_reader::_container_guard. But after that
nothing prevents it from being destroyed. The snapshot can outlive the
read if mutation_cleaner::merge_and_destroy() defers its destruction
for later. When the read ends after memtable was flushed, the snapshot
will be queued in the cache's cleaner, but internally will reference
memtable's region and cleaner. This will result in a use-after-free
when the snapshot resumes destruction.

The fix is to update snapshots's region and cleaner references at the
time of queueing to point to the cache's region and cleaner.

When memtable is destroyed without being moved to cache there is no
problem because the snapshot would be queued into memtable's cleaner,
which will be drained on destruction from all snapshots.

Introduced in f3da043 (in >= 3.0-rc1)

Fixes #4030.

Tests:

  - mvcc_test (debug)

"

* tag 'fix-snapshot-merging-use-after-free-v1.1' of github.com:tgrabiec/scylla:
  tests: mvcc: Add test_snapshot_merging_after_container_is_destroyed
  tests: mvcc: Introduce mvcc_container::migrate()
  tests: mvcc: Make mvcc_partition move-constructible
  tests: mvcc: Introduce mvcc_container::make_not_evictable()
  tests: mvcc: Allow constructing mvcc_container without a cache_tracker
  mutation_cleaner: Migrate partition_snapshots when queueing for background cleanup
  mvcc: partition_snapshot: Introduce migrate()
  mutation_cleaner: impl: Store a back-reference to the owning mutation_cleaner

(cherry picked from commit 8e2f6d0513)
2018-12-28 13:37:29 +02:00
Duarte Nunes
6375b1e5b7 streaming/stream_session: Don't use table reference across defer points
When creating a sstable from which to generate view updates, we held
on to a table reference across defer points. In case there's a
concurrent schema drop, the table object might be destroyed and we
will incur in a use-after-free. Solve this by holding on to a shared
pointer and pinning the table object.

Refs #4021

Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20181227105921.3601-1-duarte@scylladb.com>
(cherry picked from commit 66e45469b2)
2018-12-28 10:58:26 +02:00
Gleb Natapov
7ca24efb39 streaming: always read from rpc::source until end-of-stream during mutation sending
rpc::source cannot be abandoned until EOS is reached, but current code
does not obey it if error code is received, it throws exception instead that
aborts the reading loop. Fix it by moving exception throwing out of the
loop.

Fixes: #4025

Message-Id: <20181227135051.GC29458@scylladb.com>
(cherry picked from commit 37b4043677)
2018-12-27 18:59:59 +02:00
Avi Kivity
32ebaaa585 Update libdeflate submodule
* libdeflate 17ec6c9...e7e54ea (1):
  > build: improve out-of-tree build with multiple output trees

(cherry picked from commit d6a22c50cb)
2018-12-25 14:41:24 +02:00
Nadav Har'El
a88c722a4c build_ami.sh: need to check out the right branch of scylla-jmx
This patch is for branch 3.0's build_ami.sh.
It checks out the latest master branch of scylla-jmx, which not only
sounds wrong, it also doesn't work: the latest master of scylla-jmx
can only build a "relocatable package" but branch 3.0 doesn't work with
those.

This patch needs to be applied only in branch 3.0.

It should probably be made more general, though... build_ami.sh should
have been able to figure out what is the *current* branch, and if it is
branch-3.0 or next-3.0, check out branch-3.0 of the other repositories.
But I'm not sure how to do this correctly.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20181217214610.4498-1-nyh@scylladb.com>
2018-12-25 12:37:11 +02:00
Tomasz Grabiec
07582d6c10 sstables: index_reader: Fix abort when _trust_pi == trust_promoted_index::no
data is not moved-from if _trust_pi == trust_promoted_index::no, which
triggers the assert on data.empty(). We should make it empty
unconditionally.

Message-Id: <1545408731-14333-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit 419c771791)
2018-12-24 11:45:14 +02:00
Tomasz Grabiec
18c89edbf7 sstables: mc: reader: Use enum class instead of variant
variant is an overkill here.

Message-Id: <1545409014-16289-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit 07d153c769)
2018-12-24 11:45:09 +02:00
Duarte Nunes
5558fa8c44 service/storage_proxy: Protect against empty mutation when storing hint
mutation_holder::get_mutation_for() can return nullptr's, so protect
against those when storing a hint.

Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20181221194853.98775-2-duarte@scylladb.com>
(cherry picked from commit e6a8883228)
2018-12-23 12:27:27 +02:00
Duarte Nunes
f678eb52cd service/storage_proxy: Protect against empty mutation in mutation_holder
The per_destination_mutation holder can contain empty mutations,
so make sure release_mutation() skips over those.

Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20181221194853.98775-1-duarte@scylladb.com>
(cherry picked from commit 6c4a34f378)
2018-12-23 12:27:25 +02:00
Tomasz Grabiec
dfb23f4b38 sstables: mc: index_reader: Handle CK_SIZE split across buffers properly
we incorrectly falled-through to the next state instead of returning
to read more data.

This can manifest in a number of ways, an abort, or incorrect read.

Introduced in 917528c

Fixes #4011.

Message-Id: <1545402032-4114-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit d2f96a60f6)
2018-12-21 20:40:35 +02:00
Tomasz Grabiec
502ddf158a sstables: mc: reader: Avoid unnecessary index reads on fast forwarding
When the next pending fragments are after the start of the new range,
we know there is no need to skip.

Caught by perf_fast_forward --datasets large-part-ds3 \
                            --run-tests=large-partition-slicing

Refs #3984
Message-Id: <1545308006-16389-1-git-send-email-tgrabiec@scylladb.com>

(cherry picked from commit 7afe2bad51)
2018-12-21 20:40:35 +02:00
Paweł Dziepak
0ccb0a127a Merge "Optimize slicing sstable readers" from Tomasz
"
Contains several improvements for fast-forwarding and slicing readers. Mainly
for the MC format, but not only:

  - Exiting the parser early when going out of the fast-forwarding window [MC-format-only]
  - Avoiding reading of the head of the partition when slicing
  - Avoiding parsing rows which are going to be skipped [MC-format-only]
"

* 'sstable-mc-optimize-slicing-reads' of github.com:tgrabiec/scylla:
  sstables: mc: reader: Skip ignored rows before parsing them
  sstables: mc: reader: Call _cells.clear() when row ends rather than when it starts
  sstables: mc: mutation_fragment_filter: Take position_in_partition rather than a clustering_row
  sstables: mc: reader: Do not call consume_row_marker_and_tombstone() for static rows
  sstables: mc: parser: Allow the consumer to skip the whole row
  sstables: continuous_data_consumer: Introduce skip()
  sstables: continuous_data_consumer: Make position() meaningful inside state_processor::process_state()
  sstables: mc: parser: Allocate dynamic_bitset once per read instead of once per row
  sstables: reader: Do not read the head of the partition when index can be used
  sstables: mc: mutation_fragment_filter: Check the fast-forward window first
  sstables: mc: writer: Avoid calling unsigned_vint::serialized_size()

(cherry picked from commit e6d26a528f)
2018-12-21 20:40:35 +02:00
Avi Kivity
b94997be0d Merge " Extract MC sstable writer to a separate compilation unit" from Tomasz
"
The motivation is to keep code related to each format separate, to make it
easier to comprehend and reduce incremental compilation times.

Also reduces dependency on sstable writer code by removing writer bits from
sstales.hh.

The ka/la format writers are still left in sstables.cc, they could be also extracted.
"

* 'extract-sstable-writer-code' of github.com:tgrabiec/scylla:
  sstables: Make variadic write() not picked on substitution error
  sstables: Extract MC format writer to mc/writer.cc
  sstables: Extract maybe_add_summary_entry() out of components_writer
  sstables: Publish functions used by writers in writer.hh
  sstables: Move common write functions to writer.hh
  sstables: Extract sstable_writer_impl to a header
  sstables: Do not include writer.hh from sstables.hh
  sstables: mc: Extract bound_kind_m related stuff into mc/types.hh
  sstables: types: Extract sstable_enabled_features::all()
  sstables: Move components_writer to .cc
  tests: sstable_datafile_test: Avoid dependency on components_writer

(cherry picked from commit b023e8b45d)
2018-12-21 20:40:35 +02:00
Benny Halevy
d3a5b10cb8 sstables_stats: writer_impl: move common members to base class
To be used by sstable_writer for stats collection.

Note that this patch is factored out so it can be verified with no
other change in functionality.

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
(cherry picked from commit 6853c1677d)
2018-12-21 20:40:35 +02:00
Benny Halevy
48f3f899ac sstable: make write_crc, write_digest, and new_sstable_component_file private methods
Prepare for per-sstable sub directory.
Also, these functions get most of their parameters from the sst at hand so they might
as well be first class members.

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
(cherry picked from commit ad5f1e4fbb)
2018-12-21 20:40:35 +02:00
Paweł Dziepak
c4f745276c Merge "Optimize sstable writing of large partitions" from Tomasz
"
This series contains several optimizations of the MC format sstable writer, mainly:
  - Avoiding output_stream when serializing into memory (e.g. a row)
  - Faster serialization of primitive types when serializing into memory

I measured the improvement in throughput (frag/s) using perf_fast_forward for
datasets with a single large partition with many small rows:

  - 10% for a row with a single cell of 8 bytes
  - 10% for a row with a single cell of 100 bytes
  -  9% for a row with a single cell of 1000 bytes
  - 13% for a row with 6 cells of 100 bytes
"

* tag 'avoid-output-stream-in-sstable-writer-v2' of github.com:tgrabiec/scylla:
  bytes_ostream: Optimize writing of fixed-size types
  sstables: mc: Write temporary data to bytes_ostream rather than file_writer
  sstables: mc: Avoid double-serialization of a range tombstone marker
  sstables: file_writer: Generalize bytes& writer to accept bytes_view
  sstables: Templetize write() functions on the writer
  sstables: Turn m_format_write_helpers.cc into an impl header
  sstables: De-futurize file_writer
  bytes_ostream: Implement clear()
  bytes_ostream: Make initial chunk size configurable

(cherry picked from commit e3f53542c9)
2018-12-21 20:40:35 +02:00
Hagit Segev
392c7dee3c release: prepare for 3.0-rc3 2018-12-21 20:19:50 +02:00
Gleb Natapov
04e982f909 streaming: hold to sink while close() is running and call close on error as well
Currently if something throws while streaming in mutation sending loop
sink is not closed. Also when close() is running the code does not hold
onto sink object. close() is async, so sink should be kept alive until
it completes. The patch uses do_with() to hold onto sink while close is
running and run close() on error path too.

Fixes #4004.

Message-Id: <20181220155931.GL3075@scylladb.com>
(cherry picked from commit 393269d34b)
2018-12-20 19:50:47 +02:00
Amnon Heiman
97a8cc149e node_exporter_install: switch to node_exporter 0.17
The newer version of node_exporter comes with important bug fixes, that
is especially important for I3.metal is not supported with the older
version of node_exporter.

The dashboards can now support both the new and the old version of
node_exporter.

Fixes #3927

Signed-off-by: Amnon Heiman <amnon@scylladb.com>
Message-Id: <20181210085251.23312-1-amnon@scylladb.com>
(cherry picked from commit 09c2b8b48a)
2018-12-20 19:12:00 +02:00
Avi Kivity
dbe347811c Merge "materialized views: Apply backpressure from view replicas" from Duarte
"
As the amount of pending view updates increases we know that there’s a
mismatch between the rate at which the base receives writes and the
rate at which the view retires them. We react by applying backpressure
to decrease the rate of incoming base writes, allowing the slow view
replicas to catch up. We want to delay the client’s next writes to a
base replica and we use the base’s backlog of view updates to derive
this delay.

To validate this approach we tested a 3 node Scylla cluster on GCE,
using n1-standard-4 instances with NVMEs. A loader running on a
n1-standard-8 instance run cassandra-stress with 100 threads. With the
delay function d(x) set to 1s, we see no base write timeouts. With the
delay function as defined in the series, we see that backlogs stabilize
at some (arbitrary) point, as predicted, but this stabilization
co-exists with base write timeouts. However, the system overall behaves
better than the current version, with the 100 view update limit, and
also better than the version without such limit or any backpressure.

More work is necessary to further stabilize the system. Namely, we want
to keep delaying until we see the backlog is decreasing. This will
require us to add more delay beyond the stabilization point, which in
turn should minimize the base write timeouts, and will also minimize the
amount of memory the backlog takes at each base replica.

Design document:
    https://docs.google.com/document/d/1J6GeLBvN8_c3SbLVp8YsOXHcLc9nOLlRY7pC6MH3JWo

Fixes #2538
"

Reviewed-by: Nadav Har'El <nyh@scylladb.com>

* 'materialized-views/backpressure/v2' of https://github.com/duarten/scylla: (32 commits)
  service/storage_proxy: Release mutation as early as possible
  service/storage_proxy: Delay replica writes based on view update backlog
  service/storage_proxy: Get the backlog of a particular base replica
  service/storage_proxy: Add counters for delayed base writes
  main: Start and stop the view_update_backlog_broker
  service: Distribute a node's view update backlog
  service: Advertise view update backlog over gossip
  service/storage_proxy: Send view update backlog from replicas
  service/storage_proxy: Prepare to receive replica view update backlog
  service/storage_proxy: Expose local view update backlog
  tests/view_schema_test: Add simple test for db::view::node_update_backlog
  db/view: Introduce node_update_backlog class
  db/hints: Initialize current backlog
  database: Add counter for current view backlog
  database: Expose current memory view update backlog
  idl: Add db::view::update_backlog
  db/view: Add view_update_backlog
  database: Wait on view update semaphore for view building
  service/storage_proxy: Use near-infinite timeouts for view updates
  database: generate_and_propagate_view_updates no longer needs a timeout
  ...

(cherry picked from commit b66f59aa3d)
2018-12-20 19:11:56 +02:00
Avi Kivity
8f2d24bb8f config: remove "to be removed before release" notice mc sstable config
The "enable_sstables_mc_format" config item help text wants to remove itself
before release. Since scylla-3.0 did not get enough mc format mileage, we
decided to leave it in, so the notice should be removed.

Fixes #4003.
Message-Id: <20181219082554.23923-1-avi@scylladb.com>

(cherry picked from commit dd51c659f7)
2018-12-19 19:08:36 +02:00
Nadav Har'El
689e11c892 scylla.spec: add libatomic
In some cases which I've yet to understand, build fails without libatomic.
We need to add it to the mock build machine.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20181218154757.25236-1-nyh@scylladb.com>
2018-12-19 12:55:00 +00:00
Nadav Har'El
1766c793a8 build_rpm.sh: put temporary mock in build/, not /var/lib.
build_rpm.sh uses "mock" to build an entire Scylla build environment,
which easily spans more than 15 gigabytes. mock, by defaults, puts this
build directory in a subdirectory of /var/lib/mock. There is no reason
why temporary build products need to be in the root directory: Some machines
(like mine) don't have that much free space in the root directory making it
impossible to use this script on such machines. and it's too easy
to leave temporary files there without noticing.

With this patch, the mock directories are put in build/mock/ instead of
/var/lib/mock.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20181217195952.15154-1-nyh@scylladb.com>
2018-12-19 12:54:37 +02:00
Avi Kivity
0b09008cde Merge "Make sstable reader fail on unknown colum names in MC format" from Piotr
"
Before the reader was just ignoring such columns but this creates a risk of data loss.

Refs #2598
"

* 'haaawk/2598/v3' of github.com:scylladb/seastar-dev:
  sstables: Add test_sstable_reader_on_unknown_column
  sstables: Exception on sstable's column not present in schema
  sstables: store column name in column_translation::column_info
  sstables: Make test_dropped_column_handling test dropped columns

(cherry picked from commit b0cb69ec25)
2018-12-18 16:23:51 +00:00
Piotr Jastrzebski
713e60f690 sstables: Extract mp_row_cosumer_m::check_schema_mismatch
This method will contain common logic used in multiple places
and reduce code duplication.

Signed-off-by: Piotr Jastrzebski <piotr@scylladb.com>
Message-Id: <bbda2f4ea4f9325055f096dc549f63b1bb03d3b6.1543311990.git.piotr@scylladb.com>
(cherry picked from commit 4366302c4c)
2018-12-18 16:23:47 +00:00
Paweł Dziepak
7b6841f947 Merge "Check for schema mismatch after dropping dead cells" from Piotr
"
Previously we were checking for schema incompatibility between current schema and sstable
serialization header before reading any data. This isn't the best approach because
data in sstable may be already irrelevant due to column drop for example.

This patchset moves the check after actual data is read and verified that it has
a timestamp new enough to classify it as nonobsolete.

Fixes #3924
"

* 'haaawk/3924/v3' of github.com:scylladb/seastar-dev:
  sstables: Enable test_schema_change for MC format
  sstables3: Throw error on schema mismatch only for live cells
  sstables: Pass column_info to consume_*_column
  sstables: Add schema_mismatch to column_info
  sstables: Store column data type in column_info
  sstables: Remove code duplication in column_translation

(cherry picked from commit 62ea153629)
2018-12-18 15:27:53 +00:00
Tomasz Grabiec
f124b7026f Merge 'Add tests for schema changes' from Paweł
This series adds a generic test for schema changes that generates
various schema and data before and after an ALTER TABLE operation. It is
then used to check correctness of mutation::upgrade() and sstable
readers and lead to the discovery of #3924 and #3925.

Fixes #3925.

* https://github.com/pdziepak/scylla.git schema-change-test/v3.1
  schema_builder: make member function names less confusing
  converting_mutation_partition_applier: fix collection type changes
  converting_mutation_partition_applier: do not emit empty collections
  sstable: use format() instead of sprint()
  tests/random-utils: make functions and variables inline
  tests: add models for schemas and data
  tests: generate schema changes
  tests/mutation: add test for schema changes
  tests/sstable: add test for schema changes

(cherry picked from commit 564b328b2e)
2018-12-18 14:57:50 +00:00
Avi Kivity
28cca751d1 Merge "Don't binary compare compressed sstables in test_write_many_partitions_* tests" from Piotr
"
Compression is not deterministic so instead of binary comparing the sstable files we just read data back
and make sure everything that was written down is still present.

Tests: unit(release)
"

* 'haaawk/binary-compare-of-compressed-sstables/v3' of github.com:scylladb/seastar-dev:
  sstables: Remove compressed parameter from get_write_test_path
  sstables: Remove unused sstable test files
  sstables: Ensure compare_sstables isn't used for compressed files
  sstables: Don't binary compare compressed sstables
  sstables: Remove debug printout from test_write_many_partitions

(cherry picked from commit 1ff6b8fb96)
2018-12-18 14:53:52 +00:00
Duarte Nunes
21d08aa41e Merge 'Fix evictable shard reader related issues' from Botond
"
Recently some additional issues were discovered related to recent
changes to the way inactive readers are evicted and making shard readers
evictable.
One such issue is that the `querier_cache` is not prepared for the
querier to be immediately evicted by the reader concurrency semaphore,
when registered with it as an inactive read (#3987).
The other issue is that the multishard mutation query code was not
fully prepared for evicted shard readers being re-created, or failing
why being re-created (#3991).

This series fixes both of these issues and adds a unit test which covers
the second one. I am working on a unit test which would cover the second
issue, but it's proving to be a difficult one and I don't want to delay
the fixes for these issues any longer as they also affect 3.0.

Fixes: #3987
Fixes: #3991
"

* 'evictable-reader-related-issues/branch-3.0/v1' of https://github.com/denesb/scylla:
  multishard_mutation_query: reset failed readers to inexistent state
  multishard_mutation_query: handle missing readers when dismantling
  multishard_mutation_query: add support for keeping stats for discarded partitions
  multishard_mutation_query: expect evicted reader state when creating reader
  multishard_mutation_query: pretty-print the reader state in log messages
  querier_cache: check that the query wasn't evicted during registering
  reader_concurrency_semaphore: use the correct types in the constructor
  reader_concurrency_semaphore: add consume_resources()
  reader_concurrency_semaphore::inactive_read_handle: add operator bool()
2018-12-18 13:36:42 +00:00
Botond Dénes
f0b5170fa6 multishard_mutation_query: reset failed readers to inexistent state
When attempting to dismantling readers, some of the to-be-dismantled
readers might be in a failed state. The code waiting on the reader to
stop is expecting failures, however it didn't do anything besides
logging the failure and bumping a counter. Code in the lower layers did
not know how to deal with a failed reader and would trigger
`std::bad_variant_access` when trying to process (save or cleanup) it.
To prevent this, reset the state of failed readers to `inexistent_state`
so code in the lower layers doesn't attempt to further process them.

(cherry picked from commit b4c3aab4a7)
2018-12-18 14:46:56 +02:00
Botond Dénes
3b617e873c multishard_mutation_query: handle missing readers when dismantling
When dismantling the combined buffer and the compaction state we are no
longer guaranteed to have the reader each partition originated from. The
reader might have been evicted and not resumed, or resuming it might
have failed. In any case we can no longer assume the originating reader
of each partition will be present. If a reader no longer exists,
discard the partitions that it emitted.

(cherry picked from commit 9cef043841)
2018-12-18 14:46:56 +02:00
Botond Dénes
4eb9836e64 multishard_mutation_query: add support for keeping stats for discarded partitions
In the next patches we will add code that will have to discard some of
the dismantled partitions/fragments/bytes. Prepare the
`dismantle_buffer_stats` struct for being able to track the discarded
partitions/fragments/bytes in addition to those that were successfully
dismantled.

(cherry picked from commit 438bef333b)
2018-12-18 14:46:56 +02:00
Botond Dénes
46af353209 multishard_mutation_query: expect evicted reader state when creating reader
Previously readers were created once, so `make_remote_reader()` had a
validation to ensure readers were not attempted at being created more
than once. This validation was done by checking that the reader-state is
either `inexistent` or `successful_lookup`. However with the
introduction of pausing shard readers, it is now possible that a reader
will have to be created and then re-created several times, however this
validation was not updated to expect this.
Update the validation so it also expects the reader-state to be
`evicted`, the state the reader will be if it was evicted while paused.

(cherry picked from commit ce52436af4)
2018-12-18 14:46:54 +02:00
Botond Dénes
76f70c676e multishard_mutation_query: pretty-print the reader state in log messages
(cherry picked from commit 1effb1995b)
2018-12-18 14:34:33 +02:00
Botond Dénes
afc9f0e177 querier_cache: check that the query wasn't evicted during registering
The reader concurrency semaphore can evict the querier when it is
registered as an inactive read. Make the `querier_cache` aware of this
so that it doesn't continue to process the inserted querier when this
happens.
Also add a unit test for this.

(cherry picked from commit 5780f2ce7a)
2018-12-18 14:34:33 +02:00
Botond Dénes
c899191ad5 reader_concurrency_semaphore: use the correct types in the constructor
Previously there was a type mismatch for `count` and `memory`, between
the actual type used to store them in the class (signed) and the type
of the parameters in the constructor (unsigned).
Although negative numbers are completely valid for these members,
initializing them to negative numbers don't make sense, this is why they
used unsigned types in the constructor. This restriction can backfire
however when someone intends to give these parameters the maximum
possible value, which, when interpreted as a signed value will be `-1`.
What's worse the caller might not even be aware of this unsigned->signed
conversion and be very suprised when they find out.
So to prevent surprises, expose the real type of these members, trusting
the clients of knowing what they are doing.

Also add a `no_limits` constructor, so clients don't have to make sure
they don't overflow internal types.

(cherry picked from commit e1d8237e6b)
2018-12-18 14:34:33 +02:00
Botond Dénes
a3563e5f7d reader_concurrency_semaphore: add consume_resources()
(cherry picked from commit dfd649a6b4)
2018-12-18 14:34:33 +02:00
Botond Dénes
78c5b09694 reader_concurrency_semaphore::inactive_read_handle: add operator bool()
(cherry picked from commit 21b44adbfe)
2018-12-18 14:34:33 +02:00
Amnon Heiman
a51878205a node-exporter.service: Update command line to fix service startup
The upgrade to node_exporter 0.17 commit
09c2b8b48a ("node_exporter_install: switch
to node_exporter 0.17") caused the service to no longer start. Turns out
node_exported broke backwards compatibility of the command line between
0.15 to 0.16. Fix it up.

While fixing the command line, all the collector that are enabled by
default were removed.

Fixes #3989

Signed-off-by: Amnon Heiman <amnon@scylladb.com>
[ penberg@scylladb.com: edit commit message ]
Message-Id: <20181213114831.27216-1-amnon@scylladb.com>
(cherry picked from commit 571755e117)
2018-12-18 08:33:04 +02:00
Avi Kivity
46efc08882 Update seastar submodule
* seastar 6700dc3...08f1258 (1):
  > reactor: disable nowait aio due to a kernel bug

Fixes #3996.
2018-12-17 17:00:14 +02:00
Avi Kivity
c95433c967 Update seastar submodule
* seastar 1651a2a...6700dc3 (3):
  > build: link against libatomic
  > core/semaphore: Allow combining semaphore_units()
  > core/shared_ptr: Allow releasing a lw_shared_ptr to a non-const object

Fixes #3996.
2018-12-17 15:53:01 +02:00
Piotr Sarna
df3b6fb4a8 cql3: refuse to create index on COMPACT STORAGE with ck
To follow C* compatibility, creating an index on COMPACT STORAGE
table should be disallowed not only on base primary keys,
but also when the base table contains clustering keys.
Message-Id: <ab40c39730aff2e164d11ee5159ff62b8ec9e8e8.1544698186.git.sarna@scylladb.com>

(cherry picked from commit 6743af5dbd)
2018-12-17 09:45:43 +02:00
Piotr Sarna
44ee43bb17 cql3: add refusing to create an index on static column
Secondary indexes on static columns are not yet supported,
so creating such index should return an appropriate error.

Fixes #3993
Message-Id: <700b0a71e80da52d2d5250edacc12626b55681fa.1544785127.git.sarna@scylladb.com>

(cherry picked from commit 63bd43e57e)
2018-12-17 09:44:52 +02:00
Asias He
aac363ca86 storage_service: Notify NEW_NODE only when a node is new node
This is a backport of CASSANDRA-11038.

Before this, a restarted node will be reported as new node with NEW_NODE
cql notification.

To fix, only send NEW_NODE notification when the node was not part of
the cluster

Fixes: #3979
Tests: pushed_notifications_test.py:TestPushedNotifications.restart_node_test
Message-Id: <453d750b98b5af510c4637db25b629f07dd90140.1544583244.git.asias@scylladb.com>
(cherry picked from commit 71c1681f6c)
2018-12-16 13:59:19 +02:00
Duarte Nunes
9e6cc5b024 service/storage_proxy: Embed the expire timer in the response handler
Embedding the expire timer for a write response in the
abstract_write_response_handler simplifies the code as it allows
removing the rh_entry type.

It will also make the timeout easily accessible inside the handler,
for future patches.

Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20181213111818.39983-1-duarte@scylladb.com>
(cherry picked from commit f8878238ed)
2018-12-13 13:24:09 +00:00
Duarte Nunes
13b72c7b92 Merge branch 'gossip: Send node UP event to cql client after cql server is up' from Asias
"
This is a backport of CASSANDRA-8236.

Before this patch, scylla sends the node UP event to cql client when it
sees a new node joins the cluster, i.e., when a new node's status
becomes NORMAL. The problem is, at this time, the cql server might not
be ready yet. Once the client receives the UP event, it tries to
connect to the new node's cql port and fails.

To fix, a new application_sate::RPC_READY is introduced, new node sets
RPC_READY to false when it starts gossip in the very beginning and sets
RPC_READY to true when the cql server is ready.

The RPC_READY is a bad name but I think it is better to follow Cassandra.

Nodes with or without this patch are supposed to work together with no
problem.

Refs #3843
"

* 'asias/node_up_down.upstream.v4.1' of github.com:scylladb/seastar-dev:
  storage_service: Use cql_ready facility
  storage_service: Handle application_state::RPC_READY
  storage_service: Add notify_cql_change
  storage_service: Add debug log in notify_joined
  storage_service: Add extra check in notify_joined
  storage_service: Add notify_joined
  storage_service: Add debug log in notify_up
  storage_service: Add extra check in notify_up
  storage_service: Add notify_up
  storage_service: Make notify_left log debug level
  storage_service: Introduce notify_left
  storage_service: Add debug log in notify_down
  storage_service: Introduce notify_down
  storage_service: Add set_cql_ready
  gossip: Add gossiper::is_cql_ready
  gms: Add endpoint_state::is_cql_ready
  gms: Add application_state::RPC_READY
  gms: Introduce cql_ready in versioned_value

(cherry picked from commit a42b2895c2)
2018-12-13 12:06:59 +00:00
Avi Kivity
6b011fbe0a build: pass C compiler configuration in dist package build
Just like we allow customizing the C++ compiler, we should allow customizing
the C compiler.

Ref #3978
Message-Id: <20181211172821.30830-1-avi@scylladb.com>

(cherry picked from commit fa96e07e6b)
2018-12-12 14:41:38 +02:00
Tomasz Grabiec
9dd4e1b01f sstables: index_reader: Avoid schema copy in advance_to()
Introduced in 7e15e43.

Exposed by perf_fast_forward:

  running: large-partition-skips on dataset large-part-ds1
  Testing scanning large partition with skips.
  Reads whole range interleaving reads with skips according to read-skip pattern:
  read    skip      time (s)     frags     frag/s (...)
  1       0         5.268780   8000000    1518378

  1       1        31.695985   4000000     126199
Message-Id: <1544614272-21970-1-git-send-email-tgrabiec@scylladb.com>

(cherry picked from commit 0a853b8866)
2018-12-12 14:38:49 +02:00
Nadav Har'El
e91c741ef5 secondary indexes: fail attempts to create a CUSTOM INDEX
Cassandra supports a "CREATE CUSTOM INDEX" to create a secondary index
with a custom implementation. The only custom implementation that Cassandra
supports is SASI. But Scylla doesn't support this, or any other custom
index implementation. If a CREATE CUSTOM INDEX statement is used, we
shouldn't silently ignore the "CUSTOM" tag, we should generate an error.

This patch also includes a regression test that "CREATE CUSTOM INDEX"
statements with valid syntax fail (before this patch, they succeeded).

Fixes #3977

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20181211224545.18349-2-nyh@scylladb.com>
(cherry picked from commit a0379209e6)
2018-12-12 00:32:35 +00:00
Nadav Har'El
b18e9e115d Fix typo in error message
Interestingly, this typo was copied from the original Cassandra source
code :-)

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20181211224545.18349-1-nyh@scylladb.com>
(cherry picked from commit 36db4fba23)
2018-12-12 00:32:35 +00:00
Avi Kivity
0b86ab0d2a build: build libdeflate with user selected C compiler
If the user specified a C compiler, use it to build libdeflate.

Fixes #3978.
Message-Id: <20181211145604.14847-1-avi@scylladb.com>

(cherry picked from commit 34a31a807d)
2018-12-11 19:24:24 +02:00
Duarte Nunes
97cd9108d6 db/system_distributed_keyspace: Create the schema with min_timestamp
Different nodes can concurrently create the distributed system
keyspace on boot, before the "if not exists" clause can take effect.

However, the resulting schema mutations will be different since
different nodes use different timestamps. This patch forces the
timestamps to be the same across all nodes, so we save some schema
mismatches.

This fixes a bug exposed by ca5dfdf, whereby the initialization of the
distributed system keyspace is done before waiting for schema
agreement. While waiting for schema agreement in
storage_service::join_token_ring(), the node still hasn't joined the
ring and schemas can't be pulled from it, so nodes can deadlock. A
similar situation can happen between a seed node and a non-seed node,
where the seed node progresses to a different "wait for schema
agreement" barrier, but still can't make progress because it can't
pull the schema from the non-seed node still trying to join the ring.

Finally, it is assumed that changes to the schema of the current
distributed system keyspace tables will be protected by a cluster
feature and a subsequent schema synchronization, such that all nodes
will be at a point where schemas can be transferred around.

Fixes #3976

Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20181211113407.20075-1-duarte@scylladb.com>
(cherry picked from commit 89ae3fbf11)
2018-12-11 14:53:30 +00:00
Hagit Segev
f81fe96b0b release: prepare for 3.0-rc2 2018-12-11 12:32:34 +02:00
Avi Kivity
91ce3a7957 sstables: fix overflow in clustering key blocks header bit access
_ck_blocks_header is a 64-bit variable, so the mask should be 64 bits too.
Otherwise, a shift in the range 32-63 will produce wrong results.

Fix by using a 64-bit mask.

Found by Fedora 29's ubsan.

Fixes #3973.
Message-Id: <20181209120549.21371-1-avi@scylladb.com>

(cherry picked from commit 7c7da0b462)
2018-12-10 14:10:27 +02:00
Takuya ASADA
af7e58f4c5 dist/offline_installer/redhat: fix missing dependencies
Offline installer with Scylla 3.0 causes dependency error on CentOS, added
missing packages.

Fixes #3969

Signed-off-by: Takuya ASADA <syuu@scylladb.com>
Message-Id: <20181207020711.23055-1-syuu@scylladb.com>
(cherry picked from commit a2d0ebf4d9)
2018-12-10 14:10:15 +02:00
Amos Kong
bd3373b511 scylla_setup: only ask for nic in interactive mode
Current scylla_setup still asks for nic even nic is already assigned in cmdline.

Fixes #3908

Signed-off-by: Amos Kong <amos@scylladb.com>
Message-Id: <6b867e17a5583c495c771a37d5fa1e8366b1d61b.1542337635.git.amos@scylladb.com>
(cherry picked from commit 09a3b11c2f)
2018-12-09 19:26:34 +02:00
Gleb Natapov
4820130abe storage_proxy: fix crash during write timeout callback invocation
rh_entry address is captured inside timeout's callback lambda, so the
structure should not be moved after it is created. Change the code to
create rh_entry in-place instead of moving it into the map.

Fixes #3972.

Message-Id: <20181206164043.GN25283@scylladb.com>
(cherry picked from commit 9fb79bf379)
2018-12-09 15:25:52 +02:00
Tomasz Grabiec
9b299241e5 Merge "Fixes for collecting stats in SST3 + more tests" from Vladimir
This patchset fixes several remaining issues found during thorough
testing of SSTables 3.x statistics and enriches ~30 unit tests with
statistics validation against Cassandra-generated golden copies.

* https://github.com/argenet/scylla/tree/projects/sstables-30/sst3-tests-statistics/v1:
  sstables: Enforce estimated_partitions in generate_summary() to be
    always positive.
  sstables: Don't enforce default max_local_deletion_time value for 'mc'
    files.
  sstables: Update TTL/local deletion stats for non-expiring and live
    liveness_info.
  sstables: Collect statistics when writing RT markers to SSTables 3.x.
  tests: Return sstable_assertions from validate_read() helper.
  tests: Introduce helper for validating stats metadata in SSTables 3.x
    tests.
  tests: Add stats metadata validation to test_write_static_row.
  tests: Add stats metadata validation to
    test_write_composite_partition_key.
  tests: Add stats metadata validation to
    test_write_composite_clustering_key.
  tests: Add stats metadata validation to test_write_wide_partitions.
  tests: Add stats metadata validation to write_ttled_row
  tests: Add stats metadata validation to write_ttled_column
  tests: Add stats metadata validation to write_deleted_column
  tests: Add stats metadata validation to write_deleted_row
  tests: Add stats metadata validation to write_collection_wide_update
  tests: Add stats metadata validation to
    write_collection_incremental_update
  tests: Add stats metadata validation to write_multiple_partitions
  tests: Add stats metadata validation to write_multiple_rows
  tests: Add stats metadata validation to
    write_missing_columns_large_set
  tests: Add stats metadata validation to write_different_types
  tests: Add stats metadata validation to write_empty_clustering_values
  tests: Add stats metadata validation to write_large_clustering_key
  tests: Add stats metadata validation to write_compact_table
  tests: Add stats metadata validation to write_user_defined_type_table
  tests: Add stats metadata validation to write_simple_range_tombstone
  tests: Add stats metadata validation to
    write_adjacent_range_tombstones
  tests: Add stats metadata validation to
    write_non_adjacent_range_tombstones
  tests: Add stats metadata validation to
    write_mixed_rows_and_range_tombstones
  tests: Add stats metadata validation to
    write_adjacent_range_tombstones_with_rows
  tests: Add stats metadata validation to
    write_range_tombstone_same_start_with_row
  tests: Add stats metadata validation to
    write_range_tombstone_same_end_with_row
  tests: Add stats metadata validation to
    write_two_non_adjacent_range_tombstones
  tests: Delete unused (bogus) Statistics.db file from write_ SST3
    tests.

(cherry picked from commit bb24d378b2)
2018-12-08 14:08:46 +02:00
Avi Kivity
745a98e151 Merge "Fix deadlocking multishard readers" from Botond
"
Multishard combining readers, running concurrently, with limited
concurrency and no timeout may deadlock, due to inactive shard readers
sitting on permits. To avoid this we have to make sure that all shard
readers belonging to a multishard combining readers, that are not
currently active, can be evicted to free up their permits, ensuring that
all readers can make progress.
Making inactive shard readers evictable is the solution for this
problem, however the original series introducing this solution
(414b14a6bd) did not go all they way and
left some loose ends. These loose ends are tied up by this mini-series.
Namely, two issues remained:
* The last reader to reach EOS was not paused (made evictable).
* Readers created/resumed as part of a read-ahead were not paused
  immediately after finishing the read-ahead.

This series fixes both of these.

Fixes: #3865
Tests: unit(release, debug)
"

* 'fix-multishard-reader-deadlock/v1' of https://github.com/denesb/scylla:
  multishard_combining_reader: pause readers after reading ahead
  multishard_combining_reader: pause *all* EOS'd readers

(cherry picked from commit 21b4b2b9a1)
2018-12-08 14:08:46 +02:00
Avi Kivity
b9c99af18b Merge "Fix tombstone histogram when writing SSTables 3.x" from Vladimir
"
This patchset extends a number of existing tests to check SSTables
statistics for 'mc' format and fixes an issue discovered with the help
of one of the tests.

Tests: unit {release}
"

* 'projects/sstables-30/check-stats/v2' of https://github.com/argenet/scylla:
  tests: Run sstable_timestamp_metadata_correcness_with_negative with all SSTables versions.
  tests: Run sstable_tombstone_histogram_test for all SSTables versions.
  tests: Run min_max_clustering_key_test on all SSTables versions.
  tests: Expand test_sstable_max_local_deletion_time_2 to run for all SSTables versions.
  tests: Run test_sstable_max_local_deletion_time on all SSTables versions.
  tests: Extend test checking tombstones histogram to cover all SSTables versions.
  sstables: Properly track row-level tombstones when writing SSTables 3.x.
  tests: Run min_max_clustering_key_test_2 for all SSTables versions.
  tests: Make reusable_sst() helper accept SSTables version parameter.

(cherry picked from commit f073ea5f87)
2018-12-08 14:08:44 +02:00
Asias He
cded9c7ac7 gossip: Fix race in real_mark_alive and shutdown msg
In dtest, we have

   self.check_rows_on_node(node1, 2000)
   self.check_rows_on_node(node2, 2000)

which introduce the following cluster operations:

1) Initially:

- node1 up
- node2 up

2) self.check_rows_on_node(node1, 2000)
- node2 down
- node2 up (A: node2 will call gossiper::real_mark_alive when node2 boots
up to mark node1 up)

3) self.check_rows_on_node(node2, 2000)
- node1 down (B: node1 will send shutdown gossip message to node2, node2
will mark node1 down)
- node1 up (C: when node1 is up, node2 will call
gossiper::real_mark_alive)

Since there is no guarantee the order of Operation A and Operation B, it
is possible node2 will mark node1 as status=shutdown and mark node1 is
UP.

In Operation C, node2 will call gossiper::real_mark_alive to mark node1
up, but since node2 might think node1 is already up, node2 will exit
early in gossiper::real_mark_alive and not log "InetAddress 127.0.0.1 is
now UP, status={}"

As a result, dtest fails to see node2 reports node1 is up when it boots
node1 and fail the test.

   TimeoutError: 23 Nov 2018 10:44:19 [node2] Missing: ['127.0.0.1.* now UP']

In the log we can see node1 marked as DOWN and UP almost at the same time on node2:

   INFO  2018-11-23 22:31:29,999 [shard 0] gossip - InetAddress 127.0.0.1 is now DOWN, status = shutdown
   INFO  2018-11-23 22:31:30,006 [shard 0] gossip - InetAddress 127.0.0.1 is now UP, status = shutdown

Fixes #3940

Tests: dtest with 20 consecutive succesful runs
Message-Id: <996dc325cbcc3f94fc0b7569217aa65464eaaa1c.1543213511.git.asias@scylladb.com>
(cherry picked from commit eeeb2da7bb)
2018-12-08 13:42:43 +02:00
Gleb Natapov
4acfc5ed8f hints: make hints manager more resilient to unexpected directory content
Currently if hints directory contains unexpected directories Scylla fails to
start with unhandled std::invalid_argument exception. Make the manager
ignore malformed files instead and try to proceed anyway.
Message-Id: <20181121134618.29936-2-gleb@scylladb.com>

(cherry picked from commit b4a8802edc)
2018-12-08 13:42:43 +02:00
Gleb Natapov
cb9199bc7f hints: add auxiliary function for scanning high level hints directory
We scan hints directory in two places: to search for files to replay and
to search for directories to remove after resharding. The code that
translates directory name to a shard is duplicated. It is simple now, so
not a bit issue but in case it grows better have it in one place.
Message-Id: <20181121134618.29936-1-gleb@scylladb.com>

(cherry picked from commit 9433d02624)
2018-12-08 13:42:43 +02:00
Tomasz Grabiec
695ff5383f Merge "Correct the usage of row ttl and add write-read test" from Piotr
Fixes the condition which determines whether a row ttl should be used for a cell
and adds a test that uses each generated mutation to populate mutation source
and then verifies that it can read back the same mutation.

* seastar-dev.git haaawk/sst3/write-read-test/v3:
  Fix use_row_ttl condition
  Add test_all_data_is_read_back

(cherry picked from commit b8c405c019)
2018-12-08 13:42:43 +02:00
Tomasz Grabiec
730e48bf60 configure.py: Always add a rule for building gen_crc_combine_table
Fixes a build failure when only the scylla binary was selected for
building like this:

  ./configure.py --with scylla

In this case the rule for gen_crc_combine_table was missing, but it is
needed to build crc_combine_table.o

Message-Id: <1544010138-21282-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit edbef7400b)
2018-12-08 13:42:43 +02:00
Tomasz Grabiec
af6d4f40e1 utils/gz: Fix compilation on non-x86 archs
gen_crc_combine_table is now executed on every build, so it should not
fail on unsupported archs. The generated file will not contain data,
but this is fine since it should not be used.

Another problem is that u32 and u64 aliases were not visible in the #else
branch in crc_combine.cc
Message-Id: <1543864425-5650-1-git-send-email-tgrabiec@scylladb.com>

(cherry picked from commit 9a4c00beb7)
2018-12-08 13:42:43 +02:00
Avi Kivity
9d8507de09 Merge "Optimize checksum_combine() for CRC32" from Tomek
"
zlib's crc32_combine() is not very efficient. It is faster to re-combine
the buffer using crc32(). It's still substantial amount of work which
could be avoided.

This patch introduces a fast implementation of crc32_combine() which
uses a different algorithm than zlib. It also utilizes intrinsics for
carry-less multiplication instruction to perform the computation faster.
The details of the algorithm can be found in code comments.

Performance results using perf_checksum and second buffer of length 64 KiB:

zlib CRC32 combine:   38'851   ns
libdeflate CRC32:      4'797   ns
fast_crc32_combine():     11   ns

So the new implementation is 3500x faster than zlib's, and 417x faster than
re-checksumming the buffer using libdeflate.

Tested on i7-5960X CPU @ 3.00GHz

Performance was also evaluated using sstable writer benchmark:

  perf_fast_forward --populate --sstable-format=mc --data-directory /tmp/perf-mc \
     --value-size=10000 --rows 1000000 --datasets small-part

It yielded 9% improvement in median frag/s (129'055 vs 117'977).

Refs #3874
"

* tag 'fast-crc32-combine-v2' of github.com:tgrabiec/scylla:
  tests: perf_checksum: Test fast_crc32_combine()
  tests: Rename libdeflate_test to checksum_utils_test
  tests: libdeflate: Add more tests for checksum_combine()
  tests: libdeflate: Check both libdeflate and default checksummers
  sstables: Use fast_crc_combine() in the default checksummer
  utils/gz: Add fast implementation of crc32_combine()
  utils/gz: Add pre-computed polynomials
  utils/gz: Import Barett reduction implementation from libdeflate
  utils: Extract clmul() from crc.hh

(cherry picked from commit b098b5b987)
2018-12-08 13:42:43 +02:00
Tomasz Grabiec
07c980845d utils/crc: Add clmul_u32() implementation
Needed for backporting dependent changes.

Extracted from:

      commit 79136e895f
      Author: Yibo Cai (Arm Technology China) <Yibo.Cai@arm.com>
      Date:   Thu Nov 1 03:26:16 2018 +0000

        utils/crc: calculate crc in parallel
2018-12-08 13:42:43 +02:00
Tomasz Grabiec
c52b8239d0 configure.py: Compile against Westmere on x86
Needed for backporting dependent changes.

Extracted from:

  commit 79136e895f
  Author: Yibo Cai (Arm Technology China) <Yibo.Cai@arm.com>
  Date:   Thu Nov 1 03:26:16 2018 +0000

    utils/crc: calculate crc in parallel
2018-12-08 13:42:43 +02:00
Tomasz Grabiec
5a07a4fac8 configure.py: Use armv8-a+crc+crypto ISA on aarch64
Needed for backporting dependent changes.

Extracted from:

  commit 1c48e3fbec
  Author: Yibo Cai (Arm Technology China) <Yibo.Cai@arm.com>
  Date:   Mon Oct 29 02:58:19 2018 +0000

    utils/crc: leverage arm64 crc extension
2018-12-08 13:42:43 +02:00
Avi Kivity
b9c046b17b Merge "Optimize checksum computation for the MC sstable format" from Tomek
"
One part of the improvement comes from replacing zlib's CRC32 with the one
from libdeflate, which is optimized for modern architecture and utilizes the
PCLMUL instruction.

perf_checksum test was introduced to measure performance of various
checksumming operations.

Results for 514 B (relevant for writing with compression enabled):

    test                                      iterations      median         mad         min         max
    crc_test.perf_deflate_crc32_combine            58414    16.711us     3.483ns    16.708us    16.725us
    crc_test.perf_adler_combine                165788278     6.059ns     0.031ns     6.027ns     7.519ns
    crc_test.perf_zlib_crc32_combine               59546    16.767us    26.191ns    16.741us    16.801us
    ---
    crc_test.perf_deflate_crc32_checksum        12705072    83.267ns     4.580ns    78.687ns    98.964ns
    crc_test.perf_adler_checksum                 3918014   206.701ns    23.469ns   183.231ns   258.859ns
    crc_test.perf_zlib_crc32_checksum            2329682   428.787ns     0.085ns   428.702ns   510.085ns

Results for 64 KB (relevant for writing with compression disabled):

    test                                      iterations      median         mad         min         max
    crc_test.perf_deflate_crc32_combine            25364    38.393us    17.683ns    38.375us    38.545us
    crc_test.perf_adler_combine                169797143     5.842ns     0.009ns     5.833ns     6.901ns
    crc_test.perf_zlib_crc32_combine               26067    38.663us    95.094ns    38.546us    40.523us
    ---
    crc_test.perf_deflate_crc32_checksum          202821     4.937us    14.426ns     4.912us     5.093us
    crc_test.perf_adler_checksum                   44684    22.733us   206.263ns    22.492us    25.258us
    crc_test.perf_zlib_crc32_checksum              18839    53.049us    36.117ns    53.013us    53.274us

The new CRC32 implementation (deflate_crc32) doesn't provide a fast
checksum_combine() yet, it delegates to zlib so it's as slow as the latter.

Because for CRC32 checksum_combine() is several orders of magnitude slower
than checksum(), we avoid calling checksum_combine() completely for this
checksummer. We still do it for adler32, which has combine() which is faster
than checksum().

SStable write performance was evaluated by running:

  perf_fast_forward --populate --data-directory /tmp/perf-mc \
     --rows=10000000 -c1 -m4G --datasets small-part

Below is a summary of the average frag/s for a memtable flush. Each result is
an average of about 20 flushes with stddev of about 4k.

Before:

 [1] MC,lz4: 330'903
 [2] LA,lz4: 450'157
 [3] MC,checksum: 419'716
 [4] LA,checksum: 459'559

After:

 [1'] MC,lz4: 446'917 ([1] + 35%)
 [2'] LA,lz4: 456'046 ([2] + 1.3%)
 [3'] MC,checksum: 462'894 ([3] + 10%)
 [4'] LA,checksum: 467'508 ([4] + 1.7%)

After this series, the performance of the MC format writer is similar to that
of the LA format before the series.

There seems to be a small but consistent improvement for LA too. I'm not sure
why.
"

* tag 'improve-mc-sstable-checksum-libdeflate-v3' of github.com:tgrabiec/scylla:
  tests: perf: Introduce perf_checksum
  tests: Add test for libdeflate CRC32 implementation
  sstables: compress: Use libdeflate for crc32
  sstables: compress: Rename crc32_utils to zlib_crc32_checksummer
  licenses: Add libdeflate license
  Integrate libdeflate with the build system
  Add libdeflate submodule
  sstables: Avoid checksum_combine() for the crc32 checksummer
  sstables: compress: Avoid unnecessary checksum_combine()
  sstables: checksum_utils: Add missing include

(cherry picked from commit 5e759b0c07)
2018-12-08 13:42:43 +02:00
Avi Kivity
979cb636b8 Update seastar submodule
* seastar e64281d...1651a2a (1):
  > tests: perf: Make do_not_optimize() take the argument by const&
2018-12-08 13:42:43 +02:00
Botond Dénes
59cf9d9070 querier: fix evict_one() and evict_all_for_table()
Both of these have the same problem. They remove the to-be-evicted
entries from `_entries` but they don't unregister the `entry` from the
`read_concurrency_semaphore`. This results in the
`reader_concurrency_semaphore` being left with a dangling pointer to the
entries will trigger segfault when it tries to evict the associated
inactive reads.

Also add a unit test for `evict_all_for_table()` to check that it works
properly (`evict_one()` is only used in tests, so no dedicated test for
it).

Fixes: #3962

Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <57001857e3791c6385721b624d33b667ccda2e7d.1544010868.git.bdenes@scylladb.com>
(cherry picked from commit 77dbc7d09a)
2018-12-06 11:38:44 +02:00
Duarte Nunes
c9ec9d4087 Merge seastar upstream
* seastar 880826e...e64281d (2):
  > core/semaphore: Change the access of semaphore_units main ctor
  > Merge "Add semaphore_units<>::split() function" from Duarte

Signed-off-by: Duarte Nunes <duarte@scylladb.com>
2018-12-05 20:25:17 +00:00
Gleb Natapov
2e8fefbc5a storage_proxy: store hint for CL=ANY if all nodes replied with failure
Current code assumes that request failed if all replicas replied with
failure, but this is not true for CL=ANY requests. Take it into account.

Fixed: #3565
(cherry picked from commit 17197fb005)
2018-12-05 20:14:58 +00:00
Gleb Natapov
6be0635029 storage_proxy: complete write request early if all replicas replied with success of failure
Currently if write request reaches CL and all replicas replied, but some
replied with failures, the request will wait for timeout to be retired.
Detect this case and retire request immediately instead.

Fixes #3566

(cherry picked from commit d1d04eae3c)
2018-12-05 20:14:57 +00:00
Gleb Natapov
04a544c0a2 storage_proxy: check that write failure response comes from recognized replica
Before accounting failure response we need to make sure it comes from a
replica that participates in the request.

(cherry picked from commit 76ab3d716b)
2018-12-05 20:14:57 +00:00
Gleb Natapov
028f9b95d1 storage_proxy: move code executed on write timeout into separate function
Currently the callback is in lambda, but we will want to call the code
not only during timer expiration.

(cherry picked from commit 7bc68aa0eb)
2018-12-05 20:14:57 +00:00
Avi Kivity
54258ca8eb Merge "db/hints: Use frozen_mutation in hinted handoff" from Duarte
"
This series changes hinted handoff to work with `frozen_mutation`s
instead of naked `mutation`s. Instead of unfreezing a mutation from
the commitlog entry and then freezing it again for sending, now we'll
just keep the read, frozen mutation.

Tests: unit(release)
"

* 'hh-manager-cleanup/v1' of https://github.com/duarten/scylla:
  db/hints/manager: Use frozen_mutation instead of mutation
  db/hints/manager: Use database::find_schema()
  db/commitlog/commitlog_entry: Allow moving the contained mutation
  service/storage_proxy: send_to_endpoint overload accepting frozen_mutation
  service/storage_proxy: Build a shared_mutation from a frozen_mutation
  service/storage_proxy: Lift frozen_mutation_and_schema
  service/storage_proxy: Allow non-const ranges in mutate_prepare()

(cherry picked from commit 1891779e64)
2018-12-05 20:14:57 +00:00
Gleb Natapov
c9a030f1f0 storage_proxy: count number of timed out write attempts after CL is reached
It is useful to have this counter to investigate the reason for read
repairs. Non zero value means that writes were lost after CL is reached
and RR is expected.

Message-Id: <20181009120900.GF22665@scylladb.com>
(cherry picked from commit 207b57a892)
2018-12-05 20:14:57 +00:00
Gleb Natapov
1c7daef554 storage_proxy: do not pass write_stats down to send_to_live_endpoints
write_stats is referenced from write handler which is available in
send_to_live_endpoints already. No need to pass it down.

Message-Id: <20181009133017.GA14449@scylladb.com>
(cherry picked from commit 319ece8180)
2018-12-05 20:14:57 +00:00
Duarte Nunes
f8195a77b0 db/view/view_builder: Don't timeout waiting for view to be built
Remove the timeout argument to
db::view::view_builder::wait_until_built(), a test-only function to
wait until a given materialized view has finished building.

This change is motivated by the fact that some tests running on slow
environments will timeout. Instead of incrementally increasing the
timeout, remove it completely since tests are already run under an
exterior timeout.

Fixes #3920

Tests: unit release(view_build_test, view_schema_test)

Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20181115173902.19048-1-duarte@scylladb.com>
(cherry picked from commit 6fbf792777)
2018-12-05 19:20:36 +00:00
Duarte Nunes
5b724c80ab db/view: Don't copy keyspace name
Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20181022104527.14555-1-duarte@scylladb.com>
(cherry picked from commit f3a5ec0fd9)
2018-12-05 19:19:26 +00:00
Nadav Har'El
4a7ae81b3f materialized views: update stats.write statistics in all cases
mutate_MV usually calls send_to_endpoint() to push view update to remote
view replicas. This function gets passed a statistics object,
service::storage_proxy_stats::write_stats and, in particular, updates
its "writes" statistic which counts the number of ongoing writes.

In the case that the paired view replica happens to be the *same* node,
we avoid calling send_to_endpoint() and call mutate_locally() instead.
That function does not take a write_stats object, so the "writes" statistic
doesn't get incremented for the duration of the write. So we should do
this explicitly.

Co-authored-by: Nadav Har'El <nyh@scylladb.com>
Co-authored-by: Duarte Nunes <duarte@scylladb.com>
(cherry picked from commit 1d5f8d0015)
2018-12-05 19:19:26 +00:00
Piotr Sarna
3cf26a60a2 auth: add abort_source to waiting for schema agreement
When the auth service is requested to stop during bootstrap,
it might have still not reached schema agreement.
Currently, waiting for this agreement is done in an infinite loop,
without taking abort_source into account.
This patch introduces checking if abort was requested
and breaking the loop in such case, so auth service can terminate.

Tests:
unit (release)
dtest (bootstrap_test.py:TestBootstrap.shutdown_wiped_node_cannot_join_test)
Message-Id: <1b7ded14b7c42254f02b5d2e10791eb767aae7fc.1543914769.git.sarna@scylladb.com>

(cherry picked from commit 7b0a3fbf8a)
2018-12-04 14:33:05 +00:00
Tomasz Grabiec
2103d0d52b sstables: Write Statistics.db offset map entries in the same order as Cassandra
Before this patch we were writing offset map enteies in unspecified
order, the one returned by std::unorderd_map. Cassandra writes them
sorted by metadata_type. Use the same order for improved
compatibility.

Fixes #3955.

Message-Id: <1543846649-22861-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit aa19f98d18)
2018-12-04 14:30:19 +02:00
Avi Kivity
16ee3b3ebe Merge "Make inactive shard readers evictable" from Botond
"
This series attempts to solve the regressions recently discovered in
performance of multi-partition range-scans. Namely that they:
* Flood the reader concurrency semaphore's queues, trampling other
  reads.
* Behave very badly when too many of them is running concurrently
  (trashing).
* May deadlock if enough of them is running without a timeout.

The solution for these problems is to make inactive shard readers
evictable. This should address all three issues listed above, to varying
degrees:
* Shard readers will now not cling onto their permits for the entire
  duration of the scan, which might be a lot of time.
* Will be less affected by infinite concurrency (more than the node can
  handle) as each scan now can make progress by evicting inactive shard
  readers belonging to other scans.
* Will not deadlock at all.

In addition to the above fix, this series also bundles two further
improvements:
* Add a mechanism to `reader_concurrecy_semaphore` to be notified of
  newly inserted evictables.
* General cleanups and fixes for `multishard_combining_reader` and
  `foreign_reader`.

I can unbundle these mini series and send them separately, if the
maintainers so prefer, altough considering that this series will have to
be backported to 3.0, I think this present form is better.

Fixes: #3835
"

* 'evictable-inactive-shard-readers/v7' of https://github.com/denesb/scylla: (27 commits)
  tests/multishard_mutation_query_test: test stateless query too
  tests/querier_cache: fail resource-based eviction test gracefully
  tests/querier_cache: simplify resource-based eviction test
  tests/mutation_reader_test: add test_multishard_combining_reader_next_partition
  tests/mutation_reader_test: restore indentation
  tests/mutation_reader_test: enrich pause-related multishard reader test
  multishard_combining_reader: use pause-resume API
  query::partition_slice: add clear_ranges() method
  position_in_partition: add region() accessor
  foreign_reader: add pause-resume API
  tests/mutation_reader_test: implement the pause-resume API
  query_mutations_on_all_shards(): implement pause-resume API
  make_multishard_streaming_reader(): implement the pause-resume API
  database: add accessors for user and streaming concurrency semaphores
  reader_lifecycle_policy: extend with a pause-resume API
  query_mutations_on_all_shards(): restore indentation
  query_mutations_on_all_shards(): simplify the state-machine
  multishard_combining_reader: use the reader lifecycle policy
  multishard_combining_reader: add reader lifecycle policy
  multishard_combining_reader: drop unnecessary `reader_promise` member
  ...

(cherry picked from commit 414b14a6bd)
2018-12-04 12:13:13 +02:00
Duarte Nunes
b0a9c40ab1 service/storage_proxy: Consider target liveness in sent_to_endpoint()
So we don't attempt to send mutations to unreachable endpoints and
instead store a hint for them, we now check the endpoint status and
populate dead_endpoints accordingly in
storage_proxy::send_to_endpoint().

Fixes #3820

Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20181007100640.2182-1-duarte@scylladb.com>
(cherry picked from commit 30d6ed8f92)
2018-12-03 18:38:05 +00:00
Duarte Nunes
53924e5c7f service/storage_proxy: Fix formatting of send_to_endpoint()
Signed-off-by: Duarte Nunes <duarte@scylladb.com>
Message-Id: <20181006204756.32232-1-duarte@scylladb.com>
(cherry picked from commit a69d468101)
2018-12-03 18:37:59 +00:00
Avi Kivity
befe0012f5 Merge "Fix multiple summary regeneration bugs." from Vladimir
"
This patchset addresses two recently discovered bugs both triggered by
summary regeneration:

Tests: unit {release}

+

Validated with debug build of Scylla (ASAN) that no use-after-free
occurs when re-generating Summary.db.
"

* 'projects/sstables-30/summary-regeneration/v1' of https://github.com/argenet/scylla:
  tests: Add test reading SSTables in 'mc' format with missing summary.
  sstables: When loading, read statistics before summary.
  database: Capture io_priority_class by reference to avoid dangling ref.

(cherry picked from commit 009cbd3dcb)
2018-12-02 13:32:09 +02:00
Duarte Nunes
1953c5fa61 Merge 'Fix filtering with LIMIT' from Piotr
"
This series adds proper handling of filtering queries with LIMIT.
Previously the limit was erroneously applied before filtering,
which leads to truncated results.

To avoid that, paged filtering queries now use an enhanced pager,
which remembers how many rows dropped and uses that information
to fetch for more pages if the limit is not yet reached.

For unpaged filtering queries, paging is done internally as in case
of aggregations to avoid returning keeping huge results in memory.

Also, previously, all limited queries used the page size counted
from max(page size, limit). It's not good for filtering,
because with LIMIT 1 we would then query for rows one-by-one.
To avoid that, filtered queries ask for the whole page and the results
are truncated if need be afterwards.

Tests: unit (release)
"

* 'fix_filtering_with_limit_2' of https://github.com/psarna/scylla:
  tests: add filtering with LIMIT test
  tests: split filtering tests from cql_query_test
  cql3: add proper handling of filtering with LIMIT
  service/pager: use dropped_rows to adjust how many rows to read
  service/pager: virtualize max_rows_to_fetch function
  cql3: add counting dropped rows in filtering pager

(cherry picked from commit 1afda28cf3)
2018-12-02 12:07:46 +02:00
Duarte Nunes
b72a94b53e Merge 'Fix checking if system tables need view updates' from Piotr
"
This miniseries ensures that system tables are not checked
for having view updates, because they never do.
What's more, distributed system table is used in the process,
so it's unsafe to query the table while streaming it.

Tests: unit (release), dtest(update_cluster_layout_tests.py:TestUpdateClusterLayout.simple_decommission_node_2_test)
"

* 'fix_checking_if_system_tables_need_view_updates_3' of https://github.com/psarna/scylla:
  streaming: don't check view building of system tables
  database: add is_internal_keyspace
  streaming: remove unused sstable_is_staging bool class

(cherry picked from commit d09d4bbd91)
2018-11-28 15:39:34 +00:00
Piotr Sarna
3f82b697f2 main: fix deinitialization order for view update generator
View update generator should be stopped only after
drain_on_shutdown() is performed on storage service.
Message-Id: <4d2bda4c73422a2ebf46d6dcd06c95d960839889.1543230849.git.sarna@scylladb.com>

(cherry picked from commit 6ab8235369)
2018-11-27 12:34:50 +00:00
Takuya ASADA
ee1ef853e5 dist/common/systemd/scylla-housekeeping-restart.service.mustache: specify correct repo for Debian variants
We do specify correct repo for both Red Hat/Debian variants on -deily, but
mistakenly don't for -restart, so do same on -restart.

Fixes #3906

Signed-off-by: Takuya ASADA <syuu@scylladb.com>
Message-Id: <20181109224509.27380-1-syuu@scylladb.com>
(cherry picked from commit 7740cd2142)
2018-11-27 09:59:05 +02:00
Raphael S. Carvalho
6e7e7f3822 sstables: deprecate sstable metadata's ancestors
The reason for that is that it's not available in sstable format mc,
so we can no longer rely on it in common code for the currently
supported formats.

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Message-Id: <20181121170057.20900-1-raphaelsc@scylladb.com>
(cherry picked from commit d29482dce8)
2018-11-24 12:36:40 +02:00
Paweł Dziepak
82a36edc9d Merge "Optimize sstable writing of the MC format" from Tomasz
"
Tested with perf_fast_forward from:

  github.com/tgrabiec/scylla.git perf_fast_forward-for-sst3-opt-write-v1

Using the following command line:

  build/release/tests/perf/perf_fast_forward_g --populate --sstable-format=mc \
     --data-directory /tmp/perf-mc --rows=10000000 -c1 -m4G \
     --datasets small-part

The average reported flush throughput was (stdev for the avergages is around 4k):
  - for mc before the series: 367848 frag/s
  - for lc before the series: 463458 frag/s (= mc.before +25%)
  - for mc after the series: 429276 frag/s (= mc.before +16%)
  - for lc after the series: 466495 frag/s (= mc.before +26%)

Refs #3874.
"

* tag 'sst3-opt-write-v2' of github.com:tgrabiec/scylla:
  sstables: mc: Avoid serialization of promoted index when empty
  sstables: mc: Avoid double serialization of rows
  tests: sstable 3.x: Do not compare Statistics component
  utils: Introduce memory_data_sink
  schema: Optimize column count getters
  sstables: checksummed_file_data_sink_impl: Bypass output_stream

(cherry picked from commit 4aa5d83590)
2018-11-24 12:36:40 +02:00
Avi Kivity
d4efa3c9b2 Update seastar submodule
* seastar d6647df...880826e (1):
  > fstream: Introduce make_file_data_sink()
2018-11-24 12:36:40 +02:00
Avi Kivity
324dae3e12 Merge "compress: Restore lz4 as default compressor" from Duarte
"
Enables sstable compression with LZ4 by default, which was the
long-time behavior until a regression turned off compression by
default.

Fixes #3926
"

* 'restore-default-compression/v2' of https://github.com/duarten/scylla:
  tests/cql_query_test: Assert default compression options
  compress: Restore lz4 as default compressor
  tests: Be explicit about absence of compression

(cherry picked from commit bb85a21a8f)
2018-11-21 16:45:22 +02:00
Tomasz Grabiec
c0ffc9a2b7 utils: phased_barrier: Make advance_and_await() have strong exception guarantees
Currently, when advance_and_await() fails to allocate the new gate
object, it will throw bad_alloc and leave the phased_barrier object in
an invalid state. Calling advance_and_await() again on it will result
in undefined behavior (typically SIGSEGV) beacuse _gate will be
disengaged.

One place affected by this is table::seal_active_memtable(), which
calls _flush_barrier.advance_and_await(). If this throws, subsequent
flush attempts will SIGSEGV.

This patch rearranges the code so that advance_and_await() has strong
exception guarantees.
Message-Id: <1542645562-20932-1-git-send-email-tgrabiec@scylladb.com>

Fixes #3931.

(cherry picked from commit 57e25fa0f8)
2018-11-21 12:17:27 +02:00
Glauber Costa
f81fa5f75c remove monitor if sstable write failed
In (almost) all SSTable write paths, we need to inform the monitor that
the write has failed as well. The monitor will remove the SSTable from
controller's tracking at that point.

Except there is one place where we are not doing that: streaming of big
mutations. Streaming of big mutations is an interesting use case, in
which it is done in 2 parts: if the writing of the SSTable fails right
away, then we do the correct thing.

But the SSTables are not commited at that point and the monitors are
still kept around with the SSTables until a later time, when they are
finally committed. Between those two points in time, it is possible that
the streaming code will detect a failure and manually call
fail_streaming_mutations(), which marks the SSTable for deletions. At
that point we should propagate that information to the monitor as well,
but we don't.

Fixes #3732 (hopefully)
Tests: unit (release)

Signed-off-by: Glauber Costa <glauber@scylladb.com>
Message-Id: <20181114213618.16789-1-glauber@scylladb.com>
(cherry picked from commit 9f403334c8)
2018-11-20 19:27:54 +02:00
Glauber Costa
6fd1cfcfce sstables: correctly parse estimated histograms
In commit a33f0d6, we changed the way we handle arrays during the write
and parse code to avoid reactor stalls. Some potentially big loops were
transformed into futurized loops, and also some calls to vector resizes
were replaced by a reserve + push_back idiom.

The latter broke parsing of the estimated histogram. The reason being
that the vectors that are used here are already initialized internally
by the estimated_histogram object. Therefore, when we push_back, we
don't fill the array all the way from index 0, but end up with a zeroed
beginning and only push back some of the elements we need.

We could revert this array to a resize() call. After all, the reason we
are using reserve + push_back is to avoid calling the constructor member
for each element, but We don't really expect the integer specialization
to do any of that.

However, to avoid confusion with future developers that may feel tempted
to converted this as well for the sake of consistency, it is safer to
just make sure these arrays are zeroed.

Fixes #3918

Signed-off-by: Glauber Costa <glauber@scylladb.com>
Message-Id: <20181116130853.10473-1-glauber@scylladb.com>
(cherry picked from commit c6811bd877)
2018-11-17 17:20:00 +02:00
Nadav Har'El
9d458ffea9 Materialized Views and Secondary Index: no longer experimental
After this patch, the Materialized Views and Secondary Index features
are considered generally-available and no longer require passing an
explicit "--experimental=on" flag to Scylla.

The "--experimental=on" flag and the db::config::check_experimental()
function remain unused, as we graduated the only two features which used
this flag. However, we leave the support for experimental features in
the code, to make it easier to add new experimental features in the future.
Another reason to leave the command-line parameter behind is so existing
scripts that still use it will not break.

Fixes #3917

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20181115144456.25518-1-nyh@scylladb.com>
(cherry picked from commit 78ed7d6d0c)
2018-11-15 19:50:30 +02:00
Duarte Nunes
9776a048e7 Merge 'Generating view updates during streaming' from Piotr
During streaming, there are cases when we should invoke the view write
path. In particular, if we're streaming because of repair or if a view
has not yet finished building and we're bootstrapping a new node.

The design constraints are:
1) The streamed writes should be visible to new writes, but the
   sstable should not participate in compaction, or we would lose the
   ability to exclude the streamed writes on a restart;
2) The streamed writes must not be considered when generating view
   updates for them;
3) Resilient to node restarts;
4) Resilient to concurrent stream sessions, possibly streaming mutations for overlapping ranges.

We achieve this by writing the streamed writes to an sstable in a
different folder, call it "staging". We achieve 1) by publishing the
sstable to the column family sstable set, but excluding it from
compactions. We do these steps upon boot, by looking at the staging
directory, thus achieving 3).

Fixes #3275

* 'streaming_view_to_staging_sstables_9' of https://github.com/psarna/scylla: (29 commits)
  tests: add materialized views test
  tests: add view update generator to cql test env
  main: add registering staging sstables read from disk
  database: add a check if loaded sstable is already staging
  database: add get_staging_sstable method
  streaming: stream tables with views through staging sstables
  streaming: add system distributed keyspace ref to streaming
  streaming: add view update generator reference to streaming
  main: add generating missed mv updates from staging sstables
  storage_service: move initializing sys_dist_ks before bootstrap
  db/view: add view_update_from_staging_generator service
  db/view: add view updating consumer
  table: add stream_view_replica_updates
  table: split push_view_replica_updates
  table: add as_mutation_source_excluding
  table: move push_view_replica_updates to table.cc
  database: add populating tables with staging sstables
  database: add creating /staging directory for sstables
  database: add sstable-excluding reader
  table: add move_sstable_from_staging_in_thread function
  ...

(cherry picked from commit a38f6078fb)
2018-11-15 17:46:20 +02:00
Asias He
10cf97375e streaming: Expose reason for streaming
On receiving a mutation_fragment or a mutation triggered by a streaming
operation, we pass an enum stream_reason to notify the receiver what
the streaming is used for. So the receiver can decide further operation,
e.g., send view updates, beyond applying the streaming data on disk.

Fixes #3276
Message-Id: <f15ebcdee25e87a033dcdd066770114a499881c0.1539498866.git.asias@scylladb.com>

(cherry picked from commit 7f826d3343)
2018-11-15 17:45:31 +02:00
Paweł Dziepak
e6355a9a01 Merge "Write static rows for all partitions if there are static columns" from Vladimir
"
It appears that in case when there are any static columns in serialization header,
Cassandra would write a (possibly empty) static row to every partition
in the SSTables file.

This patchset alings Scylla's logic with that of Cassandra.

Note that Scylla optimizes the case when no partition contains a static
row because it keeps track of updated columns that Scylla currently does
not do - see #3901 for details.

Fixes #3900.
"

* 'projects/sstables-30/write-all-static-rows/v1' of https://github.com/argenet/scylla:
  tests: Test writing empty static rows for partitions in tables with static columns.
  sstables: Ignore empty static rows on reading.
  sstables: Write empty static rows when there are static columns in the table.

(cherry picked from commit 6469a1b451)
2018-11-12 15:59:35 -08:00
Raphael S. Carvalho
e57907a1d5 sstables: fix procedure to get fully expired sstables with MC format
MC format lacks ancestors metadata, so we need to workaround it by using
ancestors in metadata collector, which is only available for a sstable
written during this instance. It works fine here because we only want
to know if a sstable recently compacted has an ancestor which wasn't
yet deleted.

Fixes #3852.

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Reviewed-by: Vladimir Krivopalov <vladimir@scylladb.com>
Message-Id: <20181102154951.22950-1-raphaelsc@scylladb.com>
(cherry picked from commit 1c5934c934)
2018-11-06 16:03:18 +02:00
Pekka Enberg
f94b46e7e0 docker: Switch to 3.0 RPM repository 2018-11-01 19:40:10 +02:00
Avi Kivity
6847c12668 Merge "dist: use perftune.py for disks tuning" from Vlad
"
Use perftune.py for tuning disks:
   - Distribute/pin disks' IRQs:
      - For NVMe drives: evenly among all present CPUs.
      - For non-NVMe drives: according to chosen tuning mode.
   - For all disks used by scylla:
      - Tune nomerges
      - Tune I/O scheduler.

It's important to tune NIC and disks together in order to keep IRQ
pinning in the same mode.

Disk are detected and tuned based on the current content of
/etc/scylla/scylla.yaml configuration file.
"

Fixes #3831.

* 'use_perftune_for_disks-v3' of https://github.com/vladzcloudius/scylla:
  dist: change the sysconfig parameter name to reflect the new semantics
  scylla_util.py::sysconfig_parser: introduce has_option()
  dist: scylla_setup and scylla_sysconfig_setup: change paremeters names to reflect new semantics
  dist: don't distribute posix_net_conf.sh any more
  dist: use perftune.py to tune disks and NIC

(cherry picked from commit f170e3e589)
2018-11-01 19:19:04 +02:00
Avi Kivity
80b86def1f Update seastar submodule
* seastar 0c8a2c8...d6647df (3):
  > scripts: perftune.py: properly merge parameters from the command line and the configuration file
  > scripts: perftune.py: prioritize I/O schedulers
  > Merge "scripts: perftune.py: support different I/O schedulers" from Vlad

Ref #3831.
2018-11-01 19:18:07 +02:00
Vlad Zolotarov
c6de9ea39b config: enable hinted handoff by default
Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
Message-Id: <20181019180401.12400-1-vladz@scylladb.com>
(cherry picked from commit 4d1bb719a4)
2018-11-01 10:41:44 +02:00
Avi Kivity
94bed81c1d Update seastar submodule
* seastar 39b89de...0c8a2c8 (1):
  > prometheus: Allow preemption between each metric

See scylladb/seastar#469.
2018-10-31 19:21:21 +02:00
Hagit Segev
0f3a21f0bb release: prepare for 3.0-rc1 2018-10-31 12:08:43 +02:00
Tomasz Grabiec
976db7e9e0 Merge "Proper support for static rows in SSTables 3.x" from Vladimir
This patchset addresses two issues with static rows support in SSTables
3.x. ('mc' format):

1. Since collections are allowed in static rows, we need to check for
complex deletion, set corresponding flag and write tombstones, if any.
2. Column indices need to be partitioned for static columns the same way
they are partitioned for regular ones.

 * github.com/argenet/scylla.git projects/sstables-30/columns-proper-order-followup/v1:
  sstables: Partition static columns by atomicity when reading/writing
    SSTables 3.x.
  sstables: Use std::reference_wrapper<> instead of a helper structure.
  sstables: Check for complex deletion when writing static rows.
  tests: Add/fix comments to
    test_write_interleaved_atomic_and_collection_columns.
  tests: Add test covering inverleaved atomic and collection cells in
    static row.

(cherry picked from commit 62c7685b0d)
2018-10-30 14:51:21 +01:00
Nadav Har'El
996b86b804 Materalized views: fix race condition in resharding while view building
When a node reshards (i.e., restarts with a different number of CPUs), and
is in the middle of building a view for a pre-existing table, the view
building needs to find the right token from which to start building on all
shards. We ran the same code on all shards, hoping they would all make
the same decision on which token to continue. But in some cases, one
shard might make the decision, start building, and make progress -
all before a second shard goes to make the decision, which will now
be different.

This resulted, in some rare cases, in the new materialized view missing
a few rows when the build was interrupted with a resharding.

The fix is to add the missing synchronization: All shards should make
the same decision on whether and how to reshard - and only then should
start building the view.

Fixes #3890
Fixes #3452

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20181028140549.21200-1-nyh@scylladb.com>
(cherry picked from commit b8337f8c9d)
2018-10-29 09:52:25 +00:00
Avi Kivity
b7b217cc43 Merge "Re-order columns when reading/writing SSTables 3.x" from Vladimir
"
In Cassandra, row columns are stored in a BTree that uses the following
ordering on them:
    - all atomic columns go first, then all multi-cell ones
    - columns of both types (atomic and multi-cell) are
      lexicographically ordered by name regarding each other

Scylla needs to store columns and their respective indices using the
same ordering as well as when reading them back.

Fixes #3853

Tests: unit {release}

+

Checked that the following SSTables are dumped fine using Cassandra's
sstabledump:

cqlsh:sst3> CREATE TABLE atomic_and_collection3 ( pk int, ck int, rc1 text, rc2 list<text>, rc3 text, rc4 list<text>, rc5 text, rc6 list<text>, PRIMARY KEY (pk, ck)) WITH compression = {'sstable_compression': ''};
cqlsh:sst3> INSERT INTO atomic_and_collection3 (pk, ck, rc1, rc4, rc5) VALUES (0, 0, 'hello', ['beautiful','world'], 'here');
<< flush >>

sstabledump:

[
  {
    "partition" : {
      "key" : [ "0" ],
      "position" : 0
    },
    "rows" : [
      {
        "type" : "row",
        "position" : 96,
        "clustering" : [ 0 ],
        "liveness_info" : { "tstamp" : "1540599270139464" },
        "cells" : [
          { "name" : "rc1", "value" : "hello" },
          { "name" : "rc5", "value" : "here" },
          { "name" : "rc4", "deletion_info" : { "marked_deleted" : "1540599270139463", "local_delete_time" : "1540599270" } },
          { "name" : "rc4", "path" : [ "45e22cb0-d97d-11e8-9f07-000000000000" ], "value" : "beautiful" },
          { "name" : "rc4", "path" : [ "45e22cb1-d97d-11e8-9f07-000000000000" ], "value" : "world" }
        ]
      }
    ]
  }
]
"

* 'projects/sstables-30/columns-proper-order/v1' of https://github.com/argenet/scylla:
  tests: Test interleaved atomic and multi-cell columns written to SSTables 3.x.
  sstables: Re-order columns (atomic first, then collections) for SSTables 3.x.
  sstables: Use a compound structure for storing information used for reading columns.

(cherry picked from commit 75dbff984c)
2018-10-28 15:51:47 +02:00
Tomasz Grabiec
c274430933 Merge "Properly write static rows missing columns for SSTables 3.x." from Vladimir
Before this fix, write_missing_columns() helper would always deal with
regular columns even when writing static rows.

This would cause errors on reading those files.

Now, the missing columns are written correctly for regular and static
rows alike.

* github.com/argenet/scylla.git projects/sstables-30/fix-writing-static-missing-columns/v1:
  schema: Add helper method returning the count of columns of specified
    kind.
  sstables: Honour the column kind when writing missing columns in 'mc'
    format.
  tests: Add test for a static row with missing columns (SStables 3.x.).

(cherry picked from commit cf2d5c19fb)
2018-10-26 13:30:12 +03:00
Avi Kivity
893a18a7c4 Merge "Properly writing/reading shadowable deletions with SSTables 3.x." from Vladimir
"
This patchset adddresses two problems with shadowable deletions handling
in SSTables 3.x. ('mc' format).

Firstly, we previously did not set a flag indicating the presence of
extended flags byte with HAS_SHADOWABLE_DELETION bitmask on writing.
This would break subsequent reading and cause all types of failures up
to crash.

Secondly, when reading rows with this extended flag set, we need to
preserve that information and create a shadowable_tombstone for the row.

Tests: unit {release}
+

Verified manually with 'hexdump' and using modified 'sstabledump' that
second (shadowable) tombstone is written for MV tables by Scylla.

+
DTest (materialized_views_test.py:TestMaterializedViews.hundred_mv_concurrent_test)
that originally failed due to this issue has successfully passed locally.
"

* 'projects/sstables-30/shadowable-deletion/v4' of https://github.com/argenet/scylla:
  tests: Add tests writing both regular and shadowable tombstones to SSTables 3.x.
  tests: Add test covering writing and reading a shadowable tombstone with SSTables 3.x.
  sstables: Support Scylla-specific extension for writing shadowable tombstones.
  sstables: Introduce a feature for shadowable tombstones in Scylla.db.
  memtable: Track regular and shadowable tombstones separately in encoding_stats_collector.
  sstables: Error out when reading SSTables 3.x with Cassandra shadowable deletion.
  sstables: Support checking row extension flags for Cassandra shadowable deletion.

(cherry picked from commit 8210f4c982)
2018-10-24 19:32:57 +03:00
Tomasz Grabiec
39b39058fc sstable_mutation_reader: Do not read partition index when scanning
Even when we're using a full clustering range, need_skip() will return
true when we start a new partition and advance_context() will be
called with position_in_partition::before_all_clustered_rows(). We
should detect that there is no need to skip to that position before
the call to advance_to(*_current_partition_key), which will read the
index page.

Fixes #3868.

Message-Id: <1539881775-8578-1-git-send-email-tgrabiec@scylladb.com>
(cherry picked from commit 9e756d3863)
2018-10-24 19:32:40 +03:00
Avi Kivity
6bf4a73d88 thrift: limit message size
Limit message size according to the configuration, to avoid a huge message from
allocating all of the server's memory.

We also need to limit memory used in aggregate by thrift, but that is left to
another patch.

Fixes #3878.
Message-Id: <20181024081042.13067-1-avi@scylladb.com>

(cherry picked from commit a9836ad758)
2018-10-24 19:32:25 +03:00
Gleb Natapov
ca4846dd63 stream_session: remove unused capture
'Consumer function' parameter for distribute_reader_and_consume_on_shards()
captures schema_ptr (which is a seastar::shared_ptr), but the function
is later copied on another shard at which point schema_ptr is also copied
and its counter is incremented by the wrong shard. The capture is not
even used, so lets just drop it.

Fixes #3838

Message-Id: <20181011075500.GN14449@scylladb.com>
(cherry picked from commit ceb361544a)
2018-10-24 09:47:02 +03:00
Takuya ASADA
2663ff7bc1 dist/common/sysctl.d: add new conf file to set fs.aio-max-nr
We need raise fs.aio-max-nr to larger value since Seastar may allocates
more then 65535 AIO events (= kernel default value)

Fixes #3842

Signed-off-by: Takuya ASADA <syuu@scylladb.com>
Message-Id: <20181023030449.15445-1-syuu@scylladb.com>
(cherry picked from commit 950dbdb466)
2018-10-24 09:45:51 +03:00
Tomasz Grabiec
043a575fcd Merge "Correctly handle dropped columns in SSTable 3" from Piotr J.
Previously we were making assumptions about missing columns
(the size of its value, whether it's a collection or a counter) but
they didn't have to be always true. Now we're using column type
from serialization header to use the right values.

Fixes #3859

* seastar-dev.git haaawk/projects/sstables-30/handling-dropped-columns/v4:
  sstables 3: Correctly handle dropped columns in column_translation
  sstables 3: Add test for dropped columns handling

(cherry picked from commit fc37b80d24)
2018-10-24 09:45:25 +03:00
Vlad Zolotarov
00dc400993 storage_proxy::query_result_local: create a single tracing span on a replica shard
Every call of a tracing::global_trace_state_ptr object instead of a
tracing::tracing_state_ptr or a call to tracing::global_trace_state_ptr::get()
creates a new tracing session (span) object.

This should never be done unless query handling moves to a different shard.

Fixes #3862

Signed-off-by: Vlad Zolotarov <vladz@scylladb.com>
Message-Id: <20181018003500.10030-1-vladz@scylladb.com>
(cherry picked from commit a87c11bad2)
2018-10-24 09:45:14 +03:00
Duarte Nunes
522a48a244 Merge 'Fix for a select statement with filtered columns' from Eliran
"
This patchset fixes #3803. When a select statement with filtering
is executed and the column that is needed for the filtering is not
present in the select clause, rows that should have been filtered out
according to this column will still be present in the result set.

Tests:
 1. The testcase from the issue.
 2. Unit tests (release) including the
 newly added test from this patchset.
"

* 'issues/3803/v10' of https://github.com/eliransin/scylla:
  unit test: add test for filtering queries without the filtered column
  cql3 unit test: add assertion for the number of serialized columns
  cql3: ensure retrieval of columns for filtering
  cql3: refactor find_idx to be part of statement restrictions object
  cql3: add prefix size common functionality to all clustering restrictions
  cql3: rename selection metadata manipulation functions

(cherry picked from commit 3fe92663d4)
2018-10-24 09:44:46 +03:00
Paweł Dziepak
5faa28ce45 cql3: restore original timeout behaviour for aggregate queries
Commit 1d34ef38a8 "cql3: make pagers use
time_point instead of duration" has unintentionally altered the timeout
semantics for aggregate queries. Such requests fetch multiple pages before
sending a response to the client. Originally, each of those fetches had
a timeout-duration to finish, after the problematic commit the whole
request needs to complete in a single timeout-duration. This,
unsurprisingly, makes some queries that were successful before fail with
a timeout. This patch restores the original behaviour.

Fixes #3877.

Message-Id: <20181022125318.4384-1-pdziepak@scylladb.com>
(cherry picked from commit c94d2b6aa6)
2018-10-24 09:43:59 +03:00
Avi Kivity
52be02558e config: mark range_request_timeout_in_ms and request_timeout_in_ms as Used
This makes them available in scylla --help.

Fixes #3884.
Message-Id: <20181023101150.29856-1-avi@scylladb.com>

(cherry picked from commit d9e0ea6bb0)
2018-10-24 09:43:54 +03:00
Avi Kivity
a7cbfbe63f Merge "hinted handoff: give a sender a low priority" from Vlad
"
Hinted handoff should not overpower regular flows like READs, WRITEs or
background activities like memtable flushes or compactions.

In order to achieve this put its sending in the STEAMING CPU scheduling
group and its commitlog object into the STREAMING I/O scheduling group.

Fixes #3817
"

* 'hinted_handoff_scheduling_groups-v2' of https://github.com/vladzcloudius/scylla:
  db::hints::manager: use "streaming" I/O scheduling class for reads
  commitlog::read_log_file(): set the a read I/O priority class explicitly
  db::hints::manager: add hints sender to the "streaming" CPU scheduling group

(cherry picked from commit 1533487ba8)
2018-10-24 09:43:39 +03:00
Duarte Nunes
28fd2044d2 Merge 'hinted handoff: add manager::state and split storing and replaying enablement' from Vlad
"
Refs #3828
(Probably fixes it)

We found a few flaws in a way we enable hints replaying.
First of all it was allowed before manager::start() is complete.
Then, since manager::start() is called after messaging_service is
initialized there was a time window when hints are rejected and this
creates an issue for MV.

Both issues above were found in the context of #3828.

This series fixes them both.

Tested {release}:
dtest: materialized_views_test.py:TestMaterializedViews.write_to_hinted_handoff_for_views_test
dtest: hintedhandoff_additional_test.py
"

* 'hinted_handoff_dont_create_hints_until_started-v1' of https://github.com/vladzcloudius/scylla:
  hinted handoff: enable storing hints before starting messaging_service
  db::hints::manager: add a "started" state
  db::hints::manager: introduce a _state

(cherry picked from commit 3a53b3cebc)
2018-10-24 09:43:03 +03:00
Calle Wilund
76ff2e5c3d messaging_service: Make rpc streaming sink respect tls connection
Fixes #3787

Message service streaming sink was created using direct call to
rpc::client::make_sink. This in turn needs a new socker, which it
creates completely ignoring what underlying transport is active for the
client in question.

Fix by retaining the tls credential pointer in the client wrapper, and
using this in a sink method to determine whether to create a new tls
socker, or just go ahead with a plain one.

Message-Id: <20181010003249.30526-1-calle@scylladb.com>
(cherry picked from commit 3cb50c861d)
2018-10-23 07:36:21 +00:00
Avi Kivity
7b34d54a96 locator: fix abstract_replication_strategy::get_ranges() and friends violating sort order
get_ranges() is supposed to return ranges in sorted order. However, a35136533d
broke this and returned the range that was supposed to be last in the second
position (e.g. [0, 10, 1, 2, 3, 4, 5, 6, 7, 8, 9]). The broke cleanup, which
relied on the sort order to perform a binary search. Other users of the
get_ranges() family did not rely on the sort order.

Fixes #3872.
Message-Id: <20181019113613.1895-1-avi@scylladb.com>

(cherry picked from commit 1ce52d5432)
2018-10-23 07:36:21 +00:00
Duarte Nunes
26c31f6798 Merge "db/hints: Expose current backlog" from Duarte
"
Hints are stored on disk by a hints::manager, ensuring they are
eventually sent. A hints::resource_manager ensures the hints::managers
it tracks don't consume more than their allocated resources by
monitoring disk space and disabling new hints if needed. This series
fixes some bugs related to the backlog calculation, but mainly exposes
the backlog through a hints::manager so upper layers can apply flow
control.

Refs #2538
"

* 'hh-manager-backlog/v3' of https://github.com/duarten/scylla:
  db/hints/manager: Expose current backlog
  db/hints/manager: Move decision about blocking hints to the manager
  db/hints/resource_manager: Correctly account resources in space_watchdog
  db/hints/resource_manager: Replace timer with seastar::thread
  db/hints/resource_manager: Ensure managers are correctly registered
  db/hints/resource_manager: Fix formatting
  db/hints: Disallow moving or copying the managers
2018-10-23 07:36:21 +00:00
Glauber Costa
28fa66591a sstables: print sstable path in case of an exception
Without that, we don't know where to look for the problems

Before:

 compaction failed: sstables::malformed_sstable_exception (Too big ttl: 3163676957)

After:

 compaction_manager - compaction failed: sstables::malformed_sstable_exception (Too big ttl: 4294967295 in sstable /var/lib/scylla/data/system_traces/events-8826e8e9e16a372887533bc1fc713c25/mc-832-big-Data.db)

Signed-off-by: Glauber Costa <glauber@scylladb.com>
Message-Id: <20181016181004.17838-1-glauber@scylladb.com>
(cherry picked from commit 7edae5421d)
2018-10-23 07:36:14 +00:00
Piotr Sarna
0fee1d9e43 cql3: add asking for pk/ck in the base query
Base query partition and clustering keys are used to generate
paging state for an index query, so they always need to be present
when a paged base query is processed.
Message-Id: <f3bf69453a6fd2bc842c8bdbd602d62c91cf9218.1538568953.git.sarna@scylladb.com>

Fixes #3855.
(cherry picked from commit 4a23297117)
2018-10-16 19:59:42 +03:00
Piotr Sarna
76e72e28f4 cql3: add checking for may_need_paging when executing base query
It's not sufficient to check for positive page_size when preparing
a base query for indexed select statement - may_need_paging() should
be called as well.
Message-Id: <d435820019e4082a64ca9807541f0c9ad334e6a8.1538568953.git.sarna@scylladb.com>

(cherry picked from commit 50d3de0693)
2018-10-16 19:58:58 +03:00
Piotr Sarna
f969e80965 cql3: move base query command creation to a separate function
Message-Id: <6b48b8cbd6312da4a17bfd3c85af628b4215e9f4.1538568953.git.sarna@scylladb.com>
(cherry picked from commit 11b8831c04)
2018-10-16 19:58:56 +03:00
Vladimir Krivopalov
2029134063 sstables: Reset opened range tombstone when moving to another partition.
Signed-off-by: Vladimir Krivopalov <vladimir@scylladb.com>
Message-Id: <f6dc6b0bd88ca44f2ef84c2a8bee43fde82c89cc.1539396572.git.vladimir@scylladb.com>
(cherry picked from commit 092276b13d)
2018-10-15 13:26:22 +03:00
Vladimir Krivopalov
f30fe7bd17 sstables: Factor out code resetting values for a new partition.
Signed-off-by: Vladimir Krivopalov <vladimir@scylladb.com>
Message-Id: <83a3a4ce6942b036be447bcfeb66142828e75293.1539396572.git.vladimir@scylladb.com>
(cherry picked from commit 926b6430fd)
2018-10-15 13:26:20 +03:00
Piotr Sarna
aeb418af9e service/pager: avoid dereferencing null partition key
The pager::state() function returns a valid paging object even
if the pager itself is exhausted. It may also not contain the partition
key, so using it unconditionally was a bug - now, in case there is no
partition key present, paging state will contain an empty partition key.

Fixes #3829

Message-Id: <28401eb21ab8f12645c0a33d9e92ada9de83e96b.1539074813.git.sarna@scylladb.com>
(cherry picked from commit b3685342a6)
2018-10-15 12:47:25 +03:00
Glauber Costa
714e6d741f api: use longs instead of ints for snapshot sizes
Int types in json will be serialized to int types in C++. They will then
only be able to handle 4GB, and we tend to store more data than that.

Without this patch, listsnapshots is broken in all versions.

Fixes: #3845

Signed-off-by: Glauber Costa <glauber@scylladb.com>
Message-Id: <20181012155902.7573-1-glauber@scylladb.com>
(cherry picked from commit 98332de268)
2018-10-12 22:01:59 +03:00
Tomasz Grabiec
95c5872450 Merge "Enable sstable_mutation_test with SSTables 3.x." from Vladimir
Introduce uppermost_bound() method instead of upper_bound() in mutation_fragment_filter and clustering_ranges_walker.

For now, this has been only used to produce the final range tombstone
for sliced reads inside consume_partition_end().

Usage of the upper bound of the current range causes problems of two
kinds:
    1. If not all the slicing ranges have been traversed with the
    clustering range walker, which is possible when the last read
    mutation fragment was before some of the ranges and reading was limited
    to a specific range of positions taken from index, the emitted range
    tombstone will not cover the untraversed slices.

    2. At the same time, if all ranges have been walked past, the end
    bound is set to after_all_clustered_rows and the emitted RT may span
    more data than it should.

To avoid both situations, the uppermost bound is used instead, which
refers to the upper bound of the last range in the sequence.

* github.com/scylladb/seastar-dev.git haaawk/projects/sstables-30/enable-mc-with-sstable-mutation-test/v2
  sstables: Use uppermost_bound() instead of upper_bound() in
    mutation_fragment_filter.
  tests: Enable sstable_mutation_test for SSTables 'mc' format.

Rebased by Piotr J.

(cherry picked from commit b89556512a)
2018-10-12 17:46:49 +03:00
Tomasz Grabiec
87f8968553 Merge "Make SST3 pass test_clustering_slices test" from Piotr
* seastar-dev.git haaawk/sst3/test_clustering_slices/v8:
  sstables: Extract on_end_of_stream from consume_partition_end
  sstables: Don't call consume_range_tombstone_end in
    consume_partition_end
  sstables: Change the way fragments are returned from consumer

(cherry picked from commit 193efef950)
2018-10-12 17:46:46 +03:00
Tomasz Grabiec
2895428d44 Merge "Handle dead row markers when writing to SSTables 3.x" from Vladimir
There is a mismatch between row markers used in SSTables 2.x (ka/la) and
liveness_info used by SSTables 3.x (mc) in that a row marker can be
written as a deleted cell but liveness_info cannot.

To handle this, for a dead row marker the corresponding liveness_info is
written as expiring liveness_info with a fake TTL set to 1.
This approach is adapted from the solution for CASSANDRA-13395 that
exercised similar issue during SSTables upgrades.

* github.com/argenet/scylla.git projects/sstables-30/dead-row-marker/v7:
  sstables: Introduce TTL limitation and special 'expired TTL' value.
  sstables: Write dead row marker as expired liveness info.
  tests: Add test covering dead row marker writing to SSTables 3.x.

(cherry picked from commit a7a14e3af2)
2018-10-11 15:03:58 +03:00
Botond Dénes
e18f182cfc multishard_mutation_query(): don't attempt to stop broken readers
Currently, when stopping a reader fails, it simply won't be attempted to
be saved, and it will be left in the `_readers` array as-is. This can
lead to an assertion failure as the reader state will contain futures
that were already waited upon, and that the cleanup code will attempt to
wait on again. To prevent this, when stopping a reader fails, reset it
to nonexistent state, so that the cleanup code doesn't attempt to do
anything with it.

Refs: #3830

Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <a1afc1d3d74f196b772e6c218999c57c15ca05be.1539088164.git.bdenes@scylladb.com>
(cherry picked from commit d467b518bc)
2018-10-10 10:12:00 +03:00
Vladimir Krivopalov
cf8cdbf87d sstables: Add missing 'mc' format into format strings map in sstable::filename().
Fixes #3832.

Signed-off-by: Vladimir Krivopalov <vladimir@scylladb.com>
Message-Id: <269421fb2ac8ab389231cbe9ed501da7e7ff936a.1539048008.git.vladimir@scylladb.com>
(cherry picked from commit e9aba6a9c3)
2018-10-10 05:53:28 +03:00
Avi Kivity
eb2814067d Update seastar submodule
* seastar 5712816...39b89de (1):
  > prometheus: Fix histogram text representation

Fixes #3827.
2018-10-09 16:35:54 +03:00
Avi Kivity
0c722d4547 Point seastar submodule at scylla-seastar.git
This allows us to freeze this branch's Seastar and only backport selected fixes.
2018-10-09 16:29:53 +03:00
Nadav Har'El
54cf463430 materialized views: refuse to filter by non-key column
A materialized views can provide a filter so as to pick up only a subset
of the rows from the base table. Usually, the filter operates on columns
from the base table's primary key. If we use a filter on regular (non-key)
columns, things get hairy, and as issue #3430 showed, wrong: merely updating
this column in the base table may require us to delete, or resurrect, the
view row. But normally we need to do the above when the "new view key column"
was updated, when there is one. We use shadowable tombstones with one
timestamp to do this, so it cannot take into account the two timestamp from
those two columns (the filtered column and the new key column).

So in the current code, filtering by a non-key column does not work correctly.
In this patch we provide two test cases (one involving TTLs, and one involves
only normal updates), which demonstrate vividly that it does *not* work
correctly. With normal updates, trying to resurect a view row that has
previously disappeared, fails. With TTLs, things are even worse, and the view
row fails to disappear when the filtered column is TTLed.

In Cassandra, the same thing doesn't work correctly as well (see
CASSANDRA-13798 and CASSANDRA-13832) so they decided to refuse creating
a materialized view filtering a non-key column. In this patch we also
do this - fail the creation of such an unsupported view. For this reason,
the two tests mentioned above are commented out in a "#if", with, instead,
a trivial test verifying a failure to create such a view.

Note that as explained above, when the filtered column and new view key
column are *different* we have a problem. But when they are the *same* - namely
we filter by a non-key base column which actually *is* a key in the view -
we are actually fine. This patch includes additional test cases verifying
that this case is really fine and provides correct results. Accordingly,
this case is *not* forbidden in the view creation code.

Fixes #3430.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20181008185633.24616-1-nyh@scylladb.com>
(cherry picked from commit b8668dc0f8)
2018-10-09 10:18:58 +01:00
Nadav Har'El
d2a0622edd materialized views: enable two tests in view_schema_test
We had two commented out tests based on Cassandra's MV unit tests, for
the case that the view's filter (the "SELECT" clause used to define the
view) filtered by a non-primary-key column. These tests used to fail
because of problems we had in the filtering code, but they now succeed,
so we can enable them. This patch also adds some comments about what
the tests do, and adds a few more cases to one of the tests.

Refs #3430.

However, note that the success of these tests does not really prove that
the non-PK-column filtering feature works fully correctly and that issue
forbidding it, as explained in
https://issues.apache.org/jira/browse/CASSANDRA-13798. We can probably
fix this feature with our "virtual cells" mechanism, but will need to add
a test to confirm the possible problem and its (probably needed fix).
We do not add such a test in this patch.

In the meantime, issue #3430 should remain open: we still *allow* users
to create MV with such a filter, and, as the tests in this patch show,
this "mostly" works correctly. We just need to prove and/or fix what happens
with the complex row liveness issues a la issue #3362.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20181004213637.32330-1-nyh@scylladb.com>
(cherry picked from commit e4ef7fc40a)
2018-10-09 10:18:54 +01:00
Duarte Nunes
60edaec757 Merge 'Fix issues with endpoint state replication to other shards' from Tomasz
Fixes #3798
Fixes #3694

Tests:

  unit(release), dtest([new] cql_tests.py:TruncateTester.truncate_after_restart_test)

* tag 'fix-gossip-shard-replication-v1' of github.com:tgrabiec/scylla:
  gms/gossiper: Replicate enpoint states in add_saved_endpoint()
  gms/gossiper: Make reset_endpoint_state_map() have effect on all shards
  gms/gossiper: Replicate STATUS change from mark_as_shutdown() to other shards
  gms/gossiper: Always override states from older generations

(cherry picked from commit 48ebe6552c)
2018-10-09 10:14:30 +03:00
Avi Kivity
5802532cb3 Merge "Fix mutation fragments clobbering on fast_forward" from Vladimir
"
This patchset fixes a bug in SSTables 3.x reading when fast-forwarding
is enabled. It is possible that a mutation fragment, row or RT marker,
is read and then stored because it falls outside the current
fast-forwarding range.

If the reader is further fast-forwarded but the
row still falls outside of it, the reader would still continue reading
and get the next fragment, if any, that would clobber the currently
stored one. With this fix, the reader does not attempt to read on
after storing the current fragment.

Tests: unit {release}
"

* 'projects/sstables-30/row-skipped-on-double-ff/v2' of https://github.com/argenet/scylla:
  tests: Add test for reading rows after multiple fast-forwarding with SSTables 3.x.
  sstables: mp_row_consumer_m to notify reader on end of stream when storing a mutation fragment.
  sstables: In mp_row_consumer_m::push_mutation_fragments(), return the called helper's value.

(cherry picked from commit 0fa60660b8)
2018-10-09 09:35:51 +03:00
Eliran Sinvani
83ea91055e cql3 : add workaround to antlr3 null dereference bug
The Antlr3 exception class has a null dereference bug that crashes
the system when trying to extract the exception message using
ANTLR_Exception<...>::displayRecognitionError(...) function. When
a parsing error occurs the CqlParser throws an exception which in
turn processesed for some special cases in scylla to generate a custom
message. The default case however, creates the message using
displayRecognitionError, causing the system to crash.
The fix is a simple workaround, making sure the pointer is not null
before the call to the function. A "proper" fix can't be implemented
because the exception class itself is implemented outside scylla
in antlr headers that resides on the host machine os.

Tested manualy 2 testcases, a typo causing scylla to crash and
a cql comment without a newline at the end also caused scylla to crash.
Ran unit tests (release).

Fixes #3740
Fixes #3764

Signed-off-by: Eliran Sinvani <eliransin@scylladb.com>
Message-Id: <cfc7e0d758d7a855d113bb7c8191b0fd7d2e8921.1538566542.git.eliransin@scylladb.com>
(cherry picked from commit 20f49566a2)
2018-10-04 14:07:25 +03:00
Piotr Sarna
e7863d3d54 tests: add missing get() calls in threaded context
One test case missed a few get() calls in order to wait
for continuations, which only accidentally worked,
because it was followed by 'eventually()' blocks.
Message-Id: <69c145575ac81154c4b5f500d01c6b045a267088.1536839959.git.sarna@scylladb.com>

(cherry picked from commit a5570cb288)
2018-10-04 14:06:50 +03:00
Piotr Sarna
57f124b905 tests: add collections test for secondary indexing
Test case regarding creating indexes on collection columns
is added to the suite.

Refs #3654
Refs #2962
Message-Id: <1b6844634b6e9a353028545813571647c92fb330.1536839959.git.sarna@scylladb.com>

(cherry picked from commit 8a2abd45fb)
2018-10-04 14:06:48 +03:00
Piotr Sarna
40d8de5784 cql3: prevent creation of indexes on non-frozen collections
Until indexes for non-frozen collections is implemented,
creating such indexes should be disallowed to prevent unnecessary
errors on insertions/selections.

Fixes #3653
Refs #2962
Message-Id: <218cf96d5e38340806fb9446b8282d2296ba5f43.1536839959.git.sarna@scylladb.com>

(cherry picked from commit 2d355bdf47)
2018-10-04 14:06:47 +03:00
Avi Kivity
1468ec62de Merge "Handle simple column type schema changes in SST3" from Piotr
"
This patchset enables very simple column type conversions.
It covers only handling variable and fixed size type differences.
Two types still have to be compatiple on bits level to be able to convert a field from one to the other.
"

* 'haaawk/sst3/column_type_schema_change/v4' of github.com:scylladb/seastar-dev:
  Fix check_multi_schema to actually check the column type change
  Handle very basic column type conversions in SST3
  Enable check_multi_schema for SST3

(cherry picked from commit b9702222f8)
2018-10-03 17:44:26 +03:00
Avi Kivity
c6ef56ae1e Revert "compaction: demote compaction start/end messages to DEBUG level"
This reverts commit b443a9b930. The compaction
history table doesn't have enough information to be a replacement for this
log message yet.

(cherry picked from commit 7c8143c3c4)
2018-10-03 17:44:21 +03:00
Avi Kivity
ad62313b86 utils: crc32: mark power crc32 assembly as not requiring an executable stack
The linker uses an opt-in system for non-executable stack: if all object files
opt into a non-executable stack, the binary will have a non-executable stack,
which is very desirable for security. The compiler cooperates by opting into
a non-executable stack whenever possible (always for our code).

However, we also have an assembly file (for fast power crc32 computations).
Since it doesn't opt into a non-executable stack, we get a binary with
executable stack, which Gentoo's build system rightly complains about.

Fix by adding the correct incantation to the file.

Fixes #3799.

Reported-by: Alexys Jacob <ultrabug@gmail.com>
Message-Id: <20181002151251.26383-1-avi@scylladb.com>
(cherry picked from commit aaab8a3f46)
2018-10-02 23:22:56 +03:00
Avi Kivity
de87f798e1 release: prepare for 3.0-rc0 2018-10-02 12:00:50 +03:00
3956 changed files with 63676 additions and 192672 deletions

View File

@@ -1,4 +0,0 @@
.git
build
seastar/build
testlog

87
.github/CODEOWNERS vendored
View File

@@ -1,87 +0,0 @@
# AUTH
auth/* @elcallio @vladzcloudius
# CACHE
row_cache* @tgrabiec @haaawk
*mutation* @tgrabiec @haaawk
tests/mvcc* @tgrabiec @haaawk
# CDC
cdc/* @haaawk @kbr- @elcallio @piodul @jul-stas
test/cql/cdc_* @haaawk @kbr- @elcallio @piodul @jul-stas
test/boost/cdc_* @haaawk @kbr- @elcallio @piodul @jul-stas
# COMMITLOG / BATCHLOG
db/commitlog/* @elcallio
db/batch* @elcallio
# COORDINATOR
service/storage_proxy* @gleb-cloudius
# COMPACTION
sstables/compaction* @raphaelsc @nyh
# CQL TRANSPORT LAYER
transport/* @penberg
# CQL QUERY LANGUAGE
cql3/* @tgrabiec @penberg @psarna
# COUNTERS
counters* @haaawk @jul-stas
tests/counter_test* @haaawk @jul-stas
# GOSSIP
gms/* @tgrabiec @asias
# DOCKER
dist/docker/* @penberg
# LSA
utils/logalloc* @tgrabiec
# MATERIALIZED VIEWS
db/view/* @nyh @psarna
cql3/statements/*view* @nyh @psarna
test/boost/view_* @nyh @psarna
# PACKAGING
dist/* @syuu1228
# REPAIR
repair/* @tgrabiec @asias @nyh
# SCHEMA MANAGEMENT
db/schema_tables* @tgrabiec @nyh
db/legacy_schema_migrator* @tgrabiec @nyh
service/migration* @tgrabiec @nyh
schema* @tgrabiec @nyh
# SECONDARY INDEXES
db/index/* @nyh @penberg @psarna
cql3/statements/*index* @nyh @penberg @psarna
test/boost/*index* @nyh @penberg @psarna
# SSTABLES
sstables/* @tgrabiec @raphaelsc @nyh
# STREAMING
streaming/* @tgrabiec @asias
service/storage_service.* @tgrabiec @asias
# ALTERNATOR
alternator/* @nyh @psarna
test/alternator/* @nyh @psarna
# HINTED HANDOFF
db/hints/* @haaawk @piodul @vladzcloudius
# REDIS
redis/* @nyh @syuu1228
redis-test/* @nyh @syuu1228
# READERS
reader_* @denesb
querier* @denesb
test/boost/mutation_reader_test.cc @denesb
test/boost/querier_cache_test.cc @denesb

4
.github/PULL_REQUEST_TEMPLATE.md vendored Normal file
View File

@@ -0,0 +1,4 @@
Scylla doesn't use pull-requests, please send a patch to the [mailing list](mailto:scylladb-dev@googlegroups.com) instead.
See our [contributing guidelines](../CONTRIBUTING.md) and our [Scylla development guidelines](../HACKING.md) for more information.
If you have any questions please don't hesitate to send a mail to the [dev list](mailto:scylladb-dev@googlegroups.com).

6
.gitignore vendored
View File

@@ -19,9 +19,3 @@ CMakeLists.txt.user
__pycache__CMakeLists.txt.user
.gdbinit
resources
.pytest_cache
/expressions.tokens
tags
testlog
test/*/*.reject
.vscode

17
.gitmodules vendored
View File

@@ -1,23 +1,14 @@
[submodule "seastar"]
path = seastar
url = ../seastar
url = ../scylla-seastar
ignore = dirty
[submodule "swagger-ui"]
path = swagger-ui
url = ../scylla-swagger-ui
ignore = dirty
[submodule "xxHash"]
path = xxHash
url = ../xxHash
[submodule "libdeflate"]
path = libdeflate
url = ../libdeflate
[submodule "abseil"]
path = abseil
url = ../abseil-cpp
[submodule "scylla-jmx"]
path = tools/jmx
url = ../scylla-jmx
[submodule "scylla-tools"]
path = tools/java
url = ../scylla-tools-java
[submodule "scylla-python3"]
path = tools/python3
url = ../scylla-python3

View File

@@ -1,755 +1,142 @@
cmake_minimum_required(VERSION 3.18)
##
## For best results, first compile the project using the Ninja build-system.
##
cmake_minimum_required(VERSION 3.7)
project(scylla)
if(NOT CMAKE_BUILD_TYPE AND NOT CMAKE_CONFIGURATION_TYPES)
message(STATUS "Setting build type to 'Release' as none was specified.")
set(CMAKE_BUILD_TYPE "Release" CACHE
STRING "Choose the type of build." FORCE)
# Set the possible values of build type for cmake-gui
set_property(CACHE CMAKE_BUILD_TYPE PROPERTY STRINGS
"Debug" "Release" "Dev" "Sanitize")
if (NOT DEFINED FOR_IDE AND NOT DEFINED ENV{FOR_IDE} AND NOT DEFINED ENV{CLION_IDE})
message(FATAL_ERROR "This CMakeLists.txt file is only valid for use in IDEs, please define FOR_IDE to acknowledge this.")
endif()
if(CMAKE_BUILD_TYPE)
string(TOLOWER "${CMAKE_BUILD_TYPE}" BUILD_TYPE)
else()
set(BUILD_TYPE "release")
endif()
# Default value. A more accurate list is populated through `pkg-config` below if `seastar.pc` is available.
set(SEASTAR_INCLUDE_DIRS "seastar")
function(default_target_arch arch)
set(x86_instruction_sets i386 i686 x86_64)
if(CMAKE_SYSTEM_PROCESSOR IN_LIST x86_instruction_sets)
set(${arch} "westmere" PARENT_SCOPE)
elseif(CMAKE_SYSTEM_PROCESSOR EQUAL "aarch64")
set(${arch} "armv8-a+crc+crypto" PARENT_SCOPE)
# These paths are always available, since they're included in the repository. Additional DPDK headers are placed while
# Seastar is built, and are captured in `SEASTAR_INCLUDE_DIRS` through parsing the Seastar pkg-config file (below).
set(SEASTAR_DPDK_INCLUDE_DIRS
seastar/dpdk/lib/librte_eal/common/include
seastar/dpdk/lib/librte_eal/common/include/generic
seastar/dpdk/lib/librte_eal/common/include/x86
seastar/dpdk/lib/librte_ether)
find_package(PkgConfig REQUIRED)
set(ENV{PKG_CONFIG_PATH} "${CMAKE_SOURCE_DIR}/seastar/build/release:$ENV{PKG_CONFIG_PATH}")
pkg_check_modules(SEASTAR seastar)
find_package(Boost COMPONENTS filesystem program_options system thread)
##
## Populate the names of all source and header files in the indicated paths in a designated variable.
##
## When RECURSIVE is specified, directories are traversed recursively.
##
## Use: scan_scylla_source_directories(VAR my_result_var [RECURSIVE] PATHS [path1 path2 ...])
##
function (scan_scylla_source_directories)
set(options RECURSIVE)
set(oneValueArgs VAR)
set(multiValueArgs PATHS)
cmake_parse_arguments(args "${options}" "${oneValueArgs}" "${multiValueArgs}" "${ARGN}")
set(globs "")
foreach (dir ${args_PATHS})
list(APPEND globs "${dir}/*.cc" "${dir}/*.hh")
endforeach()
if (args_RECURSIVE)
set(glob_kind GLOB_RECURSE)
else()
set(${arch} "" PARENT_SCOPE)
set(glob_kind GLOB)
endif()
endfunction()
default_target_arch(target_arch)
if(target_arch)
set(target_arch_flag "-march=${target_arch}")
endif()
# Configure Seastar compile options to align with Scylla
set(Seastar_CXX_FLAGS -fcoroutines ${target_arch_flag} CACHE INTERNAL "" FORCE)
set(Seastar_CXX_DIALECT gnu++20 CACHE INTERNAL "" FORCE)
file(${glob_kind} var
${globs})
add_subdirectory(seastar)
add_subdirectory(abseil)
# Exclude absl::strerror from the default "all" target since it's not
# used in Scylla build and, moreover, makes use of deprecated glibc APIs,
# such as sys_nerr, which are not exposed from "stdio.h" since glibc 2.32,
# which happens to be the case for recent Fedora distribution versions.
#
# Need to use the internal "absl_strerror" target name instead of namespaced
# variant because `set_target_properties` does not understand the latter form,
# unfortunately.
set_target_properties(absl_strerror PROPERTIES EXCLUDE_FROM_ALL TRUE)
# System libraries dependencies
find_package(Boost COMPONENTS filesystem program_options system thread regex REQUIRED)
find_package(Lua REQUIRED)
find_package(ZLIB REQUIRED)
find_package(ICU COMPONENTS uc REQUIRED)
set(scylla_build_dir "${CMAKE_BINARY_DIR}/build/${BUILD_TYPE}")
set(scylla_gen_build_dir "${scylla_build_dir}/gen")
file(MAKE_DIRECTORY "${scylla_build_dir}" "${scylla_gen_build_dir}")
# Place libraries, executables and archives in ${buildroot}/build/${mode}/
foreach(mode RUNTIME LIBRARY ARCHIVE)
set(CMAKE_${mode}_OUTPUT_DIRECTORY "${scylla_build_dir}")
endforeach()
# Generate C++ source files from thrift definitions
function(scylla_generate_thrift)
set(one_value_args TARGET VAR IN_FILE OUT_DIR SERVICE)
cmake_parse_arguments(args "" "${one_value_args}" "" ${ARGN})
get_filename_component(in_file_name ${args_IN_FILE} NAME_WE)
set(aux_out_file_name ${args_OUT_DIR}/${in_file_name})
set(outputs
${aux_out_file_name}_types.cpp
${aux_out_file_name}_types.h
${aux_out_file_name}_constants.cpp
${aux_out_file_name}_constants.h
${args_OUT_DIR}/${args_SERVICE}.cpp
${args_OUT_DIR}/${args_SERVICE}.h)
add_custom_command(
DEPENDS
${args_IN_FILE}
thrift
OUTPUT ${outputs}
COMMAND ${CMAKE_COMMAND} -E make_directory ${args_OUT_DIR}
COMMAND thrift -gen cpp:cob_style,no_skeleton -out "${args_OUT_DIR}" "${args_IN_FILE}")
add_custom_target(${args_TARGET}
DEPENDS ${outputs})
set(${args_VAR} ${outputs} PARENT_SCOPE)
set(${args_VAR} ${var} PARENT_SCOPE)
endfunction()
scylla_generate_thrift(
TARGET scylla_thrift_gen_cassandra
VAR scylla_thrift_gen_cassandra_files
IN_FILE interface/cassandra.thrift
OUT_DIR ${scylla_gen_build_dir}
SERVICE Cassandra)
## Although Seastar is an external project, it is common enough to explore the sources while doing
## Scylla development that we'll treat the Seastar sources as part of this project for easier navigation.
scan_scylla_source_directories(
VAR SEASTAR_SOURCE_FILES
RECURSIVE
# Parse antlr3 grammar files and generate C++ sources
function(scylla_generate_antlr3)
set(one_value_args TARGET VAR IN_FILE OUT_DIR)
cmake_parse_arguments(args "" "${one_value_args}" "" ${ARGN})
PATHS
seastar/core
seastar/http
seastar/json
seastar/net
seastar/rpc
seastar/tests
seastar/util)
get_filename_component(in_file_pure_name ${args_IN_FILE} NAME)
get_filename_component(stem ${in_file_pure_name} NAME_WE)
scan_scylla_source_directories(
VAR SCYLLA_ROOT_SOURCE_FILES
PATHS .)
set(outputs
"${args_OUT_DIR}/${stem}Lexer.hpp"
"${args_OUT_DIR}/${stem}Lexer.cpp"
"${args_OUT_DIR}/${stem}Parser.hpp"
"${args_OUT_DIR}/${stem}Parser.cpp")
scan_scylla_source_directories(
VAR SCYLLA_SUB_SOURCE_FILES
RECURSIVE
add_custom_command(
DEPENDS
${args_IN_FILE}
OUTPUT ${outputs}
# Remove #ifdef'ed code from the grammar source code
COMMAND sed -e "/^#if 0/,/^#endif/d" "${args_IN_FILE}" > "${args_OUT_DIR}/${in_file_pure_name}"
COMMAND antlr3 "${args_OUT_DIR}/${in_file_pure_name}"
# We replace many local `ExceptionBaseType* ex` variables with a single function-scope one.
# Because we add such a variable to every function, and because `ExceptionBaseType` is not a global
# name, we also add a global typedef to avoid compilation errors.
COMMAND sed -i -e "/^.*On :.*$/d" "${args_OUT_DIR}/${stem}Lexer.hpp"
COMMAND sed -i -e "/^.*On :.*$/d" "${args_OUT_DIR}/${stem}Lexer.cpp"
COMMAND sed -i -e "/^.*On :.*$/d" "${args_OUT_DIR}/${stem}Parser.hpp"
COMMAND sed -i
-e "s/^\\( *\\)\\(ImplTraits::CommonTokenType\\* [a-zA-Z0-9_]* = NULL;\\)$/\\1const \\2/"
-e "/^.*On :.*$/d"
-e "1i using ExceptionBaseType = int;"
-e "s/^{/{ ExceptionBaseType\\* ex = nullptr;/; s/ExceptionBaseType\\* ex = new/ex = new/; s/exceptions::syntax_exception e/exceptions::syntax_exception\\& e/"
"${args_OUT_DIR}/${stem}Parser.cpp"
VERBATIM)
PATHS
api
auth
cql3
db
dht
exceptions
gms
index
io
locator
message
repair
service
sstables
streaming
tests
thrift
tracing
transport
utils)
add_custom_target(${args_TARGET}
DEPENDS ${outputs})
scan_scylla_source_directories(
VAR SCYLLA_GEN_SOURCE_FILES
RECURSIVE
PATHS build/release/gen)
set(${args_VAR} ${outputs} PARENT_SCOPE)
endfunction()
set(antlr3_grammar_files
cql3/Cql.g
alternator/expressions.g)
set(antlr3_gen_files)
foreach(f ${antlr3_grammar_files})
get_filename_component(grammar_file_name "${f}" NAME_WE)
get_filename_component(f_dir "${f}" DIRECTORY)
scylla_generate_antlr3(
TARGET scylla_antlr3_gen_${grammar_file_name}
VAR scylla_antlr3_gen_${grammar_file_name}_files
IN_FILE ${f}
OUT_DIR ${scylla_gen_build_dir}/${f_dir})
list(APPEND antlr3_gen_files "${scylla_antlr3_gen_${grammar_file_name}_files}")
endforeach()
# Generate C++ sources from ragel grammar files
seastar_generate_ragel(
TARGET scylla_ragel_gen_protocol_parser
VAR scylla_ragel_gen_protocol_parser_file
IN_FILE redis/protocol_parser.rl
OUT_FILE ${scylla_gen_build_dir}/redis/protocol_parser.hh)
# Generate C++ sources from Swagger definitions
set(swagger_files
api/api-doc/cache_service.json
api/api-doc/collectd.json
api/api-doc/column_family.json
api/api-doc/commitlog.json
api/api-doc/compaction_manager.json
api/api-doc/config.json
api/api-doc/endpoint_snitch_info.json
api/api-doc/error_injection.json
api/api-doc/failure_detector.json
api/api-doc/gossiper.json
api/api-doc/hinted_handoff.json
api/api-doc/lsa.json
api/api-doc/messaging_service.json
api/api-doc/storage_proxy.json
api/api-doc/storage_service.json
api/api-doc/stream_manager.json
api/api-doc/system.json
api/api-doc/utils.json)
set(swagger_gen_files)
foreach(f ${swagger_files})
get_filename_component(fname "${f}" NAME_WE)
get_filename_component(dir "${f}" DIRECTORY)
seastar_generate_swagger(
TARGET scylla_swagger_gen_${fname}
VAR scylla_swagger_gen_${fname}_files
IN_FILE "${f}"
OUT_DIR "${scylla_gen_build_dir}/${dir}")
list(APPEND swagger_gen_files "${scylla_swagger_gen_${fname}_files}")
endforeach()
# Create C++ bindings for IDL serializers
function(scylla_generate_idl_serializer)
set(one_value_args TARGET VAR IN_FILE OUT_FILE)
cmake_parse_arguments(args "" "${one_value_args}" "" ${ARGN})
get_filename_component(out_dir ${args_OUT_FILE} DIRECTORY)
set(idl_compiler "${CMAKE_SOURCE_DIR}/idl-compiler.py")
find_package(Python3 COMPONENTS Interpreter)
add_custom_command(
DEPENDS
${args_IN_FILE}
${idl_compiler}
OUTPUT ${args_OUT_FILE}
COMMAND ${CMAKE_COMMAND} -E make_directory ${out_dir}
COMMAND Python3::Interpreter ${idl_compiler} --ns ser -f ${args_IN_FILE} -o ${args_OUT_FILE})
add_custom_target(${args_TARGET}
DEPENDS ${args_OUT_FILE})
set(${args_VAR} ${args_OUT_FILE} PARENT_SCOPE)
endfunction()
set(idl_serializers
idl/cache_temperature.idl.hh
idl/commitlog.idl.hh
idl/consistency_level.idl.hh
idl/frozen_mutation.idl.hh
idl/frozen_schema.idl.hh
idl/gossip_digest.idl.hh
idl/idl_test.idl.hh
idl/keys.idl.hh
idl/messaging_service.idl.hh
idl/mutation.idl.hh
idl/paging_state.idl.hh
idl/partition_checksum.idl.hh
idl/paxos.idl.hh
idl/query.idl.hh
idl/range.idl.hh
idl/read_command.idl.hh
idl/reconcilable_result.idl.hh
idl/replay_position.idl.hh
idl/result.idl.hh
idl/ring_position.idl.hh
idl/streaming.idl.hh
idl/token.idl.hh
idl/tracing.idl.hh
idl/truncation_record.idl.hh
idl/uuid.idl.hh
idl/view.idl.hh)
set(idl_gen_files)
foreach(f ${idl_serializers})
get_filename_component(idl_name "${f}" NAME)
get_filename_component(idl_target "${idl_name}" NAME_WE)
get_filename_component(idl_dir "${f}" DIRECTORY)
string(REPLACE ".idl.hh" ".dist.hh" idl_out_hdr_name "${idl_name}")
scylla_generate_idl_serializer(
TARGET scylla_idl_gen_${idl_target}
VAR scylla_idl_gen_${idl_target}_files
IN_FILE ${f}
OUT_FILE ${scylla_gen_build_dir}/${idl_dir}/${idl_out_hdr_name})
list(APPEND idl_gen_files "${scylla_idl_gen_${idl_target}_files}")
endforeach()
set(scylla_sources
absl-flat_hash_map.cc
alternator/auth.cc
alternator/base64.cc
alternator/conditions.cc
alternator/executor.cc
alternator/expressions.cc
alternator/serialization.cc
alternator/server.cc
alternator/stats.cc
alternator/streams.cc
api/api.cc
api/cache_service.cc
api/collectd.cc
api/column_family.cc
api/commitlog.cc
api/compaction_manager.cc
api/config.cc
api/endpoint_snitch.cc
api/error_injection.cc
api/failure_detector.cc
api/gossiper.cc
api/hinted_handoff.cc
api/lsa.cc
api/messaging_service.cc
api/storage_proxy.cc
api/storage_service.cc
api/stream_manager.cc
api/system.cc
atomic_cell.cc
auth/allow_all_authenticator.cc
auth/allow_all_authorizer.cc
auth/authenticated_user.cc
auth/authentication_options.cc
auth/authenticator.cc
auth/common.cc
auth/default_authorizer.cc
auth/password_authenticator.cc
auth/passwords.cc
auth/permission.cc
auth/permissions_cache.cc
auth/resource.cc
auth/role_or_anonymous.cc
auth/roles-metadata.cc
auth/sasl_challenge.cc
auth/service.cc
auth/standard_role_manager.cc
auth/transitional.cc
bytes.cc
canonical_mutation.cc
cdc/cdc_partitioner.cc
cdc/generation.cc
cdc/log.cc
cdc/metadata.cc
cdc/split.cc
clocks-impl.cc
collection_mutation.cc
compress.cc
connection_notifier.cc
converting_mutation_partition_applier.cc
counters.cc
cql3/abstract_marker.cc
cql3/attributes.cc
cql3/cf_name.cc
cql3/column_condition.cc
cql3/column_identifier.cc
cql3/column_specification.cc
cql3/constants.cc
cql3/cql3_type.cc
cql3/expr/expression.cc
cql3/functions/aggregate_fcts.cc
cql3/functions/castas_fcts.cc
cql3/functions/error_injection_fcts.cc
cql3/functions/functions.cc
cql3/functions/user_function.cc
cql3/index_name.cc
cql3/keyspace_element_name.cc
cql3/lists.cc
cql3/maps.cc
cql3/operation.cc
cql3/query_options.cc
cql3/query_processor.cc
cql3/relation.cc
cql3/restrictions/statement_restrictions.cc
cql3/result_set.cc
cql3/role_name.cc
cql3/selection/abstract_function_selector.cc
cql3/selection/selectable.cc
cql3/selection/selection.cc
cql3/selection/selector.cc
cql3/selection/selector_factories.cc
cql3/selection/simple_selector.cc
cql3/sets.cc
cql3/single_column_relation.cc
cql3/statements/alter_keyspace_statement.cc
cql3/statements/alter_table_statement.cc
cql3/statements/alter_type_statement.cc
cql3/statements/alter_view_statement.cc
cql3/statements/authentication_statement.cc
cql3/statements/authorization_statement.cc
cql3/statements/batch_statement.cc
cql3/statements/cas_request.cc
cql3/statements/cf_prop_defs.cc
cql3/statements/cf_statement.cc
cql3/statements/create_function_statement.cc
cql3/statements/create_index_statement.cc
cql3/statements/create_keyspace_statement.cc
cql3/statements/create_table_statement.cc
cql3/statements/create_type_statement.cc
cql3/statements/create_view_statement.cc
cql3/statements/delete_statement.cc
cql3/statements/drop_function_statement.cc
cql3/statements/drop_index_statement.cc
cql3/statements/drop_keyspace_statement.cc
cql3/statements/drop_table_statement.cc
cql3/statements/drop_type_statement.cc
cql3/statements/drop_view_statement.cc
cql3/statements/function_statement.cc
cql3/statements/grant_statement.cc
cql3/statements/index_prop_defs.cc
cql3/statements/index_target.cc
cql3/statements/ks_prop_defs.cc
cql3/statements/list_permissions_statement.cc
cql3/statements/list_users_statement.cc
cql3/statements/modification_statement.cc
cql3/statements/permission_altering_statement.cc
cql3/statements/property_definitions.cc
cql3/statements/raw/parsed_statement.cc
cql3/statements/revoke_statement.cc
cql3/statements/role-management-statements.cc
cql3/statements/schema_altering_statement.cc
cql3/statements/select_statement.cc
cql3/statements/truncate_statement.cc
cql3/statements/update_statement.cc
cql3/statements/use_statement.cc
cql3/token_relation.cc
cql3/tuples.cc
cql3/type_json.cc
cql3/untyped_result_set.cc
cql3/update_parameters.cc
cql3/user_types.cc
cql3/ut_name.cc
cql3/util.cc
cql3/values.cc
cql3/variable_specifications.cc
data/cell.cc
database.cc
db/batchlog_manager.cc
db/commitlog/commitlog.cc
db/commitlog/commitlog_entry.cc
db/commitlog/commitlog_replayer.cc
db/config.cc
db/consistency_level.cc
db/cql_type_parser.cc
db/data_listeners.cc
db/extensions.cc
db/heat_load_balance.cc
db/hints/manager.cc
db/hints/resource_manager.cc
db/large_data_handler.cc
db/legacy_schema_migrator.cc
db/marshal/type_parser.cc
db/schema_tables.cc
db/size_estimates_virtual_reader.cc
db/snapshot-ctl.cc
db/sstables-format-selector.cc
db/system_distributed_keyspace.cc
db/system_keyspace.cc
db/view/row_locking.cc
db/view/view.cc
db/view/view_update_generator.cc
dht/boot_strapper.cc
dht/i_partitioner.cc
dht/murmur3_partitioner.cc
dht/range_streamer.cc
dht/token.cc
distributed_loader.cc
duration.cc
exceptions/exceptions.cc
flat_mutation_reader.cc
frozen_mutation.cc
frozen_schema.cc
gms/application_state.cc
gms/endpoint_state.cc
gms/failure_detector.cc
gms/feature_service.cc
gms/gossip_digest_ack.cc
gms/gossip_digest_ack2.cc
gms/gossip_digest_syn.cc
gms/gossiper.cc
gms/inet_address.cc
gms/version_generator.cc
gms/versioned_value.cc
hashers.cc
index/secondary_index.cc
index/secondary_index_manager.cc
init.cc
keys.cc
lister.cc
locator/abstract_replication_strategy.cc
locator/ec2_multi_region_snitch.cc
locator/ec2_snitch.cc
locator/everywhere_replication_strategy.cc
locator/gce_snitch.cc
locator/gossiping_property_file_snitch.cc
locator/local_strategy.cc
locator/network_topology_strategy.cc
locator/production_snitch_base.cc
locator/rack_inferring_snitch.cc
locator/simple_snitch.cc
locator/simple_strategy.cc
locator/snitch_base.cc
locator/token_metadata.cc
lua.cc
main.cc
memtable.cc
message/messaging_service.cc
multishard_mutation_query.cc
mutation.cc
raft/fsm.cc
raft/log.cc
raft/progress.cc
raft/raft.cc
raft/server.cc
mutation_fragment.cc
mutation_partition.cc
mutation_partition_serializer.cc
mutation_partition_view.cc
mutation_query.cc
mutation_reader.cc
mutation_writer/multishard_writer.cc
mutation_writer/shard_based_splitting_writer.cc
mutation_writer/timestamp_based_splitting_writer.cc
partition_slice_builder.cc
partition_version.cc
querier.cc
query-result-set.cc
query.cc
range_tombstone.cc
range_tombstone_list.cc
reader_concurrency_semaphore.cc
redis/abstract_command.cc
redis/command_factory.cc
redis/commands.cc
redis/keyspace_utils.cc
redis/lolwut.cc
redis/mutation_utils.cc
redis/options.cc
redis/query_processor.cc
redis/query_utils.cc
redis/server.cc
redis/service.cc
redis/stats.cc
repair/repair.cc
repair/row_level.cc
row_cache.cc
schema.cc
schema_mutations.cc
schema_registry.cc
service/client_state.cc
service/migration_manager.cc
service/migration_task.cc
service/misc_services.cc
service/pager/paging_state.cc
service/pager/query_pagers.cc
service/paxos/paxos_state.cc
service/paxos/prepare_response.cc
service/paxos/prepare_summary.cc
service/paxos/proposal.cc
service/priority_manager.cc
service/storage_proxy.cc
service/storage_service.cc
sstables/compaction.cc
sstables/compaction_manager.cc
sstables/compaction_strategy.cc
sstables/compress.cc
sstables/integrity_checked_file_impl.cc
sstables/kl/writer.cc
sstables/leveled_compaction_strategy.cc
sstables/m_format_read_helpers.cc
sstables/metadata_collector.cc
sstables/mp_row_consumer.cc
sstables/mx/writer.cc
sstables/partition.cc
sstables/prepended_input_stream.cc
sstables/random_access_reader.cc
sstables/size_tiered_compaction_strategy.cc
sstables/sstable_directory.cc
sstables/sstable_version.cc
sstables/sstables.cc
sstables/sstables_manager.cc
sstables/time_window_compaction_strategy.cc
sstables/writer.cc
streaming/progress_info.cc
streaming/session_info.cc
streaming/stream_coordinator.cc
streaming/stream_manager.cc
streaming/stream_plan.cc
streaming/stream_reason.cc
streaming/stream_receive_task.cc
streaming/stream_request.cc
streaming/stream_result_future.cc
streaming/stream_session.cc
streaming/stream_session_state.cc
streaming/stream_summary.cc
streaming/stream_task.cc
streaming/stream_transfer_task.cc
table.cc
table_helper.cc
thrift/controller.cc
thrift/handler.cc
thrift/server.cc
thrift/thrift_validation.cc
timeout_config.cc
tracing/trace_keyspace_helper.cc
tracing/trace_state.cc
tracing/traced_file.cc
tracing/tracing.cc
tracing/tracing_backend_registry.cc
transport/controller.cc
transport/cql_protocol_extension.cc
transport/event.cc
transport/event_notifier.cc
transport/messages/result_message.cc
transport/server.cc
types.cc
unimplemented.cc
utils/UUID_gen.cc
utils/arch/powerpc/crc32-vpmsum/crc32_wrapper.cc
utils/array-search.cc
utils/ascii.cc
utils/big_decimal.cc
utils/bloom_calculations.cc
utils/bloom_filter.cc
utils/buffer_input_stream.cc
utils/build_id.cc
utils/config_file.cc
utils/directories.cc
utils/disk-error-handler.cc
utils/dynamic_bitset.cc
utils/error_injection.cc
utils/exceptions.cc
utils/file_lock.cc
utils/generation-number.cc
utils/gz/crc_combine.cc
utils/human_readable.cc
utils/i_filter.cc
utils/large_bitset.cc
utils/like_matcher.cc
utils/limiting_data_source.cc
utils/logalloc.cc
utils/managed_bytes.cc
utils/multiprecision_int.cc
utils/murmur_hash.cc
utils/rate_limiter.cc
utils/rjson.cc
utils/runtime.cc
utils/updateable_value.cc
utils/utf8.cc
utils/uuid.cc
validation.cc
vint-serialization.cc
zstd.cc
release.cc)
set(scylla_gen_sources
"${scylla_thrift_gen_cassandra_files}"
"${scylla_ragel_gen_protocol_parser_file}"
"${swagger_gen_files}"
"${idl_gen_files}"
"${antlr3_gen_files}")
set(SCYLLA_SOURCE_FILES
${SCYLLA_ROOT_SOURCE_FILES}
${SCYLLA_GEN_SOURCE_FILES}
${SCYLLA_SUB_SOURCE_FILES})
add_executable(scylla
${scylla_sources}
${scylla_gen_sources})
${SEASTAR_SOURCE_FILES}
${SCYLLA_SOURCE_FILES})
target_link_libraries(scylla PRIVATE
seastar
# Boost dependencies
Boost::filesystem
Boost::program_options
Boost::system
Boost::thread
Boost::regex
Boost::headers
# Abseil libs
absl::hashtablez_sampler
absl::raw_hash_set
absl::synchronization
absl::graphcycles_internal
absl::stacktrace
absl::symbolize
absl::debugging_internal
absl::demangle_internal
absl::time
absl::time_zone
absl::int128
absl::city
absl::hash
absl::malloc_internal
absl::spinlock_wait
absl::base
absl::dynamic_annotations
absl::raw_logging_internal
absl::exponential_biased
absl::throw_delegate
# System libs
ZLIB::ZLIB
ICU::uc
systemd
zstd
snappy
${LUA_LIBRARIES}
thrift
crypt)
# Note that since CLion does not undestand GCC6 concepts, we always disable them (even if users configure otherwise).
# CLion seems to have trouble with `-U` (macro undefinition), so we do it this way instead.
list(REMOVE_ITEM SEASTAR_CFLAGS "-DHAVE_GCC6_CONCEPTS")
target_link_libraries(scylla PRIVATE
-Wl,--build-id=sha1 # Force SHA1 build-id generation
# TODO: Use lld linker if it's available, otherwise gold, else bfd
-fuse-ld=lld)
# TODO: patch dynamic linker to match configure.py behavior
# If the Seastar pkg-config information is available, append to the default flags.
#
# For ease of browsing the source code, we always pretend that DPDK is enabled.
target_compile_options(scylla PUBLIC
-std=gnu++1z
-DHAVE_DPDK
-DHAVE_HWLOC
"${SEASTAR_CFLAGS}")
target_compile_options(scylla PRIVATE
-std=gnu++20
-fcoroutines # TODO: Clang does not have this flag, adjust to both variants
${target_arch_flag})
# Hacks needed to expose internal APIs for xxhash dependencies
target_compile_definitions(scylla PRIVATE XXH_PRIVATE_API HAVE_LZ4_COMPRESS_DEFAULT)
target_include_directories(scylla PRIVATE
"${CMAKE_CURRENT_SOURCE_DIR}"
libdeflate
abseil
"${scylla_gen_build_dir}")
###
### Create crc_combine_table helper executable.
### Use it to generate crc_combine_table.cc to be used in scylla at build time.
###
add_executable(crc_combine_table utils/gz/gen_crc_combine_table.cc)
target_link_libraries(crc_combine_table PRIVATE seastar)
target_include_directories(crc_combine_table PRIVATE "${CMAKE_CURRENT_SOURCE_DIR}")
target_compile_options(crc_combine_table PRIVATE
-std=gnu++20
-fcoroutines
${target_arch_flag})
add_dependencies(scylla crc_combine_table)
# Generate an additional source file at build time that is needed for Scylla compilation
add_custom_command(OUTPUT "${scylla_gen_build_dir}/utils/gz/crc_combine_table.cc"
COMMAND $<TARGET_FILE:crc_combine_table> > "${scylla_gen_build_dir}/utils/gz/crc_combine_table.cc"
DEPENDS crc_combine_table)
target_sources(scylla PRIVATE "${scylla_gen_build_dir}/utils/gz/crc_combine_table.cc")
###
### Generate version file and supply appropriate compile definitions for release.cc
###
execute_process(COMMAND ${CMAKE_SOURCE_DIR}/SCYLLA-VERSION-GEN RESULT_VARIABLE scylla_version_gen_res)
if(scylla_version_gen_res)
message(SEND_ERROR "Version file generation failed. Return code: ${scylla_version_gen_res}")
endif()
file(READ build/SCYLLA-VERSION-FILE scylla_version)
string(STRIP "${scylla_version}" scylla_version)
file(READ build/SCYLLA-RELEASE-FILE scylla_release)
string(STRIP "${scylla_release}" scylla_release)
get_property(release_cdefs SOURCE "${CMAKE_SOURCE_DIR}/release.cc" PROPERTY COMPILE_DEFINITIONS)
list(APPEND release_cdefs "SCYLLA_VERSION=\"${scylla_version}\"" "SCYLLA_RELEASE=\"${scylla_release}\"")
set_source_files_properties("${CMAKE_SOURCE_DIR}/release.cc" PROPERTIES COMPILE_DEFINITIONS "${release_cdefs}")
###
### Custom command for building libdeflate. Link the library to scylla.
###
set(libdeflate_lib "${scylla_build_dir}/libdeflate/libdeflate.a")
add_custom_command(OUTPUT "${libdeflate_lib}"
COMMAND make -C libdeflate
BUILD_DIR=../build/${BUILD_TYPE}/libdeflate/
CC=${CMAKE_C_COMPILER}
"CFLAGS=${target_arch_flag}"
../build/${BUILD_TYPE}/libdeflate//libdeflate.a) # Two backslashes are important!
# Hack to force generating custom command to produce libdeflate.a
add_custom_target(libdeflate DEPENDS "${libdeflate_lib}")
target_link_libraries(scylla PRIVATE "${libdeflate_lib}")
# TODO: create cmake/ directory and move utilities (generate functions etc) there
# TODO: Build tests if BUILD_TESTING=on (using CTest module)
# The order matters here: prefer the "static" DPDK directories to any dynamic paths from pkg-config. Some files are only
# available dynamically, though.
target_include_directories(scylla PUBLIC
.
${SEASTAR_DPDK_INCLUDE_DIRS}
${SEASTAR_INCLUDE_DIRS}
${Boost_INCLUDE_DIRS}
xxhash
libdeflate
build/release/gen)

View File

@@ -1,6 +1,6 @@
# Asking questions or requesting help
Use the [ScyllaDB user mailing list](https://groups.google.com/forum/#!forum/scylladb-users) or the [Slack workspace](http://slack.scylladb.com) for general questions and help.
Use the [ScyllaDB user mailing list](https://groups.google.com/forum/#!forum/scylladb-users) for general questions and help.
# Reporting an issue
@@ -8,4 +8,4 @@ Please use the [Issue Tracker](https://github.com/scylladb/scylla/issues/) to re
# Contributing Code to Scylla
To contribute code to Scylla, you need to sign the [Contributor License Agreement](https://www.scylladb.com/open-source/contributor-agreement/) and send your changes as [patches](https://github.com/scylladb/scylla/wiki/Formatting-and-sending-patches) to the [mailing list](https://groups.google.com/forum/#!forum/scylladb-dev). We don't accept pull requests on GitHub.
To contribute code to Scylla, you need to sign the [Contributor License Agreement](http://www.scylladb.com/opensource/cla/) and send your changes as [patches](https://github.com/scylladb/scylla/wiki/Formatting-and-sending-patches) to the [mailing list](https://groups.google.com/forum/#!forum/scylladb-dev). We don't accept pull requests on GitHub.

View File

@@ -18,36 +18,13 @@ $ git submodule update --init --recursive
### Dependencies
Scylla is fairly fussy about its build environment, requiring a very recent
version of the C++20 compiler and numerous tools and libraries to build.
Scylla depends on the system package manager for its development dependencies.
Run `./install-dependencies.sh` (as root) to use your Linux distributions's
package manager to install the appropriate packages on your build machine.
However, this will only work on very recent distributions. For example,
currently Fedora users must upgrade to Fedora 32 otherwise the C++ compiler
will be too old, and not support the new C++20 standard that Scylla uses.
Alternatively, to avoid having to upgrade your build machine or install
various packages on it, we provide another option - the **frozen toolchain**.
This is a script, `./tools/toolchain/dbuild`, that can execute build or run
commands inside a Docker image that contains exactly the right build tools and
libraries. The `dbuild` technique is useful for beginners, but is also the way
in which ScyllaDB produces official releases, so it is highly recommended.
To use `dbuild`, you simply prefix any build or run command with it. Building
and running Scylla becomes as easy as:
```bash
$ ./tools/toolchain/dbuild ./configure.py
$ ./tools/toolchain/dbuild ninja build/release/scylla
$ ./tools/toolchain/dbuild ./build/release/scylla --developer-mode 1
```
Running `./install-dependencies.sh` (as root) installs the appropriate packages based on your Linux distribution.
### Build system
**Note**: Compiling Scylla requires, conservatively, 2 GB of memory per native
thread, and up to 3 GB per native thread while linking. GCC >= 10 is
required.
**Note**: Compiling Scylla requires, conservatively, 2 GB of memory per native thread, and up to 3 GB per native thread while linking.
Scylla is built with [Ninja](https://ninja-build.org/), a low-level rule-based system. A Python script, `configure.py`, generates a Ninja file (`build.ninja`) based on configuration options.
@@ -66,9 +43,11 @@ The full suite of options for project configuration is available via
$ ./configure.py --help
```
The most important option is:
The most important options are:
- `--enable-dpdk`: [DPDK](http://dpdk.org/) is a set of libraries and drivers for fast packet processing. During development, it's not necessary to enable support even if it is supported by your platform.
- `--mode={release,debug,all}`: Debug mode enables [AddressSanitizer](https://github.com/google/sanitizers/wiki/AddressSanitizer) and allows for debugging with tools like GDB. Debugging builds are generally slower and generate much larger object files than release builds.
- `--{enable,disable}-dpdk`: [DPDK](http://dpdk.org/) is a set of libraries and drivers for fast packet processing. During development, it's not necessary to enable support even if it is supported by your platform.
Source files and build targets are tracked manually in `configure.py`, so the script needs to be updated when new files or targets are added or removed.
@@ -76,30 +55,6 @@ To save time -- for instance, to avoid compiling all unit tests -- you can also
```bash
$ ninja-build build/release/tests/schema_change_test
$ ninja-build build/release/service/storage_proxy.o
```
You can also specify a single mode. For example
```bash
$ ninja-build release
```
Will build everytihng in release mode. The valid modes are
* Debug: Enables [AddressSanitizer](https://github.com/google/sanitizers/wiki/AddressSanitizer)
and other sanity checks. It has no optimizations, which allows for debugging with tools like
GDB. Debugging builds are generally slower and generate much larger object files than release builds.
* Release: Fewer checks and more optimizations. It still has debug info.
* Dev: No optimizations or debug info. The objective is to compile and link as fast as possible.
This is useful for the first iterations of a patch.
Note that by default unit tests binaries are stripped so they can't be used with gdb or seastar-addr2line.
To include debug information in the unit test binary, build the test binary with a `_g` suffix. For example,
```bash
$ ninja-build build/release/tests/schema_change_test_g
```
### Unit testing
@@ -128,7 +83,7 @@ The `-c1 -m1G` arguments limit this Seastar-based test to a single system thread
### Preparing patches
All changes to Scylla are submitted as patches to the public [mailing list](mailto:scylladb-dev@googlegroups.com). Once a patch is approved by one of the maintainers of the project, it is committed to the maintainers' copy of the repository at https://github.com/scylladb/scylla.
All changes to Scylla are submitted as patches to the public mailing list. Once a patch is approved by one of the maintainers of the project, it is committed to the maintainers' copy of the repository at https://github.com/scylladb/scylla.
Detailed instructions for formatting patches for the mailing list and advice on preparing good patches are available at the [ScyllaDB website](http://docs.scylladb.com/contribute/). There are also some guidelines that can help you make the patch review process smoother:
@@ -153,12 +108,10 @@ In v3:
"Tests: unit ({mode}), dtest ({smp})"
```
The usual is "Tests: unit (dev)", although running debug tests is encouraged.
The usual is "Tests: unit (release)", although running debug tests is encouraged.
5. When answering review comments, prefer inline quotes as they make it easier to track the conversation across multiple e-mails.
6. The Linux kernel's [Submitting Patches](https://www.kernel.org/doc/html/v4.19/process/submitting-patches.html) document offers excellent advice on how to prepare patches and patchsets for review. Since the Scylla development process is derived from the kernel's, almost all of the advice there is directly applicable.
### Finding a person to review and merge your patches
You can use the `scripts/find-maintainer` script to find a subsystem maintainer and/or reviewer for your patches. The script accepts a filename in the git source tree as an argument and outputs a list of subsystems the file belongs to and their respective maintainers and reviewers. For example, if you changed the `cql3/statements/create_view_statement.hh` file, run the script as follows:
@@ -211,29 +164,6 @@ On a development machine, one might run Scylla as
$ SCYLLA_HOME=$HOME/scylla build/release/scylla --overprovisioned --developer-mode=yes
```
To interact with scylla it is recommended to build our versions of
cqlsh and nodetool. They are available at
https://github.com/scylladb/scylla-tools-java and can be built with
```bash
$ sudo ./install-dependencies.sh
$ ant jar
```
cqlsh should work out of the box, but nodetool depends on a running
scylla-jmx (https://github.com/scylladb/scylla-jmx). It can be build
with
```bash
$ mvn package
```
and must be started with
```bash
$ ./scripts/scylla-jmx
```
### Branches and tags
Multiple release branches are maintained on the Git repository at https://github.com/scylladb/scylla. Release 1.5, for instance, is tracked on the `branch-1.5` branch.
@@ -324,7 +254,7 @@ In this example, `10.0.0.2` will be sent up to 16 jobs and the local machine wil
When a compilation is in progress, the status of jobs on all remote machines can be visualized in the terminal with `distccmon-text` or graphically as a GTK application with `distccmon-gnome`.
One thing to keep in mind is that linking object files happens on the coordinating machine, which can be a bottleneck. See the next sections speeding up this process.
One thing to keep in mind is that linking object files happens on the coordinating machine, which can be a bottleneck. See the next section speeding up this process.
### Using the `gold` linker
@@ -334,24 +264,6 @@ Linking Scylla can be slow. The gold linker can replace GNU ld and often speeds
$ sudo alternatives --config ld
```
### Using split dwarf
With debug info enabled, most of the link time is spent copying and
relocating it. It is possible to leave most of the debug info out of
the link by writing it to a side .dwo file. This is done by passing
`-gsplit-dwarf` to gcc.
Unfortunately just `-gsplit-dwarf` would slow down `gdb` startup. To
avoid that the gold linker can be told to create an index with
`--gdb-index`.
More info at https://gcc.gnu.org/wiki/DebugFission.
Both options can be enable by passing `--split-dwarf` to configure.py.
Note that distcc is *not* compatible with it, but icecream
(https://github.com/icecc/icecream) is.
### Testing changes in Seastar with Scylla
Sometimes Scylla development is closely tied with a feature being developed in Seastar. It can be useful to compile Scylla with a particular check-out of Seastar.
@@ -365,8 +277,3 @@ $ git remote add local /home/tsmith/src/seastar
$ git remote update
$ git checkout -t local/my_local_seastar_branch
```
### Core dump debugging
Slides:
2018.11.20: https://www.slideshare.net/tomekgrabiec/scylla-core-dump-debugging-tools

View File

131
MAINTAINERS Normal file
View File

@@ -0,0 +1,131 @@
M: Maintainer with commit access
R: Reviewer with subsystem expertise
F: Filename, directory, or pattern for the subsystem
---
AUTH
M: Paweł Dziepak <pdziepak@scylladb.com>
M: Duarte Nunes <duarte@scylladb.com>
R: Calle Wilund <calle@scylladb.com>
R: Vlad Zolotarov <vladz@scylladb.com>
R: Jesse Haber-Kucharsky <jhaberku@scylladb.com>
F: auth/*
CACHE
M: Tomasz Grabiec <tgrabiec@scylladb.com>
M: Paweł Dziepak <pdziepak@scylladb.com>
R: Piotr Jastrzebski <piotr@scylladb.com>
F: row_cache*
F: *mutation*
F: tests/mvcc*
COMMITLOG / BATCHLOGa
M: Paweł Dziepak <pdziepak@scylladb.com>
M: Duarte Nunes <duarte@scylladb.com>
R: Calle Wilund <calle@scylladb.com>
F: db/commitlog/*
F: db/batch*
COORDINATOR
M: Paweł Dziepak <pdziepak@scylladb.com>
M: Duarte Nunes <duarte@scylladb.com>
R: Gleb Natapov <gleb@scylladb.com>
F: service/storage_proxy*
COMPACTION
R: Raphael S. Carvalho <raphaelsc@scylladb.com>
R: Glauber Costa <glauber@scylladb.com>
R: Nadav Har'El <nyh@scylladb.com>
F: sstables/compaction*
CQL TRANSPORT LAYER
M: Pekka Enberg <penberg@scylladb.com>
F: transport/*
CQL QUERY LANGUAGE
M: Tomasz Grabiec <tgrabiec@scylladb.com>
M: Pekka Enberg <penberg@scylladb.com>
F: cql3/*
COUNTERS
M: Paweł Dziepak <pdziepak@scylladb.com>
F: counters*
F: tests/counter_test*
GOSSIP
M: Duarte Nunes <duarte@scylladb.com>
M: Tomasz Grabiec <tgrabiec@scylladb.com>
R: Asias He <asias@scylladb.com>
F: gms/*
DOCKER
M: Pekka Enberg <penberg@scylladb.com>
F: dist/docker/*
LSA
M: Tomasz Grabiec <tgrabiec@scylladb.com>
M: Paweł Dziepak <pdziepak@scylladb.com>
F: utils/logalloc*
MATERIALIZED VIEWS
M: Duarte Nunes <duarte@scylladb.com>
M: Pekka Enberg <penberg@scylladb.com>
R: Nadav Har'El <nyh@scylladb.com>
R: Duarte Nunes <duarte@scylladb.com>
F: db/view/*
F: cql3/statements/*view*
PACKAGING
R: Takuya ASADA <syuu@scylladb.com>
F: dist/*
REPAIR
M: Tomasz Grabiec <tgrabiec@scylladb.com>
M: Duarte Nunes <duarte@scylladb.com>
R: Asias He <asias@scylladb.com>
R: Nadav Har'El <nyh@scylladb.com>
F: repair/*
SCHEMA MANAGEMENT
M: Tomasz Grabiec <tgrabiec@scylladb.com>
M: Duarte Nunes <duarte@scylladb.com>
M: Pekka Enberg <penberg@scylladb.com>
F: db/schema_tables*
F: db/legacy_schema_migrator*
F: service/migration*
F: schema*
SECONDARY INDEXES
M: Pekka Enberg <penberg@scylladb.com>
M: Duarte Nunes <duarte@scylladb.com>
R: Nadav Har'El <nyh@scylladb.com>
R: Pekka Enberg <penberg@scylladb.com>
F: db/index/*
F: cql3/statements/*index*
SSTABLES
M: Tomasz Grabiec <tgrabiec@scylladb.com>
M: Duarte Nunes <duarte@scylladb.com>
R: Raphael S. Carvalho <raphaelsc@scylladb.com>
R: Glauber Costa <glauber@scylladb.com>
R: Nadav Har'El <nyh@scylladb.com>
F: sstables/*
STREAMING
M: Tomasz Grabiec <tgrabiec@scylladb.com>
M: Duarte Nunes <duarte@scylladb.com>
R: Asias He <asias@scylladb.com>
F: streaming/*
F: service/storage_service.*
THRIFT TRANSPORT LAYER
M: Duarte Nunes <duarte@scylladb.com>
F: thrift/*
THE REST
M: Avi Kivity <avi@scylladb.com>
M: Paweł Dziepak <pdziepak@scylladb.com>
M: Duarte Nunes <duarte@scylladb.com>
M: Tomasz Grabiec <tgrabiec@scylladb.com>
F: *

View File

@@ -1,7 +1,5 @@
This project includes code developed by the Apache Software Foundation (http://www.apache.org/),
especially Apache Cassandra.
It includes files from https://github.com/antonblanchard/crc32-vpmsum (author Anton Blanchard <anton@au.ibm.com>, IBM).
It also includes files from https://github.com/antonblanchard/crc32-vpmsum (author Anton Blanchard <anton@au.ibm.com>, IBM).
These files are located in utils/arch/powerpc/crc32-vpmsum. Their license may be found in licenses/LICENSE-crc32-vpmsum.TXT.
It includes modified code from https://gitbox.apache.org/repos/asf?p=cassandra-dtest.git (owned by The Apache Software Foundation)

29
README-DPDK.md Normal file
View File

@@ -0,0 +1,29 @@
Seastar and DPDK
================
Seastar uses the Data Plane Development Kit to drive NIC hardware directly. This
provides an enormous performance boost.
To enable DPDK, specify `--enable-dpdk` to `./configure.py`, and `--dpdk-pmd` as a
run-time parameter. This will use the DPDK package provided as a git submodule with the
seastar sources.
To use your own self-compiled DPDK package, follow this procedure:
1. Setup host to compile DPDK:
- Ubuntu
`sudo apt-get install -y build-essential linux-image-extra-$(uname -r)`
2. Prepare a DPDK SDK:
- Download the latest DPDK release: `wget http://dpdk.org/browse/dpdk/snapshot/dpdk-1.8.0.tar.gz`
- Untar it.
- Edit config/common_linuxapp: set CONFIG_RTE_MBUF_REFCNT and CONFIG_RTE_LIBRTE_KNI to 'n'.
- For DPDK 1.7.x: edit config/common_linuxapp:
- Set CONFIG_RTE_LIBRTE_PMD_BOND to 'n'.
- Set CONFIG_RTE_MBUF_SCATTER_GATHER to 'n'.
- Set CONFIG_RTE_LIBRTE_IP_FRAG to 'n'.
- Start the tools/setup.sh script as root.
- Compile a linuxapp target (option 9).
- Install IGB_UIO module (option 11).
- Bind some physical port to IGB_UIO (option 17).
- Configure hugepage mappings (option 14/15).
3. Run a configure.py: `./configure.py --dpdk-target <Path to untared dpdk-1.8.0 above>/x86_64-native-linuxapp-gcc`.

143
README.md
View File

@@ -1,113 +1,78 @@
# Scylla
[![Slack](https://img.shields.io/badge/slack-scylla-brightgreen.svg?logo=slack)](http://slack.scylladb.com)
[![Twitter](https://img.shields.io/twitter/follow/ScyllaDB.svg?style=social&label=Follow)](https://twitter.com/intent/follow?screen_name=ScyllaDB)
## What is Scylla?
Scylla is the real-time big data database that is API-compatible with Apache Cassandra and Amazon DynamoDB.
Scylla embraces a shared-nothing approach that increases throughput and storage capacity to realize order-of-magnitude performance improvements and reduce hardware costs.
For more information, please see the [ScyllaDB web site].
[ScyllaDB web site]: https://www.scylladb.com
## Build Prerequisites
Scylla is fairly fussy about its build environment, requiring very recent
versions of the C++20 compiler and of many libraries to build. The document
[HACKING.md](HACKING.md) includes detailed information on building and
developing Scylla, but to get Scylla building quickly on (almost) any build
machine, Scylla offers a [frozen toolchain](tools/toolchain/README.md),
This is a pre-configured Docker image which includes recent versions of all
the required compilers, libraries and build tools. Using the frozen toolchain
allows you to avoid changing anything in your build machine to meet Scylla's
requirements - you just need to meet the frozen toolchain's prerequisites
(mostly, Docker or Podman being available).
## Building Scylla
Building Scylla with the frozen toolchain `dbuild` is as easy as:
## Quick-start
```bash
$ git submodule update --init --force --recursive
$ ./tools/toolchain/dbuild ./configure.py
$ ./tools/toolchain/dbuild ninja build/release/scylla
$ git submodule update --init --recursive
$ sudo ./install-dependencies.sh
$ ./configure.py --mode=release
$ ninja-build -j4 # Assuming 4 system threads.
$ ./build/release/scylla
$ # Rejoice!
```
For further information, please see:
* [Developer documentation] for more information on building Scylla.
* [Build documentation] on how to build Scylla binaries, tests, and packages.
* [Docker image build documentation] for information on how to build Docker images.
[developer documentation]: HACKING.md
[build documentation]: docs/building.md
[docker image build documentation]: dist/docker/redhat/README.md
Please see [HACKING.md](HACKING.md) for detailed information on building and developing Scylla.
## Running Scylla
To start Scylla server, run:
* Run Scylla
```
./build/release/scylla
```bash
$ ./tools/toolchain/dbuild ./build/release/scylla --workdir tmp --smp 1 --developer-mode 1
```
This will start a Scylla node with one CPU core allocated to it and data files stored in the `tmp` directory.
The `--developer-mode` is needed to disable the various checks Scylla performs at startup to ensure the machine is configured for maximum performance (not relevant on development workstations).
Please note that you need to run Scylla with `dbuild` if you built it with the frozen toolchain.
* run Scylla with one CPU and ./tmp as data directory
For more run options, run:
```bash
$ ./tools/toolchain/dbuild ./build/release/scylla --help
```
./build/release/scylla --datadir tmp --commitlog-directory tmp --smp 1
```
## Testing
* For more run options:
```
./build/release/scylla --help
```
See [test.py manual](docs/testing.md).
## Building Fedora RPM
## Scylla APIs and compatibility
By default, Scylla is compatible with Apache Cassandra and its APIs - CQL and
Thrift. There is also support for the API of Amazon DynamoDB™,
which needs to be enabled and configured in order to be used. For more
information on how to enable the DynamoDB™ API in Scylla,
and the current compatibility of this feature as well as Scylla-specific extensions, see
[Alternator](docs/alternator/alternator.md) and
[Getting started with Alternator](docs/alternator/getting-started.md).
As a pre-requisite, you need to install [Mock](https://fedoraproject.org/wiki/Mock) on your machine:
## Documentation
```
# Install mock:
sudo yum install mock
Documentation can be found in [./docs](./docs) and on the
[wiki](https://github.com/scylladb/scylla/wiki). There is currently no clear
definition of what goes where, so when looking for something be sure to check
both.
Seastar documentation can be found [here](http://docs.seastar.io/master/index.html).
User documentation can be found [here](https://docs.scylladb.com/).
# Add user to the "mock" group:
usermod -a -G mock $USER && newgrp mock
```
## Training
Then, to build an RPM, run:
Training material and online courses can be found at [Scylla University](https://university.scylladb.com/).
The courses are free, self-paced and include hands-on examples. They cover a variety of topics including Scylla data modeling,
administration, architecture, basic NoSQL concepts, using drivers for application development, Scylla setup, failover, compactions,
multi-datacenters and how Scylla integrates with third-party applications.
```
./dist/redhat/build_rpm.sh
```
The built RPM is stored in the ``build/mock/<configuration>/result`` directory.
For example, on Fedora 21 mock reports the following:
```
INFO: Done(scylla-server-0.00-1.fc21.src.rpm) Config(default) 20 minutes 7 seconds
INFO: Results and/or logs in: build/mock/fedora-21-x86_64/result
```
## Building Fedora-based Docker image
Build a Docker image with:
```
cd dist/docker
docker build -t <image-name> .
```
Run the image with:
```
docker run -p $(hostname -i):9042:9042 -i -t <image name>
```
## Contributing to Scylla
If you want to report a bug or submit a pull request or a patch, please read the [contribution guidelines].
If you are a developer working on Scylla, please read the [developer guidelines].
[contribution guidelines]: CONTRIBUTING.md
[developer guidelines]: HACKING.md
## Contact
* The [users mailing list] and [Slack channel] are for users to discuss configuration, management, and operations of the ScyllaDB open source.
* The [developers mailing list] is for developers and people interested in following the development of ScyllaDB to discuss technical topics.
[Users mailing list]: https://groups.google.com/forum/#!forum/scylladb-users
[Slack channel]: http://slack.scylladb.com/
[Developers mailing list]: https://groups.google.com/forum/#!forum/scylladb-dev
[Guidelines for contributing](CONTRIBUTING.md)

View File

@@ -1,7 +1,6 @@
#!/bin/sh
PRODUCT=scylla
VERSION=4.4.dev
VERSION=3.0.11
if test -f version
then
@@ -19,16 +18,7 @@ else
SCYLLA_RELEASE=$SCYLLA_BUILD.$DATE.$GIT_COMMIT
fi
if [ -f build/SCYLLA-RELEASE-FILE ]; then
RELEASE_FILE=$(cat build/SCYLLA-RELEASE-FILE)
GIT_COMMIT_FILE=$(cat build/SCYLLA-RELEASE-FILE |cut -d . -f 3)
if [ "$GIT_COMMIT" = "$GIT_COMMIT_FILE" ]; then
exit 0
fi
fi
echo "$SCYLLA_VERSION-$SCYLLA_RELEASE"
mkdir -p build
echo "$SCYLLA_VERSION" > build/SCYLLA-VERSION-FILE
echo "$SCYLLA_RELEASE" > build/SCYLLA-RELEASE-FILE
echo "$PRODUCT" > build/SCYLLA-PRODUCT-FILE

1
abseil

Submodule abseil deleted from 1e3d25b265

View File

@@ -1,26 +0,0 @@
/*
* Copyright (C) 2020 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#include "absl-flat_hash_map.hh"
size_t sstring_hash::operator()(std::string_view v) const noexcept {
return absl::Hash<std::string_view>{}(v);
}

View File

@@ -1,47 +0,0 @@
/*
* Copyright (C) 2020 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include <absl/container/flat_hash_map.h>
#include <seastar/core/sstring.hh>
using namespace seastar;
struct sstring_hash {
using is_transparent = void;
size_t operator()(std::string_view v) const noexcept;
};
struct sstring_eq {
using is_transparent = void;
bool operator()(std::string_view a, std::string_view b) const noexcept {
return a == b;
}
};
template <typename K, typename V, typename... Ts>
struct flat_hash_map : public absl::flat_hash_map<K, V, Ts...> {
};
template <typename V>
struct flat_hash_map<sstring, V>
: public absl::flat_hash_map<sstring, V, sstring_hash, sstring_eq> {};

View File

@@ -1,146 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#include "alternator/error.hh"
#include "log.hh"
#include <string>
#include <string_view>
#include <gnutls/crypto.h>
#include <seastar/util/defer.hh>
#include "hashers.hh"
#include "bytes.hh"
#include "alternator/auth.hh"
#include <fmt/format.h>
#include "auth/common.hh"
#include "auth/password_authenticator.hh"
#include "auth/roles-metadata.hh"
#include "cql3/query_processor.hh"
#include "cql3/untyped_result_set.hh"
namespace alternator {
static logging::logger alogger("alternator-auth");
static hmac_sha256_digest hmac_sha256(std::string_view key, std::string_view msg) {
hmac_sha256_digest digest;
int ret = gnutls_hmac_fast(GNUTLS_MAC_SHA256, key.data(), key.size(), msg.data(), msg.size(), digest.data());
if (ret) {
throw std::runtime_error(fmt::format("Computing HMAC failed ({}): {}", ret, gnutls_strerror(ret)));
}
return digest;
}
static hmac_sha256_digest get_signature_key(std::string_view key, std::string_view date_stamp, std::string_view region_name, std::string_view service_name) {
auto date = hmac_sha256("AWS4" + std::string(key), date_stamp);
auto region = hmac_sha256(std::string_view(date.data(), date.size()), region_name);
auto service = hmac_sha256(std::string_view(region.data(), region.size()), service_name);
auto signing = hmac_sha256(std::string_view(service.data(), service.size()), "aws4_request");
return signing;
}
static std::string apply_sha256(std::string_view msg) {
sha256_hasher hasher;
hasher.update(msg.data(), msg.size());
return to_hex(hasher.finalize());
}
static std::string format_time_point(db_clock::time_point tp) {
time_t time_point_repr = db_clock::to_time_t(tp);
std::string time_point_str;
time_point_str.resize(17);
::tm time_buf;
// strftime prints the terminating null character as well
std::strftime(time_point_str.data(), time_point_str.size(), "%Y%m%dT%H%M%SZ", ::gmtime_r(&time_point_repr, &time_buf));
time_point_str.resize(16);
return time_point_str;
}
void check_expiry(std::string_view signature_date) {
//FIXME: The default 15min can be changed with X-Amz-Expires header - we should honor it
std::string expiration_str = format_time_point(db_clock::now() - 15min);
std::string validity_str = format_time_point(db_clock::now() + 15min);
if (signature_date < expiration_str) {
throw api_error::invalid_signature(
fmt::format("Signature expired: {} is now earlier than {} (current time - 15 min.)",
signature_date, expiration_str));
}
if (signature_date > validity_str) {
throw api_error::invalid_signature(
fmt::format("Signature not yet current: {} is still later than {} (current time + 15 min.)",
signature_date, validity_str));
}
}
std::string get_signature(std::string_view access_key_id, std::string_view secret_access_key, std::string_view host, std::string_view method,
std::string_view orig_datestamp, std::string_view signed_headers_str, const std::map<std::string_view, std::string_view>& signed_headers_map,
std::string_view body_content, std::string_view region, std::string_view service, std::string_view query_string) {
auto amz_date_it = signed_headers_map.find("x-amz-date");
if (amz_date_it == signed_headers_map.end()) {
throw api_error::invalid_signature("X-Amz-Date header is mandatory for signature verification");
}
std::string_view amz_date = amz_date_it->second;
check_expiry(amz_date);
std::string_view datestamp = amz_date.substr(0, 8);
if (datestamp != orig_datestamp) {
throw api_error::invalid_signature(
format("X-Amz-Date date does not match the provided datestamp. Expected {}, got {}",
orig_datestamp, datestamp));
}
std::string_view canonical_uri = "/";
std::stringstream canonical_headers;
for (const auto& header : signed_headers_map) {
canonical_headers << fmt::format("{}:{}", header.first, header.second) << '\n';
}
std::string payload_hash = apply_sha256(body_content);
std::string canonical_request = fmt::format("{}\n{}\n{}\n{}\n{}\n{}", method, canonical_uri, query_string, canonical_headers.str(), signed_headers_str, payload_hash);
std::string_view algorithm = "AWS4-HMAC-SHA256";
std::string credential_scope = fmt::format("{}/{}/{}/aws4_request", datestamp, region, service);
std::string string_to_sign = fmt::format("{}\n{}\n{}\n{}", algorithm, amz_date, credential_scope, apply_sha256(canonical_request));
hmac_sha256_digest signing_key = get_signature_key(secret_access_key, datestamp, region, service);
hmac_sha256_digest signature = hmac_sha256(std::string_view(signing_key.data(), signing_key.size()), string_to_sign);
return to_hex(bytes_view(reinterpret_cast<const int8_t*>(signature.data()), signature.size()));
}
future<std::string> get_key_from_roles(cql3::query_processor& qp, std::string username) {
static const sstring query = format("SELECT salted_hash FROM {} WHERE {} = ?",
auth::meta::roles_table::qualified_name, auth::meta::roles_table::role_col_name);
auto cl = auth::password_authenticator::consistency_for_user(username);
return qp.execute_internal(query, cl, auth::internal_distributed_query_state(), {sstring(username)}, true).then_wrapped([username = std::move(username)] (future<::shared_ptr<cql3::untyped_result_set>> f) {
auto res = f.get0();
auto salted_hash = std::optional<sstring>();
if (res->empty()) {
throw api_error::unrecognized_client(fmt::format("User not found: {}", username));
}
salted_hash = res->one().get_opt<sstring>("salted_hash");
if (!salted_hash) {
throw api_error::unrecognized_client(fmt::format("No password found for user: {}", username));
}
return make_ready_future<std::string>(*salted_hash);
});
}
}

View File

@@ -1,46 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include <string>
#include <string_view>
#include <array>
#include "gc_clock.hh"
#include "utils/loading_cache.hh"
namespace cql3 {
class query_processor;
}
namespace alternator {
using hmac_sha256_digest = std::array<char, 32>;
using key_cache = utils::loading_cache<std::string, std::string>;
std::string get_signature(std::string_view access_key_id, std::string_view secret_access_key, std::string_view host, std::string_view method,
std::string_view orig_datestamp, std::string_view signed_headers_str, const std::map<std::string_view, std::string_view>& signed_headers_map,
std::string_view body_content, std::string_view region, std::string_view service, std::string_view query_string);
future<std::string> get_key_from_roles(cql3::query_processor& qp, std::string username);
}

View File

@@ -1,145 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
// The DynamoAPI dictates that "binary" (a.k.a. "bytes" or "blob") values
// be encoded in the JSON API as base64-encoded strings. This is code to
// convert byte arrays to base64-encoded strings, and back.
#include "base64.hh"
#include <ctype.h>
// Arrays for quickly converting to and from an integer between 0 and 63,
// and the character used in base64 encoding to represent it.
static class base64_chars {
public:
static constexpr const char to[] =
"ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789+/";
int8_t from[255];
base64_chars() {
static_assert(sizeof(to) == 64 + 1);
for (int i = 0; i < 255; i++) {
from[i] = -1; // signal invalid character
}
for (int i = 0; i < 64; i++) {
from[(unsigned) to[i]] = i;
}
}
} base64_chars;
std::string base64_encode(bytes_view in) {
std::string ret;
ret.reserve(((4 * in.size() / 3) + 3) & ~3);
int i = 0;
unsigned char chunk3[3]; // chunk of input
for (auto byte : in) {
chunk3[i++] = byte;
if (i == 3) {
ret += base64_chars.to[ (chunk3[0] & 0xfc) >> 2 ];
ret += base64_chars.to[ ((chunk3[0] & 0x03) << 4) + ((chunk3[1] & 0xf0) >> 4) ];
ret += base64_chars.to[ ((chunk3[1] & 0x0f) << 2) + ((chunk3[2] & 0xc0) >> 6) ];
ret += base64_chars.to[ chunk3[2] & 0x3f ];
i = 0;
}
}
if (i) {
// i can be 1 or 2.
for(int j = i; j < 3; j++)
chunk3[j] = '\0';
ret += base64_chars.to[ ( chunk3[0] & 0xfc) >> 2 ];
ret += base64_chars.to[ ((chunk3[0] & 0x03) << 4) + ((chunk3[1] & 0xf0) >> 4) ];
if (i == 2) {
ret += base64_chars.to[ ((chunk3[1] & 0x0f) << 2) + ((chunk3[2] & 0xc0) >> 6) ];
} else {
ret += '=';
}
ret += '=';
}
return ret;
}
static std::string base64_decode_string(std::string_view in) {
int i = 0;
int8_t chunk4[4]; // chunk of input, each byte converted to 0..63;
std::string ret;
ret.reserve(in.size() * 3 / 4);
for (unsigned char c : in) {
uint8_t dc = base64_chars.from[c];
if (dc == 255) {
// Any unexpected character, include the "=" character usually
// used for padding, signals the end of the decode.
break;
}
chunk4[i++] = dc;
if (i == 4) {
ret += (chunk4[0] << 2) + ((chunk4[1] & 0x30) >> 4);
ret += ((chunk4[1] & 0xf) << 4) + ((chunk4[2] & 0x3c) >> 2);
ret += ((chunk4[2] & 0x3) << 6) + chunk4[3];
i = 0;
}
}
if (i) {
// i can be 2 or 3, meaning 1 or 2 more output characters
if (i>=2)
ret += (chunk4[0] << 2) + ((chunk4[1] & 0x30) >> 4);
if (i==3)
ret += ((chunk4[1] & 0xf) << 4) + ((chunk4[2] & 0x3c) >> 2);
}
return ret;
}
bytes base64_decode(std::string_view in) {
// FIXME: This copy is sad. The problem is we need back "bytes"
// but "bytes" doesn't have efficient append and std::string.
// To fix this we need to use bytes' "uninitialized" feature.
std::string ret = base64_decode_string(in);
return bytes(ret.begin(), ret.end());
}
static size_t base64_padding_len(std::string_view str) {
size_t padding = 0;
padding += (!str.empty() && str.back() == '=');
padding += (str.size() > 1 && *(str.end() - 2) == '=');
return padding;
}
size_t base64_decoded_len(std::string_view str) {
return str.size() / 4 * 3 - base64_padding_len(str);
}
bool base64_begins_with(std::string_view base, std::string_view operand) {
if (base.size() < operand.size() || base.size() % 4 != 0 || operand.size() % 4 != 0) {
return false;
}
if (base64_padding_len(operand) == 0) {
return base.starts_with(operand);
}
const std::string_view unpadded_base_prefix = base.substr(0, operand.size() - 4);
const std::string_view unpadded_operand = operand.substr(0, operand.size() - 4);
if (unpadded_base_prefix != unpadded_operand) {
return false;
}
// Decode and compare last 4 bytes of base64-encoded strings
const std::string base_remainder = base64_decode_string(base.substr(operand.size() - 4, operand.size()));
const std::string operand_remainder = base64_decode_string(operand.substr(operand.size() - 4));
return base_remainder.starts_with(operand_remainder);
}

View File

@@ -1,38 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include <string_view>
#include "bytes.hh"
#include "utils/rjson.hh"
std::string base64_encode(bytes_view);
bytes base64_decode(std::string_view);
inline bytes base64_decode(const rjson::value& v) {
return base64_decode(std::string_view(v.GetString(), v.GetStringLength()));
}
size_t base64_decoded_len(std::string_view str);
bool base64_begins_with(std::string_view base, std::string_view operand);

View File

@@ -1,650 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#include <list>
#include <map>
#include <string_view>
#include "alternator/conditions.hh"
#include "alternator/error.hh"
#include "cql3/constants.hh"
#include <unordered_map>
#include "utils/rjson.hh"
#include "serialization.hh"
#include "base64.hh"
#include <stdexcept>
#include <boost/algorithm/cxx11/all_of.hpp>
#include <boost/algorithm/cxx11/any_of.hpp>
#include "utils/overloaded_functor.hh"
#include "expressions.hh"
namespace alternator {
static logging::logger clogger("alternator-conditions");
comparison_operator_type get_comparison_operator(const rjson::value& comparison_operator) {
static std::unordered_map<std::string, comparison_operator_type> ops = {
{"EQ", comparison_operator_type::EQ},
{"NE", comparison_operator_type::NE},
{"LE", comparison_operator_type::LE},
{"LT", comparison_operator_type::LT},
{"GE", comparison_operator_type::GE},
{"GT", comparison_operator_type::GT},
{"IN", comparison_operator_type::IN},
{"NULL", comparison_operator_type::IS_NULL},
{"NOT_NULL", comparison_operator_type::NOT_NULL},
{"BETWEEN", comparison_operator_type::BETWEEN},
{"BEGINS_WITH", comparison_operator_type::BEGINS_WITH},
{"CONTAINS", comparison_operator_type::CONTAINS},
{"NOT_CONTAINS", comparison_operator_type::NOT_CONTAINS},
};
if (!comparison_operator.IsString()) {
throw api_error::validation(format("Invalid comparison operator definition {}", rjson::print(comparison_operator)));
}
std::string op = comparison_operator.GetString();
auto it = ops.find(op);
if (it == ops.end()) {
throw api_error::validation(format("Unsupported comparison operator {}", op));
}
return it->second;
}
namespace {
struct size_check {
// True iff size passes this check.
virtual bool operator()(rapidjson::SizeType size) const = 0;
// Check description, such that format("expected array {}", check.what()) is human-readable.
virtual sstring what() const = 0;
};
class exact_size : public size_check {
rapidjson::SizeType _expected;
public:
explicit exact_size(rapidjson::SizeType expected) : _expected(expected) {}
bool operator()(rapidjson::SizeType size) const override { return size == _expected; }
sstring what() const override { return format("of size {}", _expected); }
};
struct empty : public size_check {
bool operator()(rapidjson::SizeType size) const override { return size < 1; }
sstring what() const override { return "to be empty"; }
};
struct nonempty : public size_check {
bool operator()(rapidjson::SizeType size) const override { return size > 0; }
sstring what() const override { return "to be non-empty"; }
};
} // anonymous namespace
// Check that array has the expected number of elements
static void verify_operand_count(const rjson::value* array, const size_check& expected, const rjson::value& op) {
if (!array && expected(0)) {
// If expected() allows an empty AttributeValueList, it is also fine
// that it is missing.
return;
}
if (!array || !array->IsArray()) {
throw api_error::validation("With ComparisonOperator, AttributeValueList must be given and an array");
}
if (!expected(array->Size())) {
throw api_error::validation(
format("{} operator requires AttributeValueList {}, instead found list size {}",
op, expected.what(), array->Size()));
}
}
struct rjson_engaged_ptr_comp {
bool operator()(const rjson::value* p1, const rjson::value* p2) const {
return rjson::single_value_comp()(*p1, *p2);
}
};
// It's not enough to compare underlying JSON objects when comparing sets,
// as internally they're stored in an array, and the order of elements is
// not important in set equality. See issue #5021
static bool check_EQ_for_sets(const rjson::value& set1, const rjson::value& set2) {
if (set1.Size() != set2.Size()) {
return false;
}
std::set<const rjson::value*, rjson_engaged_ptr_comp> set1_raw;
for (auto it = set1.Begin(); it != set1.End(); ++it) {
set1_raw.insert(&*it);
}
for (const auto& a : set2.GetArray()) {
if (!set1_raw.contains(&a)) {
return false;
}
}
return true;
}
// Check if two JSON-encoded values match with the EQ relation
static bool check_EQ(const rjson::value* v1, const rjson::value& v2) {
if (!v1) {
return false;
}
if (v1->IsObject() && v1->MemberCount() == 1 && v2.IsObject() && v2.MemberCount() == 1) {
auto it1 = v1->MemberBegin();
auto it2 = v2.MemberBegin();
if ((it1->name == "SS" && it2->name == "SS") || (it1->name == "NS" && it2->name == "NS") || (it1->name == "BS" && it2->name == "BS")) {
return check_EQ_for_sets(it1->value, it2->value);
}
}
return *v1 == v2;
}
// Check if two JSON-encoded values match with the NE relation
static bool check_NE(const rjson::value* v1, const rjson::value& v2) {
return !v1 || *v1 != v2; // null is unequal to anything.
}
// Check if two JSON-encoded values match with the BEGINS_WITH relation
static bool check_BEGINS_WITH(const rjson::value* v1, const rjson::value& v2) {
// BEGINS_WITH requires that its single operand (v2) be a string or
// binary - otherwise it's a validation error. However, problems with
// the stored attribute (v1) will just return false (no match).
if (!v2.IsObject() || v2.MemberCount() != 1) {
throw api_error::validation(format("BEGINS_WITH operator encountered malformed AttributeValue: {}", v2));
}
auto it2 = v2.MemberBegin();
if (it2->name != "S" && it2->name != "B") {
throw api_error::validation(format("BEGINS_WITH operator requires String or Binary type in AttributeValue, got {}", it2->name));
}
if (!v1 || !v1->IsObject() || v1->MemberCount() != 1) {
return false;
}
auto it1 = v1->MemberBegin();
if (it1->name != it2->name) {
return false;
}
if (it2->name == "S") {
return rjson::to_string_view(it1->value).starts_with(rjson::to_string_view(it2->value));
} else /* it2->name == "B" */ {
return base64_begins_with(rjson::to_string_view(it1->value), rjson::to_string_view(it2->value));
}
}
static bool is_set_of(const rjson::value& type1, const rjson::value& type2) {
return (type2 == "S" && type1 == "SS") || (type2 == "N" && type1 == "NS") || (type2 == "B" && type1 == "BS");
}
// Check if two JSON-encoded values match with the CONTAINS relation
bool check_CONTAINS(const rjson::value* v1, const rjson::value& v2) {
if (!v1) {
return false;
}
const auto& kv1 = *v1->MemberBegin();
const auto& kv2 = *v2.MemberBegin();
if (kv1.name == "S" && kv2.name == "S") {
return rjson::to_string_view(kv1.value).find(rjson::to_string_view(kv2.value)) != std::string_view::npos;
} else if (kv1.name == "B" && kv2.name == "B") {
return base64_decode(kv1.value).find(base64_decode(kv2.value)) != bytes::npos;
} else if (is_set_of(kv1.name, kv2.name)) {
for (auto i = kv1.value.Begin(); i != kv1.value.End(); ++i) {
if (*i == kv2.value) {
return true;
}
}
} else if (kv1.name == "L") {
for (auto i = kv1.value.Begin(); i != kv1.value.End(); ++i) {
if (!i->IsObject() || i->MemberCount() != 1) {
clogger.error("check_CONTAINS received a list whose element is malformed");
return false;
}
const auto& el = *i->MemberBegin();
if (el.name == kv2.name && el.value == kv2.value) {
return true;
}
}
}
return false;
}
// Check if two JSON-encoded values match with the NOT_CONTAINS relation
static bool check_NOT_CONTAINS(const rjson::value* v1, const rjson::value& v2) {
if (!v1) {
return false;
}
return !check_CONTAINS(v1, v2);
}
// Check if a JSON-encoded value equals any element of an array, which must have at least one element.
static bool check_IN(const rjson::value* val, const rjson::value& array) {
if (!array[0].IsObject() || array[0].MemberCount() != 1) {
throw api_error::validation(
format("IN operator encountered malformed AttributeValue: {}", array[0]));
}
const auto& type = array[0].MemberBegin()->name;
if (type != "S" && type != "N" && type != "B") {
throw api_error::validation(
"IN operator requires AttributeValueList elements to be of type String, Number, or Binary ");
}
if (!val) {
return false;
}
bool have_match = false;
for (const auto& elem : array.GetArray()) {
if (!elem.IsObject() || elem.MemberCount() != 1 || elem.MemberBegin()->name != type) {
throw api_error::validation(
"IN operator requires all AttributeValueList elements to have the same type ");
}
if (!have_match && *val == elem) {
// Can't return yet, must check types of all array elements. <sigh>
have_match = true;
}
}
return have_match;
}
// Another variant of check_IN, this one for ConditionExpression. It needs to
// check whether the first element in the given vector is equal to any of the
// others.
static bool check_IN(const std::vector<rjson::value>& array) {
const rjson::value* first = &array[0];
for (unsigned i = 1; i < array.size(); i++) {
if (check_EQ(first, array[i])) {
return true;
}
}
return false;
}
static bool check_NULL(const rjson::value* val) {
return val == nullptr;
}
static bool check_NOT_NULL(const rjson::value* val) {
return val != nullptr;
}
// Check if two JSON-encoded values match with cmp.
template <typename Comparator>
bool check_compare(const rjson::value* v1, const rjson::value& v2, const Comparator& cmp) {
if (!v2.IsObject() || v2.MemberCount() != 1) {
throw api_error::validation(
format("{} requires a single AttributeValue of type String, Number, or Binary",
cmp.diagnostic));
}
const auto& kv2 = *v2.MemberBegin();
if (kv2.name != "S" && kv2.name != "N" && kv2.name != "B") {
throw api_error::validation(
format("{} requires a single AttributeValue of type String, Number, or Binary",
cmp.diagnostic));
}
if (!v1 || !v1->IsObject() || v1->MemberCount() != 1) {
return false;
}
const auto& kv1 = *v1->MemberBegin();
if (kv1.name != kv2.name) {
return false;
}
if (kv1.name == "N") {
return cmp(unwrap_number(*v1, cmp.diagnostic), unwrap_number(v2, cmp.diagnostic));
}
if (kv1.name == "S") {
return cmp(std::string_view(kv1.value.GetString(), kv1.value.GetStringLength()),
std::string_view(kv2.value.GetString(), kv2.value.GetStringLength()));
}
if (kv1.name == "B") {
return cmp(base64_decode(kv1.value), base64_decode(kv2.value));
}
clogger.error("check_compare panic: LHS type equals RHS type, but one is in {N,S,B} while the other isn't");
return false;
}
struct cmp_lt {
template <typename T> bool operator()(const T& lhs, const T& rhs) const { return lhs < rhs; }
// We cannot use the normal comparison operators like "<" on the bytes
// type, because they treat individual bytes as signed but we need to
// compare them as *unsigned*. So we need a specialization for bytes.
bool operator()(const bytes& lhs, const bytes& rhs) const { return compare_unsigned(lhs, rhs) < 0; }
static constexpr const char* diagnostic = "LT operator";
};
struct cmp_le {
template <typename T> bool operator()(const T& lhs, const T& rhs) const { return lhs <= rhs; }
bool operator()(const bytes& lhs, const bytes& rhs) const { return compare_unsigned(lhs, rhs) <= 0; }
static constexpr const char* diagnostic = "LE operator";
};
struct cmp_ge {
template <typename T> bool operator()(const T& lhs, const T& rhs) const { return lhs >= rhs; }
bool operator()(const bytes& lhs, const bytes& rhs) const { return compare_unsigned(lhs, rhs) >= 0; }
static constexpr const char* diagnostic = "GE operator";
};
struct cmp_gt {
template <typename T> bool operator()(const T& lhs, const T& rhs) const { return lhs > rhs; }
bool operator()(const bytes& lhs, const bytes& rhs) const { return compare_unsigned(lhs, rhs) > 0; }
static constexpr const char* diagnostic = "GT operator";
};
// True if v is between lb and ub, inclusive. Throws if lb > ub.
template <typename T>
static bool check_BETWEEN(const T& v, const T& lb, const T& ub) {
if (cmp_lt()(ub, lb)) {
throw api_error::validation(
format("BETWEEN operator requires lower_bound <= upper_bound, but {} > {}", lb, ub));
}
return cmp_ge()(v, lb) && cmp_le()(v, ub);
}
static bool check_BETWEEN(const rjson::value* v, const rjson::value& lb, const rjson::value& ub) {
if (!v) {
return false;
}
if (!v->IsObject() || v->MemberCount() != 1) {
throw api_error::validation(format("BETWEEN operator encountered malformed AttributeValue: {}", *v));
}
if (!lb.IsObject() || lb.MemberCount() != 1) {
throw api_error::validation(format("BETWEEN operator encountered malformed AttributeValue: {}", lb));
}
if (!ub.IsObject() || ub.MemberCount() != 1) {
throw api_error::validation(format("BETWEEN operator encountered malformed AttributeValue: {}", ub));
}
const auto& kv_v = *v->MemberBegin();
const auto& kv_lb = *lb.MemberBegin();
const auto& kv_ub = *ub.MemberBegin();
if (kv_lb.name != kv_ub.name) {
throw api_error::validation(
format("BETWEEN operator requires the same type for lower and upper bound; instead got {} and {}",
kv_lb.name, kv_ub.name));
}
if (kv_v.name != kv_lb.name) { // Cannot compare different types, so v is NOT between lb and ub.
return false;
}
if (kv_v.name == "N") {
const char* diag = "BETWEEN operator";
return check_BETWEEN(unwrap_number(*v, diag), unwrap_number(lb, diag), unwrap_number(ub, diag));
}
if (kv_v.name == "S") {
return check_BETWEEN(std::string_view(kv_v.value.GetString(), kv_v.value.GetStringLength()),
std::string_view(kv_lb.value.GetString(), kv_lb.value.GetStringLength()),
std::string_view(kv_ub.value.GetString(), kv_ub.value.GetStringLength()));
}
if (kv_v.name == "B") {
return check_BETWEEN(base64_decode(kv_v.value), base64_decode(kv_lb.value), base64_decode(kv_ub.value));
}
throw api_error::validation(
format("BETWEEN operator requires AttributeValueList elements to be of type String, Number, or Binary; instead got {}",
kv_lb.name));
}
// Verify one Expect condition on one attribute (whose content is "got")
// for the verify_expected() below.
// This function returns true or false depending on whether the condition
// succeeded - it does not throw ConditionalCheckFailedException.
// However, it may throw ValidationException on input validation errors.
static bool verify_expected_one(const rjson::value& condition, const rjson::value* got) {
const rjson::value* comparison_operator = rjson::find(condition, "ComparisonOperator");
const rjson::value* attribute_value_list = rjson::find(condition, "AttributeValueList");
const rjson::value* value = rjson::find(condition, "Value");
const rjson::value* exists = rjson::find(condition, "Exists");
// There are three types of conditions that Expected supports:
// A value, not-exists, and a comparison of some kind. Each allows
// and requires a different combinations of parameters in the request
if (value) {
if (exists && (!exists->IsBool() || exists->GetBool() != true)) {
throw api_error::validation("Cannot combine Value with Exists!=true");
}
if (comparison_operator) {
throw api_error::validation("Cannot combine Value with ComparisonOperator");
}
return check_EQ(got, *value);
} else if (exists) {
if (comparison_operator) {
throw api_error::validation("Cannot combine Exists with ComparisonOperator");
}
if (!exists->IsBool() || exists->GetBool() != false) {
throw api_error::validation("Exists!=false requires Value");
}
// Remember Exists=false, so we're checking that the attribute does *not* exist:
return !got;
} else {
if (!comparison_operator) {
throw api_error::validation("Missing ComparisonOperator, Value or Exists");
}
comparison_operator_type op = get_comparison_operator(*comparison_operator);
switch (op) {
case comparison_operator_type::EQ:
verify_operand_count(attribute_value_list, exact_size(1), *comparison_operator);
return check_EQ(got, (*attribute_value_list)[0]);
case comparison_operator_type::NE:
verify_operand_count(attribute_value_list, exact_size(1), *comparison_operator);
return check_NE(got, (*attribute_value_list)[0]);
case comparison_operator_type::LT:
verify_operand_count(attribute_value_list, exact_size(1), *comparison_operator);
return check_compare(got, (*attribute_value_list)[0], cmp_lt{});
case comparison_operator_type::LE:
verify_operand_count(attribute_value_list, exact_size(1), *comparison_operator);
return check_compare(got, (*attribute_value_list)[0], cmp_le{});
case comparison_operator_type::GT:
verify_operand_count(attribute_value_list, exact_size(1), *comparison_operator);
return check_compare(got, (*attribute_value_list)[0], cmp_gt{});
case comparison_operator_type::GE:
verify_operand_count(attribute_value_list, exact_size(1), *comparison_operator);
return check_compare(got, (*attribute_value_list)[0], cmp_ge{});
case comparison_operator_type::BEGINS_WITH:
verify_operand_count(attribute_value_list, exact_size(1), *comparison_operator);
return check_BEGINS_WITH(got, (*attribute_value_list)[0]);
case comparison_operator_type::IN:
verify_operand_count(attribute_value_list, nonempty(), *comparison_operator);
return check_IN(got, *attribute_value_list);
case comparison_operator_type::IS_NULL:
verify_operand_count(attribute_value_list, empty(), *comparison_operator);
return check_NULL(got);
case comparison_operator_type::NOT_NULL:
verify_operand_count(attribute_value_list, empty(), *comparison_operator);
return check_NOT_NULL(got);
case comparison_operator_type::BETWEEN:
verify_operand_count(attribute_value_list, exact_size(2), *comparison_operator);
return check_BETWEEN(got, (*attribute_value_list)[0], (*attribute_value_list)[1]);
case comparison_operator_type::CONTAINS:
{
verify_operand_count(attribute_value_list, exact_size(1), *comparison_operator);
// Expected's "CONTAINS" has this artificial limitation.
// ConditionExpression's "contains()" does not...
const rjson::value& arg = (*attribute_value_list)[0];
const auto& argtype = (*arg.MemberBegin()).name;
if (argtype != "S" && argtype != "N" && argtype != "B") {
throw api_error::validation(
format("CONTAINS operator requires a single AttributeValue of type String, Number, or Binary, "
"got {} instead", argtype));
}
return check_CONTAINS(got, arg);
}
case comparison_operator_type::NOT_CONTAINS:
{
verify_operand_count(attribute_value_list, exact_size(1), *comparison_operator);
// Expected's "NOT_CONTAINS" has this artificial limitation.
// ConditionExpression's "contains()" does not...
const rjson::value& arg = (*attribute_value_list)[0];
const auto& argtype = (*arg.MemberBegin()).name;
if (argtype != "S" && argtype != "N" && argtype != "B") {
throw api_error::validation(
format("CONTAINS operator requires a single AttributeValue of type String, Number, or Binary, "
"got {} instead", argtype));
}
return check_NOT_CONTAINS(got, arg);
}
}
throw std::logic_error(format("Internal error: corrupted operator enum: {}", int(op)));
}
}
conditional_operator_type get_conditional_operator(const rjson::value& req) {
const rjson::value* conditional_operator = rjson::find(req, "ConditionalOperator");
if (!conditional_operator) {
return conditional_operator_type::MISSING;
}
if (!conditional_operator->IsString()) {
throw api_error::validation("'ConditionalOperator' parameter, if given, must be a string");
}
auto s = rjson::to_string_view(*conditional_operator);
if (s == "AND") {
return conditional_operator_type::AND;
} else if (s == "OR") {
return conditional_operator_type::OR;
} else {
throw api_error::validation(
format("'ConditionalOperator' parameter must be AND, OR or missing. Found {}.", s));
}
}
// Check if the existing values of the item (previous_item) match the
// conditions given by the Expected and ConditionalOperator parameters
// (if they exist) in the request (an UpdateItem, PutItem or DeleteItem).
// This function can throw an ValidationException API error if there
// are errors in the format of the condition itself.
bool verify_expected(const rjson::value& req, const rjson::value* previous_item) {
const rjson::value* expected = rjson::find(req, "Expected");
auto conditional_operator = get_conditional_operator(req);
if (conditional_operator != conditional_operator_type::MISSING &&
(!expected || (expected->IsObject() && expected->GetObject().ObjectEmpty()))) {
throw api_error::validation("'ConditionalOperator' parameter cannot be specified for missing or empty Expression");
}
if (!expected) {
return true;
}
if (!expected->IsObject()) {
throw api_error::validation("'Expected' parameter, if given, must be an object");
}
bool require_all = conditional_operator != conditional_operator_type::OR;
return verify_condition(*expected, require_all, previous_item);
}
bool verify_condition(const rjson::value& condition, bool require_all, const rjson::value* previous_item) {
for (auto it = condition.MemberBegin(); it != condition.MemberEnd(); ++it) {
const rjson::value* got = nullptr;
if (previous_item) {
got = rjson::find(*previous_item, rjson::to_string_view(it->name));
}
bool success = verify_expected_one(it->value, got);
if (success && !require_all) {
// When !require_all, one success is enough!
return true;
} else if (!success && require_all) {
// When require_all, one failure is enough!
return false;
}
}
// If we got here and require_all, none of the checks failed, so succeed.
// If we got here and !require_all, all of the checks failed, so fail.
return require_all;
}
static bool calculate_primitive_condition(const parsed::primitive_condition& cond,
const rjson::value* previous_item) {
std::vector<rjson::value> calculated_values;
calculated_values.reserve(cond._values.size());
for (const parsed::value& v : cond._values) {
calculated_values.push_back(calculate_value(v,
cond._op == parsed::primitive_condition::type::VALUE ?
calculate_value_caller::ConditionExpressionAlone :
calculate_value_caller::ConditionExpression,
previous_item));
}
switch (cond._op) {
case parsed::primitive_condition::type::BETWEEN:
if (calculated_values.size() != 3) {
// Shouldn't happen unless we have a bug in the parser
throw std::logic_error(format("Wrong number of values {} in BETWEEN primitive_condition", cond._values.size()));
}
return check_BETWEEN(&calculated_values[0], calculated_values[1], calculated_values[2]);
case parsed::primitive_condition::type::IN:
return check_IN(calculated_values);
case parsed::primitive_condition::type::VALUE:
if (calculated_values.size() != 1) {
// Shouldn't happen unless we have a bug in the parser
throw std::logic_error(format("Unexpected values in primitive_condition", cond._values.size()));
}
// Unwrap the boolean wrapped as the value (if it is a boolean)
if (calculated_values[0].IsObject() && calculated_values[0].MemberCount() == 1) {
auto it = calculated_values[0].MemberBegin();
if (it->name == "BOOL" && it->value.IsBool()) {
return it->value.GetBool();
}
}
throw api_error::validation(
format("ConditionExpression: condition results in a non-boolean value: {}",
calculated_values[0]));
default:
// All the rest of the operators have exactly two parameters (and unless
// we have a bug in the parser, that's what we have in the parsed object:
if (calculated_values.size() != 2) {
throw std::logic_error(format("Wrong number of values {} in primitive_condition object", cond._values.size()));
}
}
switch (cond._op) {
case parsed::primitive_condition::type::EQ:
return check_EQ(&calculated_values[0], calculated_values[1]);
case parsed::primitive_condition::type::NE:
return check_NE(&calculated_values[0], calculated_values[1]);
case parsed::primitive_condition::type::GT:
return check_compare(&calculated_values[0], calculated_values[1], cmp_gt{});
case parsed::primitive_condition::type::GE:
return check_compare(&calculated_values[0], calculated_values[1], cmp_ge{});
case parsed::primitive_condition::type::LT:
return check_compare(&calculated_values[0], calculated_values[1], cmp_lt{});
case parsed::primitive_condition::type::LE:
return check_compare(&calculated_values[0], calculated_values[1], cmp_le{});
default:
// Shouldn't happen unless we have a bug in the parser
throw std::logic_error(format("Unknown type {} in primitive_condition object", (int)(cond._op)));
}
}
// Check if the existing values of the item (previous_item) match the
// conditions given by the given parsed ConditionExpression.
bool verify_condition_expression(
const parsed::condition_expression& condition_expression,
const rjson::value* previous_item) {
if (condition_expression.empty()) {
return true;
}
bool ret = std::visit(overloaded_functor {
[&] (const parsed::primitive_condition& cond) -> bool {
return calculate_primitive_condition(cond, previous_item);
},
[&] (const parsed::condition_expression::condition_list& list) -> bool {
auto verify_condition = [&] (const parsed::condition_expression& e) {
return verify_condition_expression(e, previous_item);
};
switch (list.op) {
case '&':
return boost::algorithm::all_of(list.conditions, verify_condition);
case '|':
return boost::algorithm::any_of(list.conditions, verify_condition);
default:
// Shouldn't happen unless we have a bug in the parser
throw std::logic_error("bad operator in condition_list");
}
}
}, condition_expression._expression);
return condition_expression._negated ? !ret : ret;
}
}

View File

@@ -1,60 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
/*
* This file contains definitions and functions related to placing conditions
* on Alternator queries (equivalent of CQL's restrictions).
*
* With conditions, it's possible to add criteria to selection requests (Scan, Query)
* and use them for narrowing down the result set, by means of filtering or indexing.
*
* Ref: https://docs.aws.amazon.com/amazondynamodb/latest/APIReference/API_Condition.html
*/
#pragma once
#include "cql3/restrictions/statement_restrictions.hh"
#include "serialization.hh"
#include "expressions_types.hh"
namespace alternator {
enum class comparison_operator_type {
EQ, NE, LE, LT, GE, GT, IN, BETWEEN, CONTAINS, NOT_CONTAINS, IS_NULL, NOT_NULL, BEGINS_WITH
};
comparison_operator_type get_comparison_operator(const rjson::value& comparison_operator);
enum class conditional_operator_type {
AND, OR, MISSING
};
conditional_operator_type get_conditional_operator(const rjson::value& req);
bool verify_expected(const rjson::value& req, const rjson::value* previous_item);
bool verify_condition(const rjson::value& condition, bool require_all, const rjson::value* previous_item);
bool check_CONTAINS(const rjson::value* v1, const rjson::value& v2);
bool verify_condition_expression(
const parsed::condition_expression& condition_expression,
const rjson::value* previous_item);
}

View File

@@ -1,86 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include <seastar/http/httpd.hh>
#include "seastarx.hh"
namespace alternator {
// api_error contains a DynamoDB error message to be returned to the user.
// It can be returned by value (see executor::request_return_type) or thrown.
// The DynamoDB's error messages are described in detail in
// https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/Programming.Errors.html
// An error message has an HTTP code (almost always 400), a type, e.g.,
// "ResourceNotFoundException", and a human readable message.
// Eventually alternator::api_handler will convert a returned or thrown
// api_error into a JSON object, and that is returned to the user.
class api_error final {
public:
using status_type = httpd::reply::status_type;
status_type _http_code;
std::string _type;
std::string _msg;
api_error(std::string type, std::string msg, status_type http_code = status_type::bad_request)
: _http_code(std::move(http_code))
, _type(std::move(type))
, _msg(std::move(msg))
{ }
// Factory functions for some common types of DynamoDB API errors
static api_error validation(std::string msg) {
return api_error("ValidationException", std::move(msg));
}
static api_error resource_not_found(std::string msg) {
return api_error("ResourceNotFoundException", std::move(msg));
}
static api_error resource_in_use(std::string msg) {
return api_error("ResourceInUseException", std::move(msg));
}
static api_error invalid_signature(std::string msg) {
return api_error("InvalidSignatureException", std::move(msg));
}
static api_error unrecognized_client(std::string msg) {
return api_error("UnrecognizedClientException", std::move(msg));
}
static api_error unknown_operation(std::string msg) {
return api_error("UnknownOperationException", std::move(msg));
}
static api_error access_denied(std::string msg) {
return api_error("AccessDeniedException", std::move(msg));
}
static api_error conditional_check_failed(std::string msg) {
return api_error("ConditionalCheckFailedException", std::move(msg));
}
static api_error expired_iterator(std::string msg) {
return api_error("ExpiredIteratorException", std::move(msg));
}
static api_error trimmed_data_access_exception(std::string msg) {
return api_error("TrimmedDataAccessException", std::move(msg));
}
static api_error internal(std::string msg) {
return api_error("InternalServerError", std::move(msg), reply::status_type::internal_server_error);
}
};
}

File diff suppressed because it is too large Load Diff

View File

@@ -1,154 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include <seastar/core/future.hh>
#include <seastar/http/httpd.hh>
#include "seastarx.hh"
#include <seastar/json/json_elements.hh>
#include <seastar/core/sharded.hh>
#include "service/storage_proxy.hh"
#include "service/migration_manager.hh"
#include "service/client_state.hh"
#include "db/timeout_clock.hh"
#include "alternator/error.hh"
#include "stats.hh"
#include "utils/rjson.hh"
namespace db {
class system_distributed_keyspace;
}
namespace query {
class partition_slice;
class result;
}
namespace cql3::selection {
class selection;
}
namespace service {
class storage_service;
}
namespace alternator {
class rmw_operation;
struct make_jsonable : public json::jsonable {
rjson::value _value;
public:
explicit make_jsonable(rjson::value&& value);
std::string to_json() const override;
};
struct json_string : public json::jsonable {
std::string _value;
public:
explicit json_string(std::string&& value);
std::string to_json() const override;
};
class executor : public peering_sharded_service<executor> {
service::storage_proxy& _proxy;
service::migration_manager& _mm;
db::system_distributed_keyspace& _sdks;
service::storage_service& _ss;
// An smp_service_group to be used for limiting the concurrency when
// forwarding Alternator request between shards - if necessary for LWT.
smp_service_group _ssg;
public:
using client_state = service::client_state;
using request_return_type = std::variant<json::json_return_type, api_error>;
stats _stats;
static constexpr auto ATTRS_COLUMN_NAME = ":attrs";
static constexpr auto KEYSPACE_NAME_PREFIX = "alternator_";
static constexpr std::string_view INTERNAL_TABLE_PREFIX = ".scylla.alternator.";
executor(service::storage_proxy& proxy, service::migration_manager& mm, db::system_distributed_keyspace& sdks, service::storage_service& ss, smp_service_group ssg)
: _proxy(proxy), _mm(mm), _sdks(sdks), _ss(ss), _ssg(ssg) {}
future<request_return_type> create_table(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
future<request_return_type> describe_table(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
future<request_return_type> delete_table(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
future<request_return_type> update_table(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
future<request_return_type> put_item(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
future<request_return_type> get_item(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
future<request_return_type> delete_item(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
future<request_return_type> update_item(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
future<request_return_type> list_tables(client_state& client_state, service_permit permit, rjson::value request);
future<request_return_type> scan(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
future<request_return_type> describe_endpoints(client_state& client_state, service_permit permit, rjson::value request, std::string host_header);
future<request_return_type> batch_write_item(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
future<request_return_type> batch_get_item(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
future<request_return_type> query(client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value request);
future<request_return_type> tag_resource(client_state& client_state, service_permit permit, rjson::value request);
future<request_return_type> untag_resource(client_state& client_state, service_permit permit, rjson::value request);
future<request_return_type> list_tags_of_resource(client_state& client_state, service_permit permit, rjson::value request);
future<request_return_type> list_streams(client_state& client_state, service_permit permit, rjson::value request);
future<request_return_type> describe_stream(client_state& client_state, service_permit permit, rjson::value request);
future<request_return_type> get_shard_iterator(client_state& client_state, service_permit permit, rjson::value request);
future<request_return_type> get_records(client_state& client_state, tracing::trace_state_ptr, service_permit permit, rjson::value request);
future<> start();
future<> stop() { return make_ready_future<>(); }
future<> create_keyspace(std::string_view keyspace_name);
static tracing::trace_state_ptr maybe_trace_query(client_state& client_state, sstring_view op, sstring_view query);
static sstring table_name(const schema&);
static db::timeout_clock::time_point default_timeout();
static schema_ptr find_table(service::storage_proxy&, const rjson::value& request);
private:
friend class rmw_operation;
static bool is_alternator_keyspace(const sstring& ks_name);
static sstring make_keyspace_name(const sstring& table_name);
static void describe_key_schema(rjson::value& parent, const schema&, std::unordered_map<std::string,std::string> * = nullptr);
static void describe_key_schema(rjson::value& parent, const schema& schema, std::unordered_map<std::string,std::string>&);
public:
static std::optional<rjson::value> describe_single_item(schema_ptr,
const query::partition_slice&,
const cql3::selection::selection&,
const query::result&,
const std::unordered_set<std::string>&);
static void describe_single_item(const cql3::selection::selection&,
const std::vector<bytes_opt>&,
const std::unordered_set<std::string>&,
rjson::value&,
bool = false);
void add_stream_options(const rjson::value& stream_spec, schema_builder&) const;
void supplement_table_info(rjson::value& descr, const schema& schema) const;
void supplement_table_stream_info(rjson::value& descr, const schema& schema) const;
};
}

View File

@@ -1,728 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#include "expressions.hh"
#include "serialization.hh"
#include "base64.hh"
#include "conditions.hh"
#include "alternator/expressionsLexer.hpp"
#include "alternator/expressionsParser.hpp"
#include "utils/overloaded_functor.hh"
#include "error.hh"
#include "seastarx.hh"
#include <seastar/core/print.hh>
#include <seastar/util/log.hh>
#include <boost/algorithm/cxx11/any_of.hpp>
#include <boost/algorithm/cxx11/all_of.hpp>
#include <functional>
#include <unordered_map>
namespace alternator {
template <typename Func, typename Result = std::result_of_t<Func(expressionsParser&)>>
Result do_with_parser(std::string input, Func&& f) {
expressionsLexer::InputStreamType input_stream{
reinterpret_cast<const ANTLR_UINT8*>(input.data()),
ANTLR_ENC_UTF8,
static_cast<ANTLR_UINT32>(input.size()),
nullptr };
expressionsLexer lexer(&input_stream);
expressionsParser::TokenStreamType tstream(ANTLR_SIZE_HINT, lexer.get_tokSource());
expressionsParser parser(&tstream);
auto result = f(parser);
return result;
}
parsed::update_expression
parse_update_expression(std::string query) {
try {
return do_with_parser(query, std::mem_fn(&expressionsParser::update_expression));
} catch (...) {
throw expressions_syntax_error(format("Failed parsing UpdateExpression '{}': {}", query, std::current_exception()));
}
}
std::vector<parsed::path>
parse_projection_expression(std::string query) {
try {
return do_with_parser(query, std::mem_fn(&expressionsParser::projection_expression));
} catch (...) {
throw expressions_syntax_error(format("Failed parsing ProjectionExpression '{}': {}", query, std::current_exception()));
}
}
parsed::condition_expression
parse_condition_expression(std::string query) {
try {
return do_with_parser(query, std::mem_fn(&expressionsParser::condition_expression));
} catch (...) {
throw expressions_syntax_error(format("Failed parsing ConditionExpression '{}': {}", query, std::current_exception()));
}
}
namespace parsed {
void update_expression::add(update_expression::action a) {
std::visit(overloaded_functor {
[&] (action::set&) { seen_set = true; },
[&] (action::remove&) { seen_remove = true; },
[&] (action::add&) { seen_add = true; },
[&] (action::del&) { seen_del = true; }
}, a._action);
_actions.push_back(std::move(a));
}
void update_expression::append(update_expression other) {
if ((seen_set && other.seen_set) ||
(seen_remove && other.seen_remove) ||
(seen_add && other.seen_add) ||
(seen_del && other.seen_del)) {
throw expressions_syntax_error("Each of SET, REMOVE, ADD, DELETE may only appear once in UpdateExpression");
}
std::move(other._actions.begin(), other._actions.end(), std::back_inserter(_actions));
seen_set |= other.seen_set;
seen_remove |= other.seen_remove;
seen_add |= other.seen_add;
seen_del |= other.seen_del;
}
void condition_expression::append(condition_expression&& a, char op) {
std::visit(overloaded_functor {
[&] (condition_list& x) {
// If 'a' has a single condition, we could, instead of inserting
// it insert its single condition (possibly negated if a._negated)
// But considering it we don't evaluate these expressions many
// times, this optimization is not worth extra code complexity.
if (!x.conditions.empty() && x.op != op) {
// Shouldn't happen unless we have a bug in the parser
throw std::logic_error("condition_expression::append called with mixed operators");
}
x.conditions.push_back(std::move(a));
x.op = op;
},
[&] (primitive_condition& x) {
// Shouldn't happen unless we have a bug in the parser
throw std::logic_error("condition_expression::append called on primitive_condition");
}
}, _expression);
}
} // namespace parsed
// The following resolve_*() functions resolve references in parsed
// expressions of different types. Resolving a parsed expression means
// replacing:
// 1. In parsed::path objects, replace references like "#name" with the
// attribute name from ExpressionAttributeNames,
// 2. In parsed::constant objects, replace references like ":value" with
// the value from ExpressionAttributeValues.
// These function also track which name and value references were used, to
// allow complaining if some remain unused.
// Note that the resolve_*() functions modify the expressions in-place,
// so if we ever intend to cache parsed expression, we need to pass a copy
// into this function.
//
// Doing the "resolving" stage before the evaluation stage has two benefits.
// First, it allows us to be compatible with DynamoDB in catching unused
// names and values (see issue #6572). Second, in the FilterExpression case,
// we need to resolve the expression just once but then use it many times
// (once for each item to be filtered).
static void resolve_path(parsed::path& p,
const rjson::value* expression_attribute_names,
std::unordered_set<std::string>& used_attribute_names) {
const std::string& column_name = p.root();
if (column_name.size() > 0 && column_name.front() == '#') {
if (!expression_attribute_names) {
throw api_error::validation(
format("ExpressionAttributeNames missing, entry '{}' required by expression", column_name));
}
const rjson::value* value = rjson::find(*expression_attribute_names, column_name);
if (!value || !value->IsString()) {
throw api_error::validation(
format("ExpressionAttributeNames missing entry '{}' required by expression", column_name));
}
used_attribute_names.emplace(column_name);
p.set_root(std::string(rjson::to_string_view(*value)));
}
}
static void resolve_constant(parsed::constant& c,
const rjson::value* expression_attribute_values,
std::unordered_set<std::string>& used_attribute_values) {
std::visit(overloaded_functor {
[&] (const std::string& valref) {
if (!expression_attribute_values) {
throw api_error::validation(
format("ExpressionAttributeValues missing, entry '{}' required by expression", valref));
}
const rjson::value* value = rjson::find(*expression_attribute_values, valref);
if (!value) {
throw api_error::validation(
format("ExpressionAttributeValues missing entry '{}' required by expression", valref));
}
if (value->IsNull()) {
throw api_error::validation(
format("ExpressionAttributeValues null value for entry '{}' required by expression", valref));
}
validate_value(*value, "ExpressionAttributeValues");
used_attribute_values.emplace(valref);
c.set(*value);
},
[&] (const parsed::constant::literal& lit) {
// Nothing to do, already resolved
}
}, c._value);
}
void resolve_value(parsed::value& rhs,
const rjson::value* expression_attribute_names,
const rjson::value* expression_attribute_values,
std::unordered_set<std::string>& used_attribute_names,
std::unordered_set<std::string>& used_attribute_values) {
std::visit(overloaded_functor {
[&] (parsed::constant& c) {
resolve_constant(c, expression_attribute_values, used_attribute_values);
},
[&] (parsed::value::function_call& f) {
for (parsed::value& value : f._parameters) {
resolve_value(value, expression_attribute_names, expression_attribute_values,
used_attribute_names, used_attribute_values);
}
},
[&] (parsed::path& p) {
resolve_path(p, expression_attribute_names, used_attribute_names);
}
}, rhs._value);
}
void resolve_set_rhs(parsed::set_rhs& rhs,
const rjson::value* expression_attribute_names,
const rjson::value* expression_attribute_values,
std::unordered_set<std::string>& used_attribute_names,
std::unordered_set<std::string>& used_attribute_values) {
resolve_value(rhs._v1, expression_attribute_names, expression_attribute_values,
used_attribute_names, used_attribute_values);
if (rhs._op != 'v') {
resolve_value(rhs._v2, expression_attribute_names, expression_attribute_values,
used_attribute_names, used_attribute_values);
}
}
void resolve_update_expression(parsed::update_expression& ue,
const rjson::value* expression_attribute_names,
const rjson::value* expression_attribute_values,
std::unordered_set<std::string>& used_attribute_names,
std::unordered_set<std::string>& used_attribute_values) {
for (parsed::update_expression::action& action : ue.actions()) {
resolve_path(action._path, expression_attribute_names, used_attribute_names);
std::visit(overloaded_functor {
[&] (parsed::update_expression::action::set& a) {
resolve_set_rhs(a._rhs, expression_attribute_names, expression_attribute_values,
used_attribute_names, used_attribute_values);
},
[&] (parsed::update_expression::action::remove& a) {
// nothing to do
},
[&] (parsed::update_expression::action::add& a) {
resolve_constant(a._valref, expression_attribute_values, used_attribute_values);
},
[&] (parsed::update_expression::action::del& a) {
resolve_constant(a._valref, expression_attribute_values, used_attribute_values);
}
}, action._action);
}
}
static void resolve_primitive_condition(parsed::primitive_condition& pc,
const rjson::value* expression_attribute_names,
const rjson::value* expression_attribute_values,
std::unordered_set<std::string>& used_attribute_names,
std::unordered_set<std::string>& used_attribute_values) {
for (parsed::value& value : pc._values) {
resolve_value(value,
expression_attribute_names, expression_attribute_values,
used_attribute_names, used_attribute_values);
}
}
void resolve_condition_expression(parsed::condition_expression& ce,
const rjson::value* expression_attribute_names,
const rjson::value* expression_attribute_values,
std::unordered_set<std::string>& used_attribute_names,
std::unordered_set<std::string>& used_attribute_values) {
std::visit(overloaded_functor {
[&] (parsed::primitive_condition& cond) {
resolve_primitive_condition(cond,
expression_attribute_names, expression_attribute_values,
used_attribute_names, used_attribute_values);
},
[&] (parsed::condition_expression::condition_list& list) {
for (parsed::condition_expression& cond : list.conditions) {
resolve_condition_expression(cond,
expression_attribute_names, expression_attribute_values,
used_attribute_names, used_attribute_values);
}
}
}, ce._expression);
}
void resolve_projection_expression(std::vector<parsed::path>& pe,
const rjson::value* expression_attribute_names,
std::unordered_set<std::string>& used_attribute_names) {
for (parsed::path& p : pe) {
resolve_path(p, expression_attribute_names, used_attribute_names);
}
}
// condition_expression_on() checks whether a condition_expression places any
// condition on the given attribute. It can be useful, for example, for
// checking whether the condition tries to restrict a key column.
static bool value_on(const parsed::value& v, std::string_view attribute) {
return std::visit(overloaded_functor {
[&] (const parsed::constant& c) {
return false;
},
[&] (const parsed::value::function_call& f) {
for (const parsed::value& value : f._parameters) {
if (value_on(value, attribute)) {
return true;
}
}
return false;
},
[&] (const parsed::path& p) {
return p.root() == attribute;
}
}, v._value);
}
static bool primitive_condition_on(const parsed::primitive_condition& pc, std::string_view attribute) {
for (const parsed::value& value : pc._values) {
if (value_on(value, attribute)) {
return true;
}
}
return false;
}
bool condition_expression_on(const parsed::condition_expression& ce, std::string_view attribute) {
return std::visit(overloaded_functor {
[&] (const parsed::primitive_condition& cond) {
return primitive_condition_on(cond, attribute);
},
[&] (const parsed::condition_expression::condition_list& list) {
for (const parsed::condition_expression& cond : list.conditions) {
if (condition_expression_on(cond, attribute)) {
return true;
}
}
return false;
}
}, ce._expression);
}
// for_condition_expression_on() runs a given function over all the attributes
// mentioned in the expression. If the same attribute is mentioned more than
// once, the function will be called more than once for the same attribute.
static void for_value_on(const parsed::value& v, const noncopyable_function<void(std::string_view)>& func) {
std::visit(overloaded_functor {
[&] (const parsed::constant& c) { },
[&] (const parsed::value::function_call& f) {
for (const parsed::value& value : f._parameters) {
for_value_on(value, func);
}
},
[&] (const parsed::path& p) {
func(p.root());
}
}, v._value);
}
void for_condition_expression_on(const parsed::condition_expression& ce, const noncopyable_function<void(std::string_view)>& func) {
std::visit(overloaded_functor {
[&] (const parsed::primitive_condition& cond) {
for (const parsed::value& value : cond._values) {
for_value_on(value, func);
}
},
[&] (const parsed::condition_expression::condition_list& list) {
for (const parsed::condition_expression& cond : list.conditions) {
for_condition_expression_on(cond, func);
}
}
}, ce._expression);
}
// The following calculate_value() functions calculate, or evaluate, a parsed
// expression. The parsed expression is assumed to have been "resolved", with
// the matching resolve_* function.
// Take two JSON-encoded list values (remember that a list value is
// {"L": [...the actual list]}) and return the concatenation, again as
// a list value.
static rjson::value list_concatenate(const rjson::value& v1, const rjson::value& v2) {
const rjson::value* list1 = unwrap_list(v1);
const rjson::value* list2 = unwrap_list(v2);
if (!list1 || !list2) {
throw api_error::validation("UpdateExpression: list_append() given a non-list");
}
rjson::value cat = rjson::copy(*list1);
for (const auto& a : list2->GetArray()) {
rjson::push_back(cat, rjson::copy(a));
}
rjson::value ret = rjson::empty_object();
rjson::set(ret, "L", std::move(cat));
return ret;
}
// calculate_size() is ConditionExpression's size() function, i.e., it takes
// a JSON-encoded value and returns its "size" as defined differently for the
// different types - also as a JSON-encoded number.
// It return a JSON-encoded "null" value if this value's type has no size
// defined. Comparisons against this non-numeric value will later fail.
static rjson::value calculate_size(const rjson::value& v) {
// NOTE: If v is improperly formatted for our JSON value encoding, it
// must come from the request itself, not from the database, so it makes
// sense to throw a ValidationException if we see such a problem.
if (!v.IsObject() || v.MemberCount() != 1) {
throw api_error::validation(format("invalid object: {}", v));
}
auto it = v.MemberBegin();
int ret;
if (it->name == "S") {
if (!it->value.IsString()) {
throw api_error::validation(format("invalid string: {}", v));
}
ret = it->value.GetStringLength();
} else if (it->name == "NS" || it->name == "SS" || it->name == "BS" || it->name == "L") {
if (!it->value.IsArray()) {
throw api_error::validation(format("invalid set: {}", v));
}
ret = it->value.Size();
} else if (it->name == "M") {
if (!it->value.IsObject()) {
throw api_error::validation(format("invalid map: {}", v));
}
ret = it->value.MemberCount();
} else if (it->name == "B") {
if (!it->value.IsString()) {
throw api_error::validation(format("invalid byte string: {}", v));
}
ret = base64_decoded_len(rjson::to_string_view(it->value));
} else {
rjson::value json_ret = rjson::empty_object();
rjson::set(json_ret, "null", rjson::value(true));
return json_ret;
}
rjson::value json_ret = rjson::empty_object();
rjson::set(json_ret, "N", rjson::from_string(std::to_string(ret)));
return json_ret;
}
static const rjson::value& calculate_value(const parsed::constant& c) {
return std::visit(overloaded_functor {
[&] (const parsed::constant::literal& v) -> const rjson::value& {
return *v;
},
[&] (const std::string& valref) -> const rjson::value& {
// Shouldn't happen, we should have called resolve_value() earlier
// and replaced the value reference by the literal constant.
throw std::logic_error("calculate_value() called before resolve_value()");
}
}, c._value);
}
static rjson::value to_bool_json(bool b) {
rjson::value json_ret = rjson::empty_object();
rjson::set(json_ret, "BOOL", rjson::value(b));
return json_ret;
}
static bool known_type(std::string_view type) {
static thread_local const std::unordered_set<std::string_view> types = {
"N", "S", "B", "NS", "SS", "BS", "L", "M", "NULL", "BOOL"
};
return types.contains(type);
}
using function_handler_type = rjson::value(calculate_value_caller, const rjson::value*, const parsed::value::function_call&);
static const
std::unordered_map<std::string_view, function_handler_type*> function_handlers {
{"list_append", [] (calculate_value_caller caller, const rjson::value* previous_item, const parsed::value::function_call& f) {
if (caller != calculate_value_caller::UpdateExpression) {
throw api_error::validation(
format("{}: list_append() not allowed here", caller));
}
if (f._parameters.size() != 2) {
throw api_error::validation(
format("{}: list_append() accepts 2 parameters, got {}", caller, f._parameters.size()));
}
rjson::value v1 = calculate_value(f._parameters[0], caller, previous_item);
rjson::value v2 = calculate_value(f._parameters[1], caller, previous_item);
return list_concatenate(v1, v2);
}
},
{"if_not_exists", [] (calculate_value_caller caller, const rjson::value* previous_item, const parsed::value::function_call& f) {
if (caller != calculate_value_caller::UpdateExpression) {
throw api_error::validation(
format("{}: if_not_exists() not allowed here", caller));
}
if (f._parameters.size() != 2) {
throw api_error::validation(
format("{}: if_not_exists() accepts 2 parameters, got {}", caller, f._parameters.size()));
}
if (!std::holds_alternative<parsed::path>(f._parameters[0]._value)) {
throw api_error::validation(
format("{}: if_not_exists() must include path as its first argument", caller));
}
rjson::value v1 = calculate_value(f._parameters[0], caller, previous_item);
rjson::value v2 = calculate_value(f._parameters[1], caller, previous_item);
return v1.IsNull() ? std::move(v2) : std::move(v1);
}
},
{"size", [] (calculate_value_caller caller, const rjson::value* previous_item, const parsed::value::function_call& f) {
if (caller != calculate_value_caller::ConditionExpression) {
throw api_error::validation(
format("{}: size() not allowed here", caller));
}
if (f._parameters.size() != 1) {
throw api_error::validation(
format("{}: size() accepts 1 parameter, got {}", caller, f._parameters.size()));
}
rjson::value v = calculate_value(f._parameters[0], caller, previous_item);
return calculate_size(v);
}
},
{"attribute_exists", [] (calculate_value_caller caller, const rjson::value* previous_item, const parsed::value::function_call& f) {
if (caller != calculate_value_caller::ConditionExpressionAlone) {
throw api_error::validation(
format("{}: attribute_exists() not allowed here", caller));
}
if (f._parameters.size() != 1) {
throw api_error::validation(
format("{}: attribute_exists() accepts 1 parameter, got {}", caller, f._parameters.size()));
}
if (!std::holds_alternative<parsed::path>(f._parameters[0]._value)) {
throw api_error::validation(
format("{}: attribute_exists()'s parameter must be a path", caller));
}
rjson::value v = calculate_value(f._parameters[0], caller, previous_item);
return to_bool_json(!v.IsNull());
}
},
{"attribute_not_exists", [] (calculate_value_caller caller, const rjson::value* previous_item, const parsed::value::function_call& f) {
if (caller != calculate_value_caller::ConditionExpressionAlone) {
throw api_error::validation(
format("{}: attribute_not_exists() not allowed here", caller));
}
if (f._parameters.size() != 1) {
throw api_error::validation(
format("{}: attribute_not_exists() accepts 1 parameter, got {}", caller, f._parameters.size()));
}
if (!std::holds_alternative<parsed::path>(f._parameters[0]._value)) {
throw api_error::validation(
format("{}: attribute_not_exists()'s parameter must be a path", caller));
}
rjson::value v = calculate_value(f._parameters[0], caller, previous_item);
return to_bool_json(v.IsNull());
}
},
{"attribute_type", [] (calculate_value_caller caller, const rjson::value* previous_item, const parsed::value::function_call& f) {
if (caller != calculate_value_caller::ConditionExpressionAlone) {
throw api_error::validation(
format("{}: attribute_type() not allowed here", caller));
}
if (f._parameters.size() != 2) {
throw api_error::validation(
format("{}: attribute_type() accepts 2 parameters, got {}", caller, f._parameters.size()));
}
// There is no real reason for the following check (not
// allowing the type to come from a document attribute), but
// DynamoDB does this check, so we do too...
if (!f._parameters[1].is_constant()) {
throw api_error::validation(
format("{}: attribute_types()'s first parameter must be an expression attribute", caller));
}
rjson::value v0 = calculate_value(f._parameters[0], caller, previous_item);
rjson::value v1 = calculate_value(f._parameters[1], caller, previous_item);
if (v1.IsObject() && v1.MemberCount() == 1 && v1.MemberBegin()->name == "S") {
// If the type parameter is not one of the legal types
// we should generate an error, not a failed condition:
if (!known_type(rjson::to_string_view(v1.MemberBegin()->value))) {
throw api_error::validation(
format("{}: attribute_types()'s second parameter, {}, is not a known type",
caller, v1.MemberBegin()->value));
}
if (v0.IsObject() && v0.MemberCount() == 1) {
return to_bool_json(v1.MemberBegin()->value == v0.MemberBegin()->name);
} else {
return to_bool_json(false);
}
} else {
throw api_error::validation(
format("{}: attribute_type() second parameter must refer to a string, got {}", caller, v1));
}
}
},
{"begins_with", [] (calculate_value_caller caller, const rjson::value* previous_item, const parsed::value::function_call& f) {
if (caller != calculate_value_caller::ConditionExpressionAlone) {
throw api_error::validation(
format("{}: begins_with() not allowed here", caller));
}
if (f._parameters.size() != 2) {
throw api_error::validation(
format("{}: begins_with() accepts 2 parameters, got {}", caller, f._parameters.size()));
}
rjson::value v1 = calculate_value(f._parameters[0], caller, previous_item);
rjson::value v2 = calculate_value(f._parameters[1], caller, previous_item);
// TODO: There's duplication here with check_BEGINS_WITH().
// But unfortunately, the two functions differ a bit.
// If one of v1 or v2 is malformed or has an unsupported type
// (not B or S), what we do depends on whether it came from
// the user's query (is_constant()), or the item. Unsupported
// values in the query result in an error, but if they are in
// the item, we silently return false (no match).
bool bad = false;
if (!v1.IsObject() || v1.MemberCount() != 1) {
bad = true;
if (f._parameters[0].is_constant()) {
throw api_error::validation(format("{}: begins_with() encountered malformed AttributeValue: {}", caller, v1));
}
} else if (v1.MemberBegin()->name != "S" && v1.MemberBegin()->name != "B") {
bad = true;
if (f._parameters[0].is_constant()) {
throw api_error::validation(format("{}: begins_with() supports only string or binary in AttributeValue: {}", caller, v1));
}
}
if (!v2.IsObject() || v2.MemberCount() != 1) {
bad = true;
if (f._parameters[1].is_constant()) {
throw api_error::validation(format("{}: begins_with() encountered malformed AttributeValue: {}", caller, v2));
}
} else if (v2.MemberBegin()->name != "S" && v2.MemberBegin()->name != "B") {
bad = true;
if (f._parameters[1].is_constant()) {
throw api_error::validation(format("{}: begins_with() supports only string or binary in AttributeValue: {}", caller, v2));
}
}
bool ret = false;
if (!bad) {
auto it1 = v1.MemberBegin();
auto it2 = v2.MemberBegin();
if (it1->name == it2->name) {
if (it2->name == "S") {
std::string_view val1 = rjson::to_string_view(it1->value);
std::string_view val2 = rjson::to_string_view(it2->value);
ret = val1.starts_with(val2);
} else /* it2->name == "B" */ {
ret = base64_begins_with(rjson::to_string_view(it1->value), rjson::to_string_view(it2->value));
}
}
}
return to_bool_json(ret);
}
},
{"contains", [] (calculate_value_caller caller, const rjson::value* previous_item, const parsed::value::function_call& f) {
if (caller != calculate_value_caller::ConditionExpressionAlone) {
throw api_error::validation(
format("{}: contains() not allowed here", caller));
}
if (f._parameters.size() != 2) {
throw api_error::validation(
format("{}: contains() accepts 2 parameters, got {}", caller, f._parameters.size()));
}
rjson::value v1 = calculate_value(f._parameters[0], caller, previous_item);
rjson::value v2 = calculate_value(f._parameters[1], caller, previous_item);
return to_bool_json(check_CONTAINS(v1.IsNull() ? nullptr : &v1, v2));
}
},
};
// Given a parsed::value, which can refer either to a constant value from
// ExpressionAttributeValues, to the value of some attribute, or to a function
// of other values, this function calculates the resulting value.
// "caller" determines which expression - ConditionExpression or
// UpdateExpression - is asking for this value. We need to know this because
// DynamoDB allows a different choice of functions for different expressions.
rjson::value calculate_value(const parsed::value& v,
calculate_value_caller caller,
const rjson::value* previous_item) {
return std::visit(overloaded_functor {
[&] (const parsed::constant& c) -> rjson::value {
return rjson::copy(calculate_value(c));
},
[&] (const parsed::value::function_call& f) -> rjson::value {
auto function_it = function_handlers.find(std::string_view(f._function_name));
if (function_it == function_handlers.end()) {
throw api_error::validation(
format("UpdateExpression: unknown function '{}' called.", f._function_name));
}
return function_it->second(caller, previous_item, f);
},
[&] (const parsed::path& p) -> rjson::value {
if (!previous_item) {
return rjson::null_value();
}
std::string update_path = p.root();
if (p.has_operators()) {
// FIXME: support this
throw api_error::validation("Reading attribute paths not yet implemented");
}
const rjson::value* previous_value = rjson::find(*previous_item, update_path);
return previous_value ? rjson::copy(*previous_value) : rjson::null_value();
}
}, v._value);
}
// Same as calculate_value() above, except takes a set_rhs, which may be
// either a single value, or v1+v2 or v1-v2.
rjson::value calculate_value(const parsed::set_rhs& rhs,
const rjson::value* previous_item) {
switch (rhs._op) {
case 'v':
return calculate_value(rhs._v1, calculate_value_caller::UpdateExpression, previous_item);
case '+': {
rjson::value v1 = calculate_value(rhs._v1, calculate_value_caller::UpdateExpression, previous_item);
rjson::value v2 = calculate_value(rhs._v2, calculate_value_caller::UpdateExpression, previous_item);
return number_add(v1, v2);
}
case '-': {
rjson::value v1 = calculate_value(rhs._v1, calculate_value_caller::UpdateExpression, previous_item);
rjson::value v2 = calculate_value(rhs._v2, calculate_value_caller::UpdateExpression, previous_item);
return number_subtract(v1, v2);
}
}
// Can't happen
return rjson::null_value();
}
} // namespace alternator

View File

@@ -1,265 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*
* This file is part of Scylla. See the LICENSE.PROPRIETARY file in the
* top-level directory for licensing information.
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
/*
* The DynamoDB protocol is based on JSON, and most DynamoDB requests
* describe the operation and its parameters via JSON objects such as maps
* and lists. Nevertheless, in some types of requests an "expression" is
* passed as a single string, and we need to parse this string. These
* cases include:
* 1. Attribute paths, such as "a[3].b.c", are used in projection
* expressions as well as inside other expressions described below.
* 2. Condition expressions, such as "(NOT (a=b OR c=d)) AND e=f",
* used in conditional updates, filters, and other places.
* 3. Update expressions, such as "SET #a.b = :x, c = :y DELETE d"
*
* All these expression syntaxes are very simple: Most of them could be
* parsed as regular expressions, and the parenthesized condition expression
* could be done with a simple hand-written lexical analyzer and recursive-
* descent parser. Nevertheless, we decided to specify these parsers in the
* ANTLR3 language already used in the Scylla project, hopefully making these
* parsers easier to reason about, and easier to change if needed - and
* reducing the amount of boiler-plate code.
*/
grammar expressions;
options {
language = Cpp;
}
@parser::namespace{alternator}
@lexer::namespace{alternator}
/* TODO: explain what these traits things are. I haven't seen them explained
* in any document... Compilation fails without these fail because a definition
* of "expressionsLexerTraits" and "expressionParserTraits" is needed.
*/
@lexer::traits {
class expressionsLexer;
class expressionsParser;
typedef antlr3::Traits<expressionsLexer, expressionsParser> expressionsLexerTraits;
}
@parser::traits {
typedef expressionsLexerTraits expressionsParserTraits;
}
@lexer::header {
#include "alternator/expressions.hh"
// ANTLR generates a bunch of unused variables and functions. Yuck...
#pragma GCC diagnostic ignored "-Wunused-variable"
#pragma GCC diagnostic ignored "-Wunused-function"
}
@parser::header {
#include "expressionsLexer.hpp"
}
/* By default, ANTLR3 composes elaborate syntax-error messages, saying which
* token was unexpected, where, and so on on, but then dutifully writes these
* error messages to the standard error, and returns from the parser as if
* everything was fine, with a half-constructed output object! If we define
* the "displayRecognitionError" method, it will be called upon to build this
* error message, and we can instead throw an exception to stop the parsing
* immediately. This is good enough for now, for our simple needs, but if
* we ever want to show more information about the syntax error, Cql3.g
* contains an elaborate implementation (it would be nice if we could reuse
* it, not duplicate it).
* Unfortunately, we have to repeat the same definition twice - once for the
* parser, and once for the lexer.
*/
@parser::context {
void displayRecognitionError(ANTLR_UINT8** token_names, ExceptionBaseType* ex) {
throw expressions_syntax_error("syntax error");
}
}
@lexer::context {
void displayRecognitionError(ANTLR_UINT8** token_names, ExceptionBaseType* ex) {
throw expressions_syntax_error("syntax error");
}
}
/*
* Lexical analysis phase, i.e., splitting the input up to tokens.
* Lexical analyzer rules have names starting in capital letters.
* "fragment" rules do not generate tokens, and are just aliases used to
* make other rules more readable.
* Characters *not* listed here, e.g., '=', '(', etc., will be handled
* as individual tokens on their own right.
* Whitespace spans are skipped, so do not generate tokens.
*/
WHITESPACE: (' ' | '\t' | '\n' | '\r')+ { skip(); };
/* shortcuts for case-insensitive keywords */
fragment A:('a'|'A');
fragment B:('b'|'B');
fragment C:('c'|'C');
fragment D:('d'|'D');
fragment E:('e'|'E');
fragment F:('f'|'F');
fragment G:('g'|'G');
fragment H:('h'|'H');
fragment I:('i'|'I');
fragment J:('j'|'J');
fragment K:('k'|'K');
fragment L:('l'|'L');
fragment M:('m'|'M');
fragment N:('n'|'N');
fragment O:('o'|'O');
fragment P:('p'|'P');
fragment Q:('q'|'Q');
fragment R:('r'|'R');
fragment S:('s'|'S');
fragment T:('t'|'T');
fragment U:('u'|'U');
fragment V:('v'|'V');
fragment W:('w'|'W');
fragment X:('x'|'X');
fragment Y:('y'|'Y');
fragment Z:('z'|'Z');
/* These keywords must be appear before the generic NAME token below,
* because NAME matches too, and the first to match wins.
*/
SET: S E T;
REMOVE: R E M O V E;
ADD: A D D;
DELETE: D E L E T E;
AND: A N D;
OR: O R;
NOT: N O T;
BETWEEN: B E T W E E N;
IN: I N;
fragment ALPHA: 'A'..'Z' | 'a'..'z';
fragment DIGIT: '0'..'9';
fragment ALNUM: ALPHA | DIGIT | '_';
INTEGER: DIGIT+;
NAME: ALPHA ALNUM*;
NAMEREF: '#' ALNUM+;
VALREF: ':' ALNUM+;
/*
* Parsing phase - parsing the string of tokens generated by the lexical
* analyzer defined above.
*/
path_component: NAME | NAMEREF;
path returns [parsed::path p]:
root=path_component { $p.set_root($root.text); }
( '.' name=path_component { $p.add_dot($name.text); }
| '[' INTEGER ']' { $p.add_index(std::stoi($INTEGER.text)); }
)*;
value returns [parsed::value v]:
VALREF { $v.set_valref($VALREF.text); }
| path { $v.set_path($path.p); }
| NAME { $v.set_func_name($NAME.text); }
'(' x=value { $v.add_func_parameter($x.v); }
(',' x=value { $v.add_func_parameter($x.v); })*
')'
;
update_expression_set_rhs returns [parsed::set_rhs rhs]:
v=value { $rhs.set_value(std::move($v.v)); }
( '+' v=value { $rhs.set_plus(std::move($v.v)); }
| '-' v=value { $rhs.set_minus(std::move($v.v)); }
)?
;
update_expression_set_action returns [parsed::update_expression::action a]:
path '=' rhs=update_expression_set_rhs { $a.assign_set($path.p, $rhs.rhs); };
update_expression_remove_action returns [parsed::update_expression::action a]:
path { $a.assign_remove($path.p); };
update_expression_add_action returns [parsed::update_expression::action a]:
path VALREF { $a.assign_add($path.p, $VALREF.text); };
update_expression_delete_action returns [parsed::update_expression::action a]:
path VALREF { $a.assign_del($path.p, $VALREF.text); };
update_expression_clause returns [parsed::update_expression e]:
SET s=update_expression_set_action { $e.add(s); }
(',' s=update_expression_set_action { $e.add(s); })*
| REMOVE r=update_expression_remove_action { $e.add(r); }
(',' r=update_expression_remove_action { $e.add(r); })*
| ADD a=update_expression_add_action { $e.add(a); }
(',' a=update_expression_add_action { $e.add(a); })*
| DELETE d=update_expression_delete_action { $e.add(d); }
(',' d=update_expression_delete_action { $e.add(d); })*
;
// Note the "EOF" token at the end of the update expression. We want to the
// parser to match the entire string given to it - not just its beginning!
update_expression returns [parsed::update_expression e]:
(update_expression_clause { e.append($update_expression_clause.e); })* EOF;
projection_expression returns [std::vector<parsed::path> v]:
p=path { $v.push_back(std::move($p.p)); }
(',' p=path { $v.push_back(std::move($p.p)); } )* EOF;
primitive_condition returns [parsed::primitive_condition c]:
v=value { $c.add_value(std::move($v.v));
$c.set_operator(parsed::primitive_condition::type::VALUE); }
( ( '=' { $c.set_operator(parsed::primitive_condition::type::EQ); }
| '<' '>' { $c.set_operator(parsed::primitive_condition::type::NE); }
| '<' { $c.set_operator(parsed::primitive_condition::type::LT); }
| '<' '=' { $c.set_operator(parsed::primitive_condition::type::LE); }
| '>' { $c.set_operator(parsed::primitive_condition::type::GT); }
| '>' '=' { $c.set_operator(parsed::primitive_condition::type::GE); }
)
v=value { $c.add_value(std::move($v.v)); }
| BETWEEN { $c.set_operator(parsed::primitive_condition::type::BETWEEN); }
v=value { $c.add_value(std::move($v.v)); }
AND
v=value { $c.add_value(std::move($v.v)); }
| IN '(' { $c.set_operator(parsed::primitive_condition::type::IN); }
v=value { $c.add_value(std::move($v.v)); }
(',' v=value { $c.add_value(std::move($v.v)); })*
')'
)?
;
// The following rules for parsing boolean expressions are verbose and
// somewhat strange because of Antlr 3's limitations on recursive rules,
// common rule prefixes, and (lack of) support for operator precedence.
// These rules could have been written more clearly using a more powerful
// parser generator - such as Yacc.
boolean_expression returns [parsed::condition_expression e]:
b=boolean_expression_1 { $e.append(std::move($b.e), '|'); }
(OR b=boolean_expression_1 { $e.append(std::move($b.e), '|'); } )*
;
boolean_expression_1 returns [parsed::condition_expression e]:
b=boolean_expression_2 { $e.append(std::move($b.e), '&'); }
(AND b=boolean_expression_2 { $e.append(std::move($b.e), '&'); } )*
;
boolean_expression_2 returns [parsed::condition_expression e]:
p=primitive_condition { $e.set_primitive(std::move($p.c)); }
| NOT b=boolean_expression_2 { $e = std::move($b.e); $e.apply_not(); }
| '(' b=boolean_expression ')' { $e = std::move($b.e); }
;
condition_expression returns [parsed::condition_expression e]:
boolean_expression { e=std::move($boolean_expression.e); } EOF;

View File

@@ -1,102 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include <string>
#include <stdexcept>
#include <vector>
#include <unordered_set>
#include <string_view>
#include <seastar/util/noncopyable_function.hh>
#include "expressions_types.hh"
#include "utils/rjson.hh"
namespace alternator {
class expressions_syntax_error : public std::runtime_error {
public:
using runtime_error::runtime_error;
};
parsed::update_expression parse_update_expression(std::string query);
std::vector<parsed::path> parse_projection_expression(std::string query);
parsed::condition_expression parse_condition_expression(std::string query);
void resolve_update_expression(parsed::update_expression& ue,
const rjson::value* expression_attribute_names,
const rjson::value* expression_attribute_values,
std::unordered_set<std::string>& used_attribute_names,
std::unordered_set<std::string>& used_attribute_values);
void resolve_projection_expression(std::vector<parsed::path>& pe,
const rjson::value* expression_attribute_names,
std::unordered_set<std::string>& used_attribute_names);
void resolve_condition_expression(parsed::condition_expression& ce,
const rjson::value* expression_attribute_names,
const rjson::value* expression_attribute_values,
std::unordered_set<std::string>& used_attribute_names,
std::unordered_set<std::string>& used_attribute_values);
void validate_value(const rjson::value& v, const char* caller);
bool condition_expression_on(const parsed::condition_expression& ce, std::string_view attribute);
// for_condition_expression_on() runs the given function on the attributes
// that the expression uses. It may run for the same attribute more than once
// if the same attribute is used more than once in the expression.
void for_condition_expression_on(const parsed::condition_expression& ce, const noncopyable_function<void(std::string_view)>& func);
// calculate_value() behaves slightly different (especially, different
// functions supported) when used in different types of expressions, as
// enumerated in this enum:
enum class calculate_value_caller {
UpdateExpression, ConditionExpression, ConditionExpressionAlone
};
inline std::ostream& operator<<(std::ostream& out, calculate_value_caller caller) {
switch (caller) {
case calculate_value_caller::UpdateExpression:
out << "UpdateExpression";
break;
case calculate_value_caller::ConditionExpression:
out << "ConditionExpression";
break;
case calculate_value_caller::ConditionExpressionAlone:
out << "ConditionExpression";
break;
default:
out << "unknown type of expression";
break;
}
return out;
}
rjson::value calculate_value(const parsed::value& v,
calculate_value_caller caller,
const rjson::value* previous_item);
rjson::value calculate_value(const parsed::set_rhs& rhs,
const rjson::value* previous_item);
} /* namespace alternator */

View File

@@ -1,255 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include <vector>
#include <string>
#include <variant>
#include <seastar/core/shared_ptr.hh>
#include "utils/rjson.hh"
/*
* Parsed representation of expressions and their components.
*
* Types in alternator::parse namespace are used for holding the parse
* tree - objects generated by the Antlr rules after parsing an expression.
* Because of the way Antlr works, all these objects are default-constructed
* first, and then assigned when the rule is completed, so all these types
* have only default constructors - but setter functions to set them later.
*/
namespace alternator {
namespace parsed {
// "path" is an attribute's path in a document, e.g., a.b[3].c.
class path {
// All paths have a "root", a top-level attribute, and any number of
// "dereference operators" - each either an index (e.g., "[2]") or a
// dot (e.g., ".xyz").
std::string _root;
std::vector<std::variant<std::string, unsigned>> _operators;
public:
void set_root(std::string root) {
_root = std::move(root);
}
void add_index(unsigned i) {
_operators.emplace_back(i);
}
void add_dot(std::string(name)) {
_operators.emplace_back(std::move(name));
}
const std::string& root() const {
return _root;
}
bool has_operators() const {
return !_operators.empty();
}
};
// When an expression is first parsed, all constants are references, like
// ":val1", into ExpressionAttributeValues. This uses std::string() variant.
// The resolve_value() function replaces these constants by the JSON item
// extracted from the ExpressionAttributeValues.
struct constant {
// We use lw_shared_ptr<rjson::value> just to make rjson::value copyable,
// to make this entire object copyable as ANTLR needs.
using literal = lw_shared_ptr<rjson::value>;
std::variant<std::string, literal> _value;
void set(const rjson::value& v) {
_value = make_lw_shared<rjson::value>(rjson::copy(v));
}
void set(std::string& s) {
_value = s;
}
};
// "value" is is a value used in the right hand side of an assignment
// expression, "SET a = ...". It can be a constant (a reference to a value
// included in the request, e.g., ":val"), a path to an attribute from the
// existing item (e.g., "a.b[3].c"), or a function of other such values.
// Note that the real right-hand-side of an assignment is actually a bit
// more general - it allows either a value, or a value+value or value-value -
// see class set_rhs below.
struct value {
struct function_call {
std::string _function_name;
std::vector<value> _parameters;
};
std::variant<constant, path, function_call> _value;
void set_constant(constant c) {
_value = std::move(c);
}
void set_valref(std::string s) {
_value = constant { std::move(s) };
}
void set_path(path p) {
_value = std::move(p);
}
void set_func_name(std::string s) {
_value = function_call {std::move(s), {}};
}
void add_func_parameter(value v) {
std::get<function_call>(_value)._parameters.emplace_back(std::move(v));
}
bool is_constant() const {
return std::holds_alternative<constant>(_value);
}
bool is_path() const {
return std::holds_alternative<path>(_value);
}
bool is_func() const {
return std::holds_alternative<function_call>(_value);
}
};
// The right-hand-side of a SET in an update expression can be either a
// single value (see above), or value+value, or value-value.
class set_rhs {
public:
char _op; // '+', '-', or 'v''
value _v1;
value _v2;
void set_value(value&& v1) {
_op = 'v';
_v1 = std::move(v1);
}
void set_plus(value&& v2) {
_op = '+';
_v2 = std::move(v2);
}
void set_minus(value&& v2) {
_op = '-';
_v2 = std::move(v2);
}
};
class update_expression {
public:
struct action {
path _path;
struct set {
set_rhs _rhs;
};
struct remove {
};
struct add {
constant _valref;
};
struct del {
constant _valref;
};
std::variant<set, remove, add, del> _action;
void assign_set(path p, set_rhs rhs) {
_path = std::move(p);
_action = set { std::move(rhs) };
}
void assign_remove(path p) {
_path = std::move(p);
_action = remove { };
}
void assign_add(path p, std::string v) {
_path = std::move(p);
_action = add { constant { std::move(v) } };
}
void assign_del(path p, std::string v) {
_path = std::move(p);
_action = del { constant { std::move(v) } };
}
};
private:
std::vector<action> _actions;
bool seen_set = false;
bool seen_remove = false;
bool seen_add = false;
bool seen_del = false;
public:
void add(action a);
void append(update_expression other);
bool empty() const {
return _actions.empty();
}
const std::vector<action>& actions() const {
return _actions;
}
std::vector<action>& actions() {
return _actions;
}
};
// A primitive_condition is a condition expression involving one condition,
// while the full condition_expression below adds boolean logic over these
// primitive conditions.
// The supported primitive conditions are:
// 1. Binary operators - v1 OP v2, where OP is =, <>, <, <=, >, or >= and
// v1 and v2 are values - from the item (an attribute path), the query
// (a ":val" reference), or a function of the the above (only the size()
// function is supported).
// 2. Ternary operator - v1 BETWEEN v2 and v3 (means v1 >= v2 AND v1 <= v3).
// 3. N-ary operator - v1 IN ( v2, v3, ... )
// 4. A single function call (attribute_exists etc.). The parser actually
// accepts a more general "value" here but later stages reject a value
// which is not a function call (because DynamoDB does it too).
class primitive_condition {
public:
enum class type {
UNDEFINED, VALUE, EQ, NE, LT, LE, GT, GE, BETWEEN, IN
};
type _op = type::UNDEFINED;
std::vector<value> _values;
void set_operator(type op) {
_op = op;
}
void add_value(value&& v) {
_values.push_back(std::move(v));
}
bool empty() const {
return _op == type::UNDEFINED;
}
};
class condition_expression {
public:
bool _negated = false; // If true, the entire condition is negated
struct condition_list {
char op = '|'; // '&' or '|'
std::vector<condition_expression> conditions;
};
std::variant<primitive_condition, condition_list> _expression = condition_list();
void set_primitive(primitive_condition&& p) {
_expression = std::move(p);
}
void append(condition_expression&& c, char op);
void apply_not() {
_negated = !_negated;
}
bool empty() const {
return std::holds_alternative<condition_list>(_expression) &&
std::get<condition_list>(_expression).conditions.empty();
}
};
} // namespace parsed
} // namespace alternator

View File

@@ -1,128 +0,0 @@
/*
* Copyright 2020 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include "seastarx.hh"
#include "service/storage_proxy.hh"
#include "service/storage_proxy.hh"
#include "utils/rjson.hh"
#include "executor.hh"
namespace alternator {
// An rmw_operation encapsulates the common logic of all the item update
// operations which may involve a read of the item before the write
// (so-called Read-Modify-Write operations). These operations include PutItem,
// UpdateItem and DeleteItem: All of these may be conditional operations (the
// "Expected" parameter) which requir a read before the write, and UpdateItem
// may also have an update expression which refers to the item's old value.
//
// The code below supports running the read and the write together as one
// transaction using LWT (this is why rmw_operation is a subclass of
// cas_request, as required by storage_proxy::cas()), but also has optional
// modes not using LWT.
class rmw_operation : public service::cas_request, public enable_shared_from_this<rmw_operation> {
public:
// The following options choose which mechanism to use for isolating
// parallel write operations:
// * The FORBID_RMW option forbids RMW (read-modify-write) operations
// such as conditional updates. For the remaining write-only
// operations, ordinary quorum writes are isolated enough.
// * The LWT_ALWAYS option always uses LWT (lightweight transactions)
// for any write operation - whether or not it also has a read.
// * The LWT_RMW_ONLY option uses LWT only for RMW operations, and uses
// ordinary quorum writes for write-only operations.
// This option is not safe if the user may send both RMW and write-only
// operations on the same item.
// * The UNSAFE_RMW option does read-modify-write operations as separate
// read and write. It is unsafe - concurrent RMW operations are not
// isolated at all. This option will likely be removed in the future.
enum class write_isolation {
FORBID_RMW, LWT_ALWAYS, LWT_RMW_ONLY, UNSAFE_RMW
};
static constexpr auto WRITE_ISOLATION_TAG_KEY = "system:write_isolation";
static write_isolation get_write_isolation_for_schema(schema_ptr schema);
static write_isolation default_write_isolation;
public:
static void set_default_write_isolation(std::string_view mode);
protected:
// The full request JSON
rjson::value _request;
// All RMW operations involve a single item with a specific partition
// and optional clustering key, in a single table, so the following
// information is common to all of them:
schema_ptr _schema;
partition_key _pk = partition_key::make_empty();
clustering_key _ck = clustering_key::make_empty();
write_isolation _write_isolation;
// All RMW operations can have a ReturnValues parameter from the following
// choices. But note that only UpdateItem actually supports all of them:
enum class returnvalues {
NONE, ALL_OLD, UPDATED_OLD, ALL_NEW, UPDATED_NEW
} _returnvalues;
static returnvalues parse_returnvalues(const rjson::value& request);
// When _returnvalues != NONE, apply() should store here, in JSON form,
// the values which are to be returned in the "Attributes" field.
// The default null JSON means do not return an Attributes field at all.
// This field is marked "mutable" so that the const apply() can modify
// it (see explanation below), but note that because apply() may be
// called more than once, if apply() will sometimes set this field it
// must set it (even if just to the default empty value) every time.
mutable rjson::value _return_attributes;
public:
// The constructor of a rmw_operation subclass should parse the request
// and try to discover as many input errors as it can before really
// attempting the read or write operations.
rmw_operation(service::storage_proxy& proxy, rjson::value&& request);
// rmw_operation subclasses (update_item_operation, put_item_operation
// and delete_item_operation) shall implement an apply() function which
// takes the previous value of the item (if it was read) and creates the
// write mutation. If the previous value of item does not pass the needed
// conditional expression, apply() should return an empty optional.
// apply() may throw if it encounters input errors not discovered during
// the constructor.
// apply() may be called more than once in case of contention, so it must
// not change the state saved in the object (issue #7218 was caused by
// violating this). We mark apply() "const" to let the compiler validate
// this for us. The output-only field _return_attributes is marked
// "mutable" above so that apply() can still write to it.
virtual std::optional<mutation> apply(std::unique_ptr<rjson::value> previous_item, api::timestamp_type ts) const = 0;
// Convert the above apply() into the signature needed by cas_request:
virtual std::optional<mutation> apply(foreign_ptr<lw_shared_ptr<query::result>> qr, const query::partition_slice& slice, api::timestamp_type ts) override;
virtual ~rmw_operation() = default;
schema_ptr schema() const { return _schema; }
const rjson::value& request() const { return _request; }
rjson::value&& move_request() && { return std::move(_request); }
future<executor::request_return_type> execute(service::storage_proxy& proxy,
service::client_state& client_state,
tracing::trace_state_ptr trace_state,
service_permit permit,
bool needs_read_before_write,
stats& stats);
std::optional<shard_id> shard_for_execute(bool needs_read_before_write);
};
} // namespace alternator

View File

@@ -1,375 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#include "base64.hh"
#include "log.hh"
#include "serialization.hh"
#include "error.hh"
#include "rapidjson/writer.h"
#include "concrete_types.hh"
#include "cql3/type_json.hh"
static logging::logger slogger("alternator-serialization");
namespace alternator {
type_info type_info_from_string(std::string_view type) {
static thread_local const std::unordered_map<std::string_view, type_info> type_infos = {
{"S", {alternator_type::S, utf8_type}},
{"B", {alternator_type::B, bytes_type}},
{"BOOL", {alternator_type::BOOL, boolean_type}},
{"N", {alternator_type::N, decimal_type}}, //FIXME: Replace with custom Alternator type when implemented
};
auto it = type_infos.find(type);
if (it == type_infos.end()) {
return {alternator_type::NOT_SUPPORTED_YET, utf8_type};
}
return it->second;
}
type_representation represent_type(alternator_type atype) {
static thread_local const std::unordered_map<alternator_type, type_representation> type_representations = {
{alternator_type::S, {"S", utf8_type}},
{alternator_type::B, {"B", bytes_type}},
{alternator_type::BOOL, {"BOOL", boolean_type}},
{alternator_type::N, {"N", decimal_type}}, //FIXME: Replace with custom Alternator type when implemented
};
auto it = type_representations.find(atype);
if (it == type_representations.end()) {
throw std::runtime_error(format("Unknown alternator type {}", int8_t(atype)));
}
return it->second;
}
struct from_json_visitor {
const rjson::value& v;
bytes_ostream& bo;
void operator()(const reversed_type_impl& t) const { visit(*t.underlying_type(), from_json_visitor{v, bo}); };
void operator()(const string_type_impl& t) {
bo.write(t.from_string(rjson::to_string_view(v)));
}
void operator()(const bytes_type_impl& t) const {
bo.write(base64_decode(v));
}
void operator()(const boolean_type_impl& t) const {
bo.write(boolean_type->decompose(v.GetBool()));
}
void operator()(const decimal_type_impl& t) const {
try {
bo.write(t.from_string(rjson::to_string_view(v)));
} catch (const marshal_exception& e) {
throw api_error::validation(format("The parameter cannot be converted to a numeric value: {}", v));
}
}
// default
void operator()(const abstract_type& t) const {
bo.write(from_json_object(t, v, cql_serialization_format::internal()));
}
};
bytes serialize_item(const rjson::value& item) {
if (item.IsNull() || item.MemberCount() != 1) {
throw api_error::validation(format("An item can contain only one attribute definition: {}", item));
}
auto it = item.MemberBegin();
type_info type_info = type_info_from_string(rjson::to_string_view(it->name)); // JSON keys are guaranteed to be strings
if (type_info.atype == alternator_type::NOT_SUPPORTED_YET) {
slogger.trace("Non-optimal serialization of type {}", it->name);
return bytes{int8_t(type_info.atype)} + to_bytes(rjson::print(item));
}
bytes_ostream bo;
bo.write(bytes{int8_t(type_info.atype)});
visit(*type_info.dtype, from_json_visitor{it->value, bo});
return bytes(bo.linearize());
}
struct to_json_visitor {
rjson::value& deserialized;
const std::string& type_ident;
bytes_view bv;
void operator()(const reversed_type_impl& t) const { visit(*t.underlying_type(), to_json_visitor{deserialized, type_ident, bv}); };
void operator()(const decimal_type_impl& t) const {
auto s = to_json_string(*decimal_type, bytes(bv));
//FIXME(sarna): unnecessary copy
rjson::set_with_string_name(deserialized, type_ident, rjson::from_string(s));
}
void operator()(const string_type_impl& t) {
rjson::set_with_string_name(deserialized, type_ident, rjson::from_string(reinterpret_cast<const char *>(bv.data()), bv.size()));
}
void operator()(const bytes_type_impl& t) const {
std::string b64 = base64_encode(bv);
rjson::set_with_string_name(deserialized, type_ident, rjson::from_string(b64));
}
// default
void operator()(const abstract_type& t) const {
rjson::set_with_string_name(deserialized, type_ident, rjson::parse(to_json_string(t, bytes(bv))));
}
};
rjson::value deserialize_item(bytes_view bv) {
rjson::value deserialized(rapidjson::kObjectType);
if (bv.empty()) {
throw api_error::validation("Serialized value empty");
}
alternator_type atype = alternator_type(bv[0]);
bv.remove_prefix(1);
if (atype == alternator_type::NOT_SUPPORTED_YET) {
slogger.trace("Non-optimal deserialization of alternator type {}", int8_t(atype));
return rjson::parse(std::string_view(reinterpret_cast<const char *>(bv.data()), bv.size()));
}
type_representation type_representation = represent_type(atype);
visit(*type_representation.dtype, to_json_visitor{deserialized, type_representation.ident, bv});
return deserialized;
}
std::string type_to_string(data_type type) {
static thread_local std::unordered_map<data_type, std::string> types = {
{utf8_type, "S"},
{bytes_type, "B"},
{boolean_type, "BOOL"},
{decimal_type, "N"}, // FIXME: use a specialized Alternator number type instead of the general decimal_type
};
auto it = types.find(type);
if (it == types.end()) {
// fall back to string, in order to be able to present
// internal Scylla types in a human-readable way
return "S";
}
return it->second;
}
bytes get_key_column_value(const rjson::value& item, const column_definition& column) {
std::string column_name = column.name_as_text();
const rjson::value* key_typed_value = rjson::find(item, column_name);
if (!key_typed_value) {
throw api_error::validation(format("Key column {} not found", column_name));
}
return get_key_from_typed_value(*key_typed_value, column);
}
// Parses the JSON encoding for a key value, which is a map with a single
// entry, whose key is the type (expected to match the key column's type)
// and the value is the encoded value.
bytes get_key_from_typed_value(const rjson::value& key_typed_value, const column_definition& column) {
if (!key_typed_value.IsObject() || key_typed_value.MemberCount() != 1 ||
!key_typed_value.MemberBegin()->value.IsString()) {
throw api_error::validation(
format("Malformed value object for key column {}: {}",
column.name_as_text(), key_typed_value));
}
auto it = key_typed_value.MemberBegin();
if (it->name != type_to_string(column.type)) {
throw api_error::validation(
format("Type mismatch: expected type {} for key column {}, got type {}",
type_to_string(column.type), column.name_as_text(), it->name));
}
std::string_view value_view = rjson::to_string_view(it->value);
if (value_view.empty()) {
throw api_error::validation(
format("The AttributeValue for a key attribute cannot contain an empty string value. Key: {}", column.name_as_text()));
}
if (column.type == bytes_type) {
return base64_decode(it->value);
} else {
return column.type->from_string(rjson::to_string_view(it->value));
}
}
rjson::value json_key_column_value(bytes_view cell, const column_definition& column) {
if (column.type == bytes_type) {
std::string b64 = base64_encode(cell);
return rjson::from_string(b64);
} if (column.type == utf8_type) {
return rjson::from_string(std::string(reinterpret_cast<const char*>(cell.data()), cell.size()));
} else if (column.type == decimal_type) {
// FIXME: use specialized Alternator number type, not the more
// general "decimal_type". A dedicated type can be more efficient
// in storage space and in parsing speed.
auto s = to_json_string(*decimal_type, bytes(cell));
return rjson::from_string(s);
} else {
// Support for arbitrary key types is useful for parsing values of virtual tables,
// which can involve any type supported by Scylla.
// In order to guarantee that the returned type is parsable by alternator clients,
// they are represented simply as strings.
return rjson::from_string(column.type->to_string(bytes(cell)));
}
}
partition_key pk_from_json(const rjson::value& item, schema_ptr schema) {
std::vector<bytes> raw_pk;
// FIXME: this is a loop, but we really allow only one partition key column.
for (const column_definition& cdef : schema->partition_key_columns()) {
bytes raw_value = get_key_column_value(item, cdef);
raw_pk.push_back(std::move(raw_value));
}
return partition_key::from_exploded(raw_pk);
}
clustering_key ck_from_json(const rjson::value& item, schema_ptr schema) {
if (schema->clustering_key_size() == 0) {
return clustering_key::make_empty();
}
std::vector<bytes> raw_ck;
// FIXME: this is a loop, but we really allow only one clustering key column.
for (const column_definition& cdef : schema->clustering_key_columns()) {
bytes raw_value = get_key_column_value(item, cdef);
raw_ck.push_back(std::move(raw_value));
}
return clustering_key::from_exploded(raw_ck);
}
big_decimal unwrap_number(const rjson::value& v, std::string_view diagnostic) {
if (!v.IsObject() || v.MemberCount() != 1) {
throw api_error::validation(format("{}: invalid number object", diagnostic));
}
auto it = v.MemberBegin();
if (it->name != "N") {
throw api_error::validation(format("{}: expected number, found type '{}'", diagnostic, it->name));
}
try {
if (it->value.IsNumber()) {
// FIXME(sarna): should use big_decimal constructor with numeric values directly:
return big_decimal(rjson::print(it->value));
}
if (!it->value.IsString()) {
throw api_error::validation(format("{}: improperly formatted number constant", diagnostic));
}
return big_decimal(rjson::to_string_view(it->value));
} catch (const marshal_exception& e) {
throw api_error::validation(format("The parameter cannot be converted to a numeric value: {}", it->value));
}
}
const std::pair<std::string, const rjson::value*> unwrap_set(const rjson::value& v) {
if (!v.IsObject() || v.MemberCount() != 1) {
return {"", nullptr};
}
auto it = v.MemberBegin();
const std::string it_key = it->name.GetString();
if (it_key != "SS" && it_key != "BS" && it_key != "NS") {
return {"", nullptr};
}
return std::make_pair(it_key, &(it->value));
}
const rjson::value* unwrap_list(const rjson::value& v) {
if (!v.IsObject() || v.MemberCount() != 1) {
return nullptr;
}
auto it = v.MemberBegin();
if (it->name != std::string("L")) {
return nullptr;
}
return &(it->value);
}
// Take two JSON-encoded numeric values ({"N": "thenumber"}) and return the
// sum, again as a JSON-encoded number.
rjson::value number_add(const rjson::value& v1, const rjson::value& v2) {
auto n1 = unwrap_number(v1, "UpdateExpression");
auto n2 = unwrap_number(v2, "UpdateExpression");
rjson::value ret = rjson::empty_object();
std::string str_ret = std::string((n1 + n2).to_string());
rjson::set(ret, "N", rjson::from_string(str_ret));
return ret;
}
rjson::value number_subtract(const rjson::value& v1, const rjson::value& v2) {
auto n1 = unwrap_number(v1, "UpdateExpression");
auto n2 = unwrap_number(v2, "UpdateExpression");
rjson::value ret = rjson::empty_object();
std::string str_ret = std::string((n1 - n2).to_string());
rjson::set(ret, "N", rjson::from_string(str_ret));
return ret;
}
// Take two JSON-encoded set values (e.g. {"SS": [...the actual set]}) and
// return the sum of both sets, again as a set value.
rjson::value set_sum(const rjson::value& v1, const rjson::value& v2) {
auto [set1_type, set1] = unwrap_set(v1);
auto [set2_type, set2] = unwrap_set(v2);
if (set1_type != set2_type) {
throw api_error::validation(format("Mismatched set types: {} and {}", set1_type, set2_type));
}
if (!set1 || !set2) {
throw api_error::validation("UpdateExpression: ADD operation for sets must be given sets as arguments");
}
rjson::value sum = rjson::copy(*set1);
std::set<rjson::value, rjson::single_value_comp> set1_raw;
for (auto it = sum.Begin(); it != sum.End(); ++it) {
set1_raw.insert(rjson::copy(*it));
}
for (const auto& a : set2->GetArray()) {
if (!set1_raw.contains(a)) {
rjson::push_back(sum, rjson::copy(a));
}
}
rjson::value ret = rjson::empty_object();
rjson::set_with_string_name(ret, set1_type, std::move(sum));
return ret;
}
// Take two JSON-encoded set values (e.g. {"SS": [...the actual list]}) and
// return the difference of s1 - s2, again as a set value.
// DynamoDB does not allow empty sets, so if resulting set is empty, return
// an unset optional instead.
std::optional<rjson::value> set_diff(const rjson::value& v1, const rjson::value& v2) {
auto [set1_type, set1] = unwrap_set(v1);
auto [set2_type, set2] = unwrap_set(v2);
if (set1_type != set2_type) {
throw api_error::validation(format("Mismatched set types: {} and {}", set1_type, set2_type));
}
if (!set1 || !set2) {
throw api_error::validation("UpdateExpression: DELETE operation can only be performed on a set");
}
std::set<rjson::value, rjson::single_value_comp> set1_raw;
for (auto it = set1->Begin(); it != set1->End(); ++it) {
set1_raw.insert(rjson::copy(*it));
}
for (const auto& a : set2->GetArray()) {
set1_raw.erase(a);
}
if (set1_raw.empty()) {
return std::nullopt;
}
rjson::value ret = rjson::empty_object();
rjson::set_with_string_name(ret, set1_type, rjson::empty_array());
rjson::value& result_set = ret[set1_type];
for (const auto& a : set1_raw) {
rjson::push_back(result_set, rjson::copy(a));
}
return ret;
}
}

View File

@@ -1,89 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include <string>
#include <string_view>
#include "types.hh"
#include "schema_fwd.hh"
#include "keys.hh"
#include "utils/rjson.hh"
#include "utils/big_decimal.hh"
namespace alternator {
enum class alternator_type : int8_t {
S, B, BOOL, N, NOT_SUPPORTED_YET
};
struct type_info {
alternator_type atype;
data_type dtype;
};
struct type_representation {
std::string ident;
data_type dtype;
};
type_info type_info_from_string(std::string_view type);
type_representation represent_type(alternator_type atype);
bytes serialize_item(const rjson::value& item);
rjson::value deserialize_item(bytes_view bv);
std::string type_to_string(data_type type);
bytes get_key_column_value(const rjson::value& item, const column_definition& column);
bytes get_key_from_typed_value(const rjson::value& key_typed_value, const column_definition& column);
rjson::value json_key_column_value(bytes_view cell, const column_definition& column);
partition_key pk_from_json(const rjson::value& item, schema_ptr schema);
clustering_key ck_from_json(const rjson::value& item, schema_ptr schema);
// If v encodes a number (i.e., it is a {"N": [...]}, returns an object representing it. Otherwise,
// raises ValidationException with diagnostic.
big_decimal unwrap_number(const rjson::value& v, std::string_view diagnostic);
// Check if a given JSON object encodes a set (i.e., it is a {"SS": [...]}, or "NS", "BS"
// and returns set's type and a pointer to that set. If the object does not encode a set,
// returned value is {"", nullptr}
const std::pair<std::string, const rjson::value*> unwrap_set(const rjson::value& v);
// Check if a given JSON object encodes a list (i.e., it is a {"L": [...]}
// and returns a pointer to that list.
const rjson::value* unwrap_list(const rjson::value& v);
// Take two JSON-encoded numeric values ({"N": "thenumber"}) and return the
// sum, again as a JSON-encoded number.
rjson::value number_add(const rjson::value& v1, const rjson::value& v2);
rjson::value number_subtract(const rjson::value& v1, const rjson::value& v2);
// Take two JSON-encoded set values (e.g. {"SS": [...the actual set]}) and
// return the sum of both sets, again as a set value.
rjson::value set_sum(const rjson::value& v1, const rjson::value& v2);
// Take two JSON-encoded set values (e.g. {"SS": [...the actual list]}) and
// return the difference of s1 - s2, again as a set value.
// DynamoDB does not allow empty sets, so if resulting set is empty, return
// an unset optional instead.
std::optional<rjson::value> set_diff(const rjson::value& v1, const rjson::value& v2);
}

View File

@@ -1,499 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#include "alternator/server.hh"
#include "log.hh"
#include <seastar/http/function_handlers.hh>
#include <seastar/json/json_elements.hh>
#include "seastarx.hh"
#include "error.hh"
#include "utils/rjson.hh"
#include "auth.hh"
#include <cctype>
#include "cql3/query_processor.hh"
#include "service/storage_service.hh"
#include "utils/overloaded_functor.hh"
static logging::logger slogger("alternator-server");
using namespace httpd;
namespace alternator {
static constexpr auto TARGET = "X-Amz-Target";
inline std::vector<std::string_view> split(std::string_view text, char separator) {
std::vector<std::string_view> tokens;
if (text == "") {
return tokens;
}
while (true) {
auto pos = text.find_first_of(separator);
if (pos != std::string_view::npos) {
tokens.emplace_back(text.data(), pos);
text.remove_prefix(pos + 1);
} else {
tokens.emplace_back(text);
break;
}
}
return tokens;
}
// DynamoDB HTTP error responses are structured as follows
// https://docs.aws.amazon.com/amazondynamodb/latest/developerguide/Programming.Errors.html
// Our handlers throw an exception to report an error. If the exception
// is of type alternator::api_error, it unwrapped and properly reported to
// the user directly. Other exceptions are unexpected, and reported as
// Internal Server Error.
class api_handler : public handler_base {
public:
api_handler(const std::function<future<executor::request_return_type>(std::unique_ptr<request> req)>& _handle) : _f_handle(
[this, _handle](std::unique_ptr<request> req, std::unique_ptr<reply> rep) {
return seastar::futurize_invoke(_handle, std::move(req)).then_wrapped([this, rep = std::move(rep)](future<executor::request_return_type> resf) mutable {
if (resf.failed()) {
// Exceptions of type api_error are wrapped as JSON and
// returned to the client as expected. Other types of
// exceptions are unexpected, and returned to the user
// as an internal server error:
try {
resf.get();
} catch (api_error &ae) {
generate_error_reply(*rep, ae);
} catch (rjson::error & re) {
generate_error_reply(*rep,
api_error::validation(re.what()));
} catch (...) {
generate_error_reply(*rep,
api_error::internal(format("Internal server error: {}", std::current_exception())));
}
return make_ready_future<std::unique_ptr<reply>>(std::move(rep));
}
auto res = resf.get0();
std::visit(overloaded_functor {
[&] (const json::json_return_type& json_return_value) {
slogger.trace("api_handler success case");
if (json_return_value._body_writer) {
rep->write_body("json", std::move(json_return_value._body_writer));
} else {
rep->_content += json_return_value._res;
}
},
[&] (const api_error& err) {
generate_error_reply(*rep, err);
}
}, res);
return make_ready_future<std::unique_ptr<reply>>(std::move(rep));
});
}), _type("json") { }
api_handler(const api_handler&) = default;
future<std::unique_ptr<reply>> handle(const sstring& path,
std::unique_ptr<request> req, std::unique_ptr<reply> rep) override {
return _f_handle(std::move(req), std::move(rep)).then(
[this](std::unique_ptr<reply> rep) {
rep->done(_type);
return make_ready_future<std::unique_ptr<reply>>(std::move(rep));
});
}
protected:
void generate_error_reply(reply& rep, const api_error& err) {
rep._content += "{\"__type\":\"com.amazonaws.dynamodb.v20120810#" + err._type + "\"," +
"\"message\":\"" + err._msg + "\"}";
rep._status = err._http_code;
slogger.trace("api_handler error case: {}", rep._content);
}
future_handler_function _f_handle;
sstring _type;
};
class gated_handler : public handler_base {
seastar::gate& _gate;
public:
gated_handler(seastar::gate& gate) : _gate(gate) {}
virtual future<std::unique_ptr<reply>> do_handle(const sstring& path, std::unique_ptr<request> req, std::unique_ptr<reply> rep) = 0;
virtual future<std::unique_ptr<reply>> handle(const sstring& path, std::unique_ptr<request> req, std::unique_ptr<reply> rep) final override {
return with_gate(_gate, [this, &path, req = std::move(req), rep = std::move(rep)] () mutable {
return do_handle(path, std::move(req), std::move(rep));
});
}
};
class health_handler : public gated_handler {
public:
health_handler(seastar::gate& pending_requests) : gated_handler(pending_requests) {}
protected:
virtual future<std::unique_ptr<reply>> do_handle(const sstring& path, std::unique_ptr<request> req, std::unique_ptr<reply> rep) override {
rep->set_status(reply::status_type::ok);
rep->write_body("txt", format("healthy: {}", req->get_header("Host")));
return make_ready_future<std::unique_ptr<reply>>(std::move(rep));
}
};
class local_nodelist_handler : public gated_handler {
public:
local_nodelist_handler(seastar::gate& pending_requests) : gated_handler(pending_requests) {}
protected:
virtual future<std::unique_ptr<reply>> do_handle(const sstring& path, std::unique_ptr<request> req, std::unique_ptr<reply> rep) override {
rjson::value results = rjson::empty_array();
// It's very easy to get a list of all live nodes on the cluster,
// using gms::get_local_gossiper().get_live_members(). But getting
// just the list of live nodes in this DC needs more elaborate code:
sstring local_dc = locator::i_endpoint_snitch::get_local_snitch_ptr()->get_datacenter(
utils::fb_utilities::get_broadcast_address());
std::unordered_set<gms::inet_address> local_dc_nodes =
service::get_local_storage_service().get_token_metadata().
get_topology().get_datacenter_endpoints().at(local_dc);
for (auto& ip : local_dc_nodes) {
if (gms::get_local_gossiper().is_alive(ip)) {
rjson::push_back(results, rjson::from_string(ip.to_sstring()));
}
}
rep->set_status(reply::status_type::ok);
rep->set_content_type("json");
rep->_content = rjson::print(results);
return make_ready_future<std::unique_ptr<reply>>(std::move(rep));
}
};
future<> server::verify_signature(const request& req) {
if (!_enforce_authorization) {
slogger.debug("Skipping authorization");
return make_ready_future<>();
}
auto host_it = req._headers.find("Host");
if (host_it == req._headers.end()) {
throw api_error::invalid_signature("Host header is mandatory for signature verification");
}
auto authorization_it = req._headers.find("Authorization");
if (authorization_it == req._headers.end()) {
throw api_error::invalid_signature("Authorization header is mandatory for signature verification");
}
std::string host = host_it->second;
std::vector<std::string_view> credentials_raw = split(authorization_it->second, ' ');
std::string credential;
std::string user_signature;
std::string signed_headers_str;
std::vector<std::string_view> signed_headers;
for (std::string_view entry : credentials_raw) {
std::vector<std::string_view> entry_split = split(entry, '=');
if (entry_split.size() != 2) {
if (entry != "AWS4-HMAC-SHA256") {
throw api_error::invalid_signature(format("Only AWS4-HMAC-SHA256 algorithm is supported. Found: {}", entry));
}
continue;
}
std::string_view auth_value = entry_split[1];
// Commas appear as an additional (quite redundant) delimiter
if (auth_value.back() == ',') {
auth_value.remove_suffix(1);
}
if (entry_split[0] == "Credential") {
credential = std::string(auth_value);
} else if (entry_split[0] == "Signature") {
user_signature = std::string(auth_value);
} else if (entry_split[0] == "SignedHeaders") {
signed_headers_str = std::string(auth_value);
signed_headers = split(auth_value, ';');
std::sort(signed_headers.begin(), signed_headers.end());
}
}
std::vector<std::string_view> credential_split = split(credential, '/');
if (credential_split.size() != 5) {
throw api_error::validation(format("Incorrect credential information format: {}", credential));
}
std::string user(credential_split[0]);
std::string datestamp(credential_split[1]);
std::string region(credential_split[2]);
std::string service(credential_split[3]);
std::map<std::string_view, std::string_view> signed_headers_map;
for (const auto& header : signed_headers) {
signed_headers_map.emplace(header, std::string_view());
}
for (auto& header : req._headers) {
std::string header_str;
header_str.resize(header.first.size());
std::transform(header.first.begin(), header.first.end(), header_str.begin(), ::tolower);
auto it = signed_headers_map.find(header_str);
if (it != signed_headers_map.end()) {
it->second = std::string_view(header.second);
}
}
auto cache_getter = [&qp = _qp] (std::string username) {
return get_key_from_roles(qp, std::move(username));
};
return _key_cache.get_ptr(user, cache_getter).then([this, &req,
user = std::move(user),
host = std::move(host),
datestamp = std::move(datestamp),
signed_headers_str = std::move(signed_headers_str),
signed_headers_map = std::move(signed_headers_map),
region = std::move(region),
service = std::move(service),
user_signature = std::move(user_signature)] (key_cache::value_ptr key_ptr) {
std::string signature = get_signature(user, *key_ptr, std::string_view(host), req._method,
datestamp, signed_headers_str, signed_headers_map, req.content, region, service, "");
if (signature != std::string_view(user_signature)) {
_key_cache.remove(user);
throw api_error::unrecognized_client("The security token included in the request is invalid.");
}
});
}
future<executor::request_return_type> server::handle_api_request(std::unique_ptr<request>&& req) {
_executor._stats.total_operations++;
sstring target = req->get_header(TARGET);
std::vector<std::string_view> split_target = split(target, '.');
//NOTICE(sarna): Target consists of Dynamo API version followed by a dot '.' and operation type (e.g. CreateTable)
std::string op = split_target.empty() ? std::string() : std::string(split_target.back());
slogger.trace("Request: {} {} {}", op, req->content, req->_headers);
return verify_signature(*req).then([this, op, req = std::move(req)] () mutable {
auto callback_it = _callbacks.find(op);
if (callback_it == _callbacks.end()) {
_executor._stats.unsupported_operations++;
throw api_error::unknown_operation(format("Unsupported operation {}", op));
}
return with_gate(_pending_requests, [this, callback_it = std::move(callback_it), op = std::move(op), req = std::move(req)] () mutable {
//FIXME: Client state can provide more context, e.g. client's endpoint address
// We use unique_ptr because client_state cannot be moved or copied
return do_with(std::make_unique<executor::client_state>(executor::client_state::internal_tag()),
[this, callback_it = std::move(callback_it), op = std::move(op), req = std::move(req)] (std::unique_ptr<executor::client_state>& client_state) mutable {
tracing::trace_state_ptr trace_state = executor::maybe_trace_query(*client_state, op, req->content);
tracing::trace(trace_state, op);
// JSON parsing can allocate up to roughly 2x the size of the raw document, + a couple of bytes for maintenance.
// FIXME: by this time, the whole HTTP request was already read, so some memory is already occupied.
// Once HTTP allows working on streams, we should grab the permit *before* reading the HTTP payload.
size_t mem_estimate = req->content.size() * 3 + 8000;
auto units_fut = get_units(*_memory_limiter, mem_estimate);
if (_memory_limiter->waiters()) {
++_executor._stats.requests_blocked_memory;
}
return units_fut.then([this, callback_it = std::move(callback_it), &client_state, trace_state, req = std::move(req)] (semaphore_units<> units) mutable {
return _json_parser.parse(req->content).then([this, callback_it = std::move(callback_it), &client_state, trace_state,
units = std::move(units), req = std::move(req)] (rjson::value json_request) mutable {
return callback_it->second(_executor, *client_state, trace_state, make_service_permit(std::move(units)), std::move(json_request), std::move(req)).finally([trace_state] {});
});
});
});
});
});
}
void server::set_routes(routes& r) {
api_handler* req_handler = new api_handler([this] (std::unique_ptr<request> req) mutable {
return handle_api_request(std::move(req));
});
r.put(operation_type::POST, "/", req_handler);
r.put(operation_type::GET, "/", new health_handler(_pending_requests));
// The "/localnodes" request is a new Alternator feature, not supported by
// DynamoDB and not required for DynamoDB compatibility. It allows a
// client to enquire - using a trivial HTTP request without requiring
// authentication - the list of all live nodes in the same data center of
// the Alternator cluster. The client can use this list to balance its
// request load to all the nodes in the same geographical region.
// Note that this API exposes - openly without authentication - the
// information on the cluster's members inside one data center. We do not
// consider this to be a security risk, because an attacker can already
// scan an entire subnet for nodes responding to the health request,
// or even just scan for open ports.
r.put(operation_type::GET, "/localnodes", new local_nodelist_handler(_pending_requests));
}
//FIXME: A way to immediately invalidate the cache should be considered,
// e.g. when the system table which stores the keys is changed.
// For now, this propagation may take up to 1 minute.
server::server(executor& exec, cql3::query_processor& qp)
: _http_server("http-alternator")
, _https_server("https-alternator")
, _executor(exec)
, _qp(qp)
, _key_cache(1024, 1min, slogger)
, _enforce_authorization(false)
, _enabled_servers{}
, _pending_requests{}
, _callbacks{
{"CreateTable", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.create_table(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
{"DescribeTable", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.describe_table(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
{"DeleteTable", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.delete_table(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
{"UpdateTable", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.update_table(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
{"PutItem", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.put_item(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
{"UpdateItem", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.update_item(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
{"GetItem", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.get_item(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
{"DeleteItem", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.delete_item(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
{"ListTables", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.list_tables(client_state, std::move(permit), std::move(json_request));
}},
{"Scan", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.scan(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
{"DescribeEndpoints", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.describe_endpoints(client_state, std::move(permit), std::move(json_request), req->get_header("Host"));
}},
{"BatchWriteItem", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.batch_write_item(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
{"BatchGetItem", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.batch_get_item(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
{"Query", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.query(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
{"TagResource", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.tag_resource(client_state, std::move(permit), std::move(json_request));
}},
{"UntagResource", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.untag_resource(client_state, std::move(permit), std::move(json_request));
}},
{"ListTagsOfResource", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.list_tags_of_resource(client_state, std::move(permit), std::move(json_request));
}},
{"ListStreams", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.list_streams(client_state, std::move(permit), std::move(json_request));
}},
{"DescribeStream", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.describe_stream(client_state, std::move(permit), std::move(json_request));
}},
{"GetShardIterator", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.get_shard_iterator(client_state, std::move(permit), std::move(json_request));
}},
{"GetRecords", [] (executor& e, executor::client_state& client_state, tracing::trace_state_ptr trace_state, service_permit permit, rjson::value json_request, std::unique_ptr<request> req) {
return e.get_records(client_state, std::move(trace_state), std::move(permit), std::move(json_request));
}},
} {
}
future<> server::init(net::inet_address addr, std::optional<uint16_t> port, std::optional<uint16_t> https_port, std::optional<tls::credentials_builder> creds,
bool enforce_authorization, semaphore* memory_limiter) {
_memory_limiter = memory_limiter;
_enforce_authorization = enforce_authorization;
if (!port && !https_port) {
return make_exception_future<>(std::runtime_error("Either regular port or TLS port"
" must be specified in order to init an alternator HTTP server instance"));
}
return seastar::async([this, addr, port, https_port, creds] {
try {
_executor.start().get();
if (port) {
set_routes(_http_server._routes);
_http_server.set_content_length_limit(server::content_length_limit);
_http_server.listen(socket_address{addr, *port}).get();
_enabled_servers.push_back(std::ref(_http_server));
}
if (https_port) {
set_routes(_https_server._routes);
_https_server.set_content_length_limit(server::content_length_limit);
_https_server.set_tls_credentials(creds->build_reloadable_server_credentials([](const std::unordered_set<sstring>& files, std::exception_ptr ep) {
if (ep) {
slogger.warn("Exception loading {}: {}", files, ep);
} else {
slogger.info("Reloaded {}", files);
}
}).get0());
_https_server.listen(socket_address{addr, *https_port}).get();
_enabled_servers.push_back(std::ref(_https_server));
}
} catch (...) {
slogger.error("Failed to set up Alternator HTTP server on {} port {}, TLS port {}: {}",
addr, port ? std::to_string(*port) : "OFF", https_port ? std::to_string(*https_port) : "OFF", std::current_exception());
std::throw_with_nested(std::runtime_error(
format("Failed to set up Alternator HTTP server on {} port {}, TLS port {}",
addr, port ? std::to_string(*port) : "OFF", https_port ? std::to_string(*https_port) : "OFF")));
}
});
}
future<> server::stop() {
return parallel_for_each(_enabled_servers, [] (http_server& server) {
return server.stop();
}).then([this] {
return _pending_requests.close();
}).then([this] {
return _json_parser.stop();
});
}
server::json_parser::json_parser() : _run_parse_json_thread(async([this] {
while (true) {
_document_waiting.wait().get();
if (_as.abort_requested()) {
return;
}
try {
_parsed_document = rjson::parse_yieldable(_raw_document);
_current_exception = nullptr;
} catch (...) {
_current_exception = std::current_exception();
}
_document_parsed.signal();
}
})) {
}
future<rjson::value> server::json_parser::parse(std::string_view content) {
if (content.size() < yieldable_parsing_threshold) {
return make_ready_future<rjson::value>(rjson::parse(content));
}
return with_semaphore(_parsing_sem, 1, [this, content] {
_raw_document = content;
_document_waiting.signal();
return _document_parsed.wait().then([this] {
if (_current_exception) {
return make_exception_future<rjson::value>(_current_exception);
}
return make_ready_future<rjson::value>(std::move(_parsed_document));
});
});
}
future<> server::json_parser::stop() {
_as.request_abort();
_document_waiting.signal();
_document_parsed.broken();
return std::move(_run_parse_json_thread);
}
}

View File

@@ -1,84 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include "alternator/executor.hh"
#include <seastar/core/future.hh>
#include <seastar/http/httpd.hh>
#include <seastar/net/tls.hh>
#include <optional>
#include "alternator/auth.hh"
#include "utils/small_vector.hh"
#include <seastar/core/units.hh>
namespace alternator {
class server {
static constexpr size_t content_length_limit = 16*MB;
using alternator_callback = std::function<future<executor::request_return_type>(executor&, executor::client_state&,
tracing::trace_state_ptr, service_permit, rjson::value, std::unique_ptr<request>)>;
using alternator_callbacks_map = std::unordered_map<std::string_view, alternator_callback>;
http_server _http_server;
http_server _https_server;
executor& _executor;
cql3::query_processor& _qp;
key_cache _key_cache;
bool _enforce_authorization;
utils::small_vector<std::reference_wrapper<seastar::httpd::http_server>, 2> _enabled_servers;
gate _pending_requests;
alternator_callbacks_map _callbacks;
semaphore* _memory_limiter;
class json_parser {
static constexpr size_t yieldable_parsing_threshold = 16*KB;
std::string_view _raw_document;
rjson::value _parsed_document;
std::exception_ptr _current_exception;
semaphore _parsing_sem{1};
condition_variable _document_waiting;
condition_variable _document_parsed;
abort_source _as;
future<> _run_parse_json_thread;
public:
json_parser();
future<rjson::value> parse(std::string_view content);
future<> stop();
};
json_parser _json_parser;
public:
server(executor& executor, cql3::query_processor& qp);
future<> init(net::inet_address addr, std::optional<uint16_t> port, std::optional<uint16_t> https_port, std::optional<tls::credentials_builder> creds,
bool enforce_authorization, semaphore* memory_limiter);
future<> stop();
private:
void set_routes(seastar::httpd::routes& r);
future<> verify_signature(const seastar::httpd::request& r);
future<executor::request_return_type> handle_api_request(std::unique_ptr<request>&& req);
};
}

View File

@@ -1,109 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#include "stats.hh"
#include "utils/histogram_metrics_helper.hh"
#include <seastar/core/metrics.hh>
namespace alternator {
const char* ALTERNATOR_METRICS = "alternator";
stats::stats() : api_operations{} {
// Register the
seastar::metrics::label op("op");
_metrics.add_group("alternator", {
#define OPERATION(name, CamelCaseName) \
seastar::metrics::make_total_operations("operation", api_operations.name, \
seastar::metrics::description("number of operations via Alternator API"), {op(CamelCaseName)}),
#define OPERATION_LATENCY(name, CamelCaseName) \
seastar::metrics::make_histogram("op_latency", \
seastar::metrics::description("Latency histogram of an operation via Alternator API"), {op(CamelCaseName)}, [this]{return to_metrics_histogram(api_operations.name);}),
OPERATION(batch_write_item, "BatchWriteItem")
OPERATION(create_backup, "CreateBackup")
OPERATION(create_global_table, "CreateGlobalTable")
OPERATION(create_table, "CreateTable")
OPERATION(delete_backup, "DeleteBackup")
OPERATION(delete_item, "DeleteItem")
OPERATION(delete_table, "DeleteTable")
OPERATION(describe_backup, "DescribeBackup")
OPERATION(describe_continuous_backups, "DescribeContinuousBackups")
OPERATION(describe_endpoints, "DescribeEndpoints")
OPERATION(describe_global_table, "DescribeGlobalTable")
OPERATION(describe_global_table_settings, "DescribeGlobalTableSettings")
OPERATION(describe_limits, "DescribeLimits")
OPERATION(describe_table, "DescribeTable")
OPERATION(describe_time_to_live, "DescribeTimeToLive")
OPERATION(get_item, "GetItem")
OPERATION(list_backups, "ListBackups")
OPERATION(list_global_tables, "ListGlobalTables")
OPERATION(list_tables, "ListTables")
OPERATION(list_tags_of_resource, "ListTagsOfResource")
OPERATION(put_item, "PutItem")
OPERATION(query, "Query")
OPERATION(restore_table_from_backup, "RestoreTableFromBackup")
OPERATION(restore_table_to_point_in_time, "RestoreTableToPointInTime")
OPERATION(scan, "Scan")
OPERATION(tag_resource, "TagResource")
OPERATION(transact_get_items, "TransactGetItems")
OPERATION(transact_write_items, "TransactWriteItems")
OPERATION(untag_resource, "UntagResource")
OPERATION(update_continuous_backups, "UpdateContinuousBackups")
OPERATION(update_global_table, "UpdateGlobalTable")
OPERATION(update_global_table_settings, "UpdateGlobalTableSettings")
OPERATION(update_item, "UpdateItem")
OPERATION(update_table, "UpdateTable")
OPERATION(update_time_to_live, "UpdateTimeToLive")
OPERATION_LATENCY(put_item_latency, "PutItem")
OPERATION_LATENCY(get_item_latency, "GetItem")
OPERATION_LATENCY(delete_item_latency, "DeleteItem")
OPERATION_LATENCY(update_item_latency, "UpdateItem")
OPERATION(list_streams, "ListStreams")
OPERATION(describe_stream, "DescribeStream")
OPERATION(get_shard_iterator, "GetShardIterator")
OPERATION(get_records, "GetRecords")
OPERATION_LATENCY(get_records_latency, "GetRecords")
});
_metrics.add_group("alternator", {
seastar::metrics::make_total_operations("unsupported_operations", unsupported_operations,
seastar::metrics::description("number of unsupported operations via Alternator API")),
seastar::metrics::make_total_operations("total_operations", total_operations,
seastar::metrics::description("number of total operations via Alternator API")),
seastar::metrics::make_total_operations("reads_before_write", reads_before_write,
seastar::metrics::description("number of performed read-before-write operations")),
seastar::metrics::make_total_operations("write_using_lwt", write_using_lwt,
seastar::metrics::description("number of writes that used LWT")),
seastar::metrics::make_total_operations("shard_bounce_for_lwt", shard_bounce_for_lwt,
seastar::metrics::description("number writes that had to be bounced from this shard because of LWT requirements")),
seastar::metrics::make_total_operations("requests_blocked_memory", requests_blocked_memory,
seastar::metrics::description("Counts a number of requests blocked due to memory pressure.")),
seastar::metrics::make_total_operations("filtered_rows_read_total", cql_stats.filtered_rows_read_total,
seastar::metrics::description("number of rows read during filtering operations")),
seastar::metrics::make_total_operations("filtered_rows_matched_total", cql_stats.filtered_rows_matched_total,
seastar::metrics::description("number of rows read and matched during filtering operations")),
seastar::metrics::make_total_operations("filtered_rows_dropped_total", [this] { return cql_stats.filtered_rows_read_total - cql_stats.filtered_rows_matched_total; },
seastar::metrics::description("number of rows read and dropped during filtering operations")),
});
}
}

View File

@@ -1,103 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include <cstdint>
#include <seastar/core/metrics_registration.hh>
#include "seastarx.hh"
#include "utils/estimated_histogram.hh"
#include "cql3/stats.hh"
namespace alternator {
// Object holding per-shard statistics related to Alternator.
// While this object is alive, these metrics are also registered to be
// visible by the metrics REST API, with the "alternator" prefix.
class stats {
public:
stats();
// Count of DynamoDB API operations by types
struct {
uint64_t batch_get_item = 0;
uint64_t batch_write_item = 0;
uint64_t create_backup = 0;
uint64_t create_global_table = 0;
uint64_t create_table = 0;
uint64_t delete_backup = 0;
uint64_t delete_item = 0;
uint64_t delete_table = 0;
uint64_t describe_backup = 0;
uint64_t describe_continuous_backups = 0;
uint64_t describe_endpoints = 0;
uint64_t describe_global_table = 0;
uint64_t describe_global_table_settings = 0;
uint64_t describe_limits = 0;
uint64_t describe_table = 0;
uint64_t describe_time_to_live = 0;
uint64_t get_item = 0;
uint64_t list_backups = 0;
uint64_t list_global_tables = 0;
uint64_t list_tables = 0;
uint64_t list_tags_of_resource = 0;
uint64_t put_item = 0;
uint64_t query = 0;
uint64_t restore_table_from_backup = 0;
uint64_t restore_table_to_point_in_time = 0;
uint64_t scan = 0;
uint64_t tag_resource = 0;
uint64_t transact_get_items = 0;
uint64_t transact_write_items = 0;
uint64_t untag_resource = 0;
uint64_t update_continuous_backups = 0;
uint64_t update_global_table = 0;
uint64_t update_global_table_settings = 0;
uint64_t update_item = 0;
uint64_t update_table = 0;
uint64_t update_time_to_live = 0;
uint64_t list_streams = 0;
uint64_t describe_stream = 0;
uint64_t get_shard_iterator = 0;
uint64_t get_records = 0;
utils::time_estimated_histogram put_item_latency;
utils::time_estimated_histogram get_item_latency;
utils::time_estimated_histogram delete_item_latency;
utils::time_estimated_histogram update_item_latency;
utils::time_estimated_histogram get_records_latency;
} api_operations;
// Miscellaneous event counters
uint64_t total_operations = 0;
uint64_t unsupported_operations = 0;
uint64_t reads_before_write = 0;
uint64_t write_using_lwt = 0;
uint64_t shard_bounce_for_lwt = 0;
uint64_t requests_blocked_memory = 0;
// CQL-derived stats
cql3::cql_stats cql_stats;
private:
// The metric_groups object holds this stat object's metrics registered
// as long as the stats object is alive.
seastar::metrics::metric_groups _metrics;
};
}

File diff suppressed because it is too large Load Diff

View File

@@ -1,53 +0,0 @@
/*
* Copyright 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include "serializer.hh"
#include "schema.hh"
#include "db/extensions.hh"
namespace alternator {
class tags_extension : public schema_extension {
public:
static constexpr auto NAME = "scylla_tags";
tags_extension() = default;
explicit tags_extension(const std::map<sstring, sstring>& tags) : _tags(std::move(tags)) {}
explicit tags_extension(bytes b) : _tags(tags_extension::deserialize(b)) {}
explicit tags_extension(const sstring& s) {
throw std::logic_error("Cannot create tags from string");
}
bytes serialize() const override {
return ser::serialize_to_buffer<bytes>(_tags);
}
static std::map<sstring, sstring> deserialize(bytes_view buffer) {
return ser::deserialize_from_buffer(buffer, boost::type<std::map<sstring, sstring>>());
}
const std::map<sstring, sstring>& tags() const {
return _tags;
}
private:
std::map<sstring, sstring> _tags;
};
}

View File

@@ -13,7 +13,7 @@
{
"method":"GET",
"summary":"get row cache save period in seconds",
"type": "long",
"type":"int",
"nickname":"get_row_cache_save_period_in_seconds",
"produces":[
"application/json"
@@ -35,7 +35,7 @@
"description":"row cache save period in seconds",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -48,7 +48,7 @@
{
"method":"GET",
"summary":"get key cache save period in seconds",
"type": "long",
"type":"int",
"nickname":"get_key_cache_save_period_in_seconds",
"produces":[
"application/json"
@@ -70,7 +70,7 @@
"description":"key cache save period in seconds",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -83,7 +83,7 @@
{
"method":"GET",
"summary":"get counter cache save period in seconds",
"type": "long",
"type":"int",
"nickname":"get_counter_cache_save_period_in_seconds",
"produces":[
"application/json"
@@ -105,7 +105,7 @@
"description":"counter cache save period in seconds",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -118,7 +118,7 @@
{
"method":"GET",
"summary":"get row cache keys to save",
"type": "long",
"type":"int",
"nickname":"get_row_cache_keys_to_save",
"produces":[
"application/json"
@@ -140,7 +140,7 @@
"description":"row cache keys to save",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -153,7 +153,7 @@
{
"method":"GET",
"summary":"get key cache keys to save",
"type": "long",
"type":"int",
"nickname":"get_key_cache_keys_to_save",
"produces":[
"application/json"
@@ -175,7 +175,7 @@
"description":"key cache keys to save",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -188,7 +188,7 @@
{
"method":"GET",
"summary":"get counter cache keys to save",
"type": "long",
"type":"int",
"nickname":"get_counter_cache_keys_to_save",
"produces":[
"application/json"
@@ -210,7 +210,7 @@
"description":"counter cache keys to save",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -448,7 +448,7 @@
{
"method": "GET",
"summary": "Get key entries",
"type": "long",
"type": "int",
"nickname": "get_key_entries",
"produces": [
"application/json"
@@ -568,7 +568,7 @@
{
"method": "GET",
"summary": "Get row entries",
"type": "long",
"type": "int",
"nickname": "get_row_entries",
"produces": [
"application/json"
@@ -688,7 +688,7 @@
{
"method": "GET",
"summary": "Get counter entries",
"type": "long",
"type": "int",
"nickname": "get_counter_entries",
"produces": [
"application/json"

View File

@@ -70,7 +70,7 @@
{
"method":"POST",
"summary":"Force a major compaction of this column family",
"type":"void",
"type":"string",
"nickname":"force_major_compaction",
"produces":[
"application/json"
@@ -121,7 +121,7 @@
"description":"The minimum number of sstables in queue before compaction kicks off",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -172,7 +172,7 @@
"description":"The maximum number of sstables in queue before compaction kicks off",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -223,7 +223,7 @@
"description":"The maximum number of sstables in queue before compaction kicks off",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
},
{
@@ -231,7 +231,7 @@
"description":"The minimum number of sstables in queue before compaction kicks off",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -380,54 +380,16 @@
"operations":[
{
"method":"GET",
"summary":"check if the auto_compaction property is enabled for a given table",
"summary":"check if the auto compaction disabled",
"type":"boolean",
"nickname":"get_auto_compaction",
"nickname":"is_auto_compaction_disabled",
"produces":[
"application/json"
],
"parameters":[
{
"name":"name",
"description":"The table name in keyspace:name format",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
}
]
},
{
"method":"POST",
"summary":"Enable table auto compaction",
"type":"void",
"nickname":"enable_auto_compaction",
"produces":[
"application/json"
],
"parameters":[
{
"name":"name",
"description":"The table name in keyspace:name format",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
}
]
},
{
"method":"DELETE",
"summary":"Disable table auto compaction",
"type":"void",
"nickname":"disable_auto_compaction",
"produces":[
"application/json"
],
"parameters":[
{
"name":"name",
"description":"The table name in keyspace:name format",
"description":"The column family name in keyspace:name format",
"required":true,
"allowMultiple":false,
"type":"string",
@@ -582,7 +544,7 @@
"summary":"sstable count for each level. empty unless leveled compaction is used",
"type":"array",
"items":{
"type": "long"
"type":"int"
},
"nickname":"get_sstable_count_per_level",
"produces":[
@@ -649,54 +611,6 @@
}
]
},
{
"path":"/column_family/toppartitions/{name}",
"operations":[
{
"method":"GET",
"summary":"Toppartitions query",
"type":"toppartitions_query_results",
"nickname":"toppartitions",
"produces":[
"application/json"
],
"parameters":[
{
"name":"name",
"description":"The column family name in keyspace:name format",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
},
{
"name":"duration",
"description":"Duration (in milliseconds) of monitoring operation",
"required":true,
"allowMultiple":false,
"type": "long",
"paramType":"query"
},
{
"name":"list_size",
"description":"number of the top partitions to list",
"required":false,
"allowMultiple":false,
"type": "long",
"paramType":"query"
},
{
"name":"capacity",
"description":"capacity of stream summary: determines amount of resources used in query processing",
"required":false,
"allowMultiple":false,
"type": "long",
"paramType":"query"
}
]
}
]
},
{
"path":"/column_family/metrics/memtable_columns_count/",
"operations":[
@@ -959,7 +873,7 @@
{
"method":"GET",
"summary":"Get memtable switch count",
"type": "long",
"type":"int",
"nickname":"get_memtable_switch_count",
"produces":[
"application/json"
@@ -983,7 +897,7 @@
{
"method":"GET",
"summary":"Get all memtable switch count",
"type": "long",
"type":"int",
"nickname":"get_all_memtable_switch_count",
"produces":[
"application/json"
@@ -1120,7 +1034,7 @@
{
"method":"GET",
"summary":"Get read latency",
"type": "long",
"type":"int",
"nickname":"get_read_latency",
"produces":[
"application/json"
@@ -1273,7 +1187,7 @@
{
"method":"GET",
"summary":"Get all read latency",
"type": "long",
"type":"int",
"nickname":"get_all_read_latency",
"produces":[
"application/json"
@@ -1289,7 +1203,7 @@
{
"method":"GET",
"summary":"Get range latency",
"type": "long",
"type":"int",
"nickname":"get_range_latency",
"produces":[
"application/json"
@@ -1313,7 +1227,7 @@
{
"method":"GET",
"summary":"Get all range latency",
"type": "long",
"type":"int",
"nickname":"get_all_range_latency",
"produces":[
"application/json"
@@ -1329,7 +1243,7 @@
{
"method":"GET",
"summary":"Get write latency",
"type": "long",
"type":"int",
"nickname":"get_write_latency",
"produces":[
"application/json"
@@ -1482,7 +1396,7 @@
{
"method":"GET",
"summary":"Get all write latency",
"type": "long",
"type":"int",
"nickname":"get_all_write_latency",
"produces":[
"application/json"
@@ -1498,7 +1412,7 @@
{
"method":"GET",
"summary":"Get pending flushes",
"type": "long",
"type":"int",
"nickname":"get_pending_flushes",
"produces":[
"application/json"
@@ -1522,7 +1436,7 @@
{
"method":"GET",
"summary":"Get all pending flushes",
"type": "long",
"type":"int",
"nickname":"get_all_pending_flushes",
"produces":[
"application/json"
@@ -1538,7 +1452,7 @@
{
"method":"GET",
"summary":"Get pending compactions",
"type": "long",
"type":"int",
"nickname":"get_pending_compactions",
"produces":[
"application/json"
@@ -1562,7 +1476,7 @@
{
"method":"GET",
"summary":"Get all pending compactions",
"type": "long",
"type":"int",
"nickname":"get_all_pending_compactions",
"produces":[
"application/json"
@@ -1578,7 +1492,7 @@
{
"method":"GET",
"summary":"Get live ss table count",
"type": "long",
"type":"int",
"nickname":"get_live_ss_table_count",
"produces":[
"application/json"
@@ -1602,7 +1516,7 @@
{
"method":"GET",
"summary":"Get all live ss table count",
"type": "long",
"type":"int",
"nickname":"get_all_live_ss_table_count",
"produces":[
"application/json"
@@ -1618,7 +1532,7 @@
{
"method":"GET",
"summary":"Get live disk space used",
"type": "long",
"type":"int",
"nickname":"get_live_disk_space_used",
"produces":[
"application/json"
@@ -1642,7 +1556,7 @@
{
"method":"GET",
"summary":"Get all live disk space used",
"type": "long",
"type":"int",
"nickname":"get_all_live_disk_space_used",
"produces":[
"application/json"
@@ -1658,7 +1572,7 @@
{
"method":"GET",
"summary":"Get total disk space used",
"type": "long",
"type":"int",
"nickname":"get_total_disk_space_used",
"produces":[
"application/json"
@@ -1682,7 +1596,7 @@
{
"method":"GET",
"summary":"Get all total disk space used",
"type": "long",
"type":"int",
"nickname":"get_all_total_disk_space_used",
"produces":[
"application/json"
@@ -2138,7 +2052,7 @@
{
"method":"GET",
"summary":"Get speculative retries",
"type": "long",
"type":"int",
"nickname":"get_speculative_retries",
"produces":[
"application/json"
@@ -2162,7 +2076,7 @@
{
"method":"GET",
"summary":"Get all speculative retries",
"type": "long",
"type":"int",
"nickname":"get_all_speculative_retries",
"produces":[
"application/json"
@@ -2242,7 +2156,7 @@
{
"method":"GET",
"summary":"Get row cache hit out of range",
"type": "long",
"type":"int",
"nickname":"get_row_cache_hit_out_of_range",
"produces":[
"application/json"
@@ -2266,7 +2180,7 @@
{
"method":"GET",
"summary":"Get all row cache hit out of range",
"type": "long",
"type":"int",
"nickname":"get_all_row_cache_hit_out_of_range",
"produces":[
"application/json"
@@ -2282,7 +2196,7 @@
{
"method":"GET",
"summary":"Get row cache hit",
"type": "long",
"type":"int",
"nickname":"get_row_cache_hit",
"produces":[
"application/json"
@@ -2306,7 +2220,7 @@
{
"method":"GET",
"summary":"Get all row cache hit",
"type": "long",
"type":"int",
"nickname":"get_all_row_cache_hit",
"produces":[
"application/json"
@@ -2322,7 +2236,7 @@
{
"method":"GET",
"summary":"Get row cache miss",
"type": "long",
"type":"int",
"nickname":"get_row_cache_miss",
"produces":[
"application/json"
@@ -2346,7 +2260,7 @@
{
"method":"GET",
"summary":"Get all row cache miss",
"type": "long",
"type":"int",
"nickname":"get_all_row_cache_miss",
"produces":[
"application/json"
@@ -2362,7 +2276,7 @@
{
"method":"GET",
"summary":"Get cas prepare",
"type": "long",
"type":"int",
"nickname":"get_cas_prepare",
"produces":[
"application/json"
@@ -2386,7 +2300,7 @@
{
"method":"GET",
"summary":"Get cas propose",
"type": "long",
"type":"int",
"nickname":"get_cas_propose",
"produces":[
"application/json"
@@ -2410,7 +2324,7 @@
{
"method":"GET",
"summary":"Get cas commit",
"type": "long",
"type":"int",
"nickname":"get_cas_commit",
"produces":[
"application/json"
@@ -2902,44 +2816,6 @@
"description":"The column family type"
}
}
},
"toppartitions_record":{
"id":"toppartitions_record",
"description":"nodetool toppartitions query record",
"properties":{
"partition":{
"type":"string",
"description":"Partition key"
},
"count":{
"type":"long",
"description":"Number of read/write operations"
},
"error":{
"type":"long",
"description":"Indication of inaccuracy in counting PKs"
}
}
},
"toppartitions_query_results":{
"id":"toppartitions_query_results",
"description":"nodetool toppartitions query results",
"properties":{
"read":{
"type":"array",
"items":{
"type":"toppartitions_record"
},
"description":"Read results"
},
"write":{
"type":"array",
"items":{
"type":"toppartitions_record"
},
"description":"Write results"
}
}
}
}
}

View File

@@ -118,7 +118,7 @@
{
"method": "GET",
"summary": "Get pending tasks",
"type": "long",
"type": "int",
"nickname": "get_pending_tasks",
"produces": [
"application/json"
@@ -127,24 +127,6 @@
}
]
},
{
"path": "/compaction_manager/metrics/pending_tasks_by_table",
"operations": [
{
"method": "GET",
"summary": "Get pending tasks by table name",
"type": "array",
"items": {
"type": "pending_compaction"
},
"nickname": "get_pending_tasks_by_table",
"produces": [
"application/json"
],
"parameters": []
}
]
},
{
"path": "/compaction_manager/metrics/completed_tasks",
"operations": [
@@ -181,7 +163,7 @@
{
"method": "GET",
"summary": "Get bytes compacted",
"type": "long",
"type": "int",
"nickname": "get_bytes_compacted",
"produces": [
"application/json"
@@ -197,7 +179,7 @@
"description":"A row merged information",
"properties":{
"key":{
"type": "long",
"type":"int",
"description":"The number of sstable"
},
"value":{
@@ -262,23 +244,6 @@
}
}
},
"pending_compaction": {
"id": "pending_compaction",
"properties": {
"cf": {
"type": "string",
"description": "The column family name"
},
"ks": {
"type":"string",
"description": "The keyspace name"
},
"task": {
"type":"long",
"description": "The number of pending tasks"
}
}
},
"history": {
"id":"history",
"description":"Compaction history information",

View File

@@ -1,90 +0,0 @@
{
"apiVersion":"0.0.1",
"swaggerVersion":"1.2",
"basePath":"{{Protocol}}://{{Host}}",
"resourcePath":"/error_injection",
"produces":[
"application/json"
],
"apis":[
{
"path":"/v2/error_injection/injection/{injection}",
"operations":[
{
"method":"POST",
"summary":"Activate an injection that triggers an error in code",
"type":"void",
"nickname":"enable_injection",
"produces":[
"application/json"
],
"parameters":[
{
"name":"injection",
"description":"injection name, should correspond to an injection added in code",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
},
{
"name":"one_shot",
"description":"boolean flag indicating whether the injection should be enabled to trigger only once",
"required":false,
"allowMultiple":false,
"type":"boolean",
"paramType":"query"
}
]
},
{
"method":"DELETE",
"summary":"Deactivate an injection previously activated by the API",
"type":"void",
"nickname":"disable_injection",
"produces":[
"application/json"
],
"parameters":[
{
"name":"injection",
"description":"injection name",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
}
]
}
]
},
{
"path":"/v2/error_injection/injection",
"operations":[
{
"method":"GET",
"summary":"List all enabled injections on all shards, i.e. injections that will trigger an error in the code",
"type":"array",
"items":{
"type":"string"
},
"nickname":"get_enabled_injections_on_all",
"produces":[
"application/json"
],
"parameters":[]
},
{
"method":"DELETE",
"summary":"Deactivate all injections previously activated on all shards by the API",
"type":"void",
"nickname":"disable_on_all",
"produces":[
"application/json"
],
"parameters":[]
}
]
}
]
}

View File

@@ -110,7 +110,7 @@
{
"method":"GET",
"summary":"Get count down endpoint",
"type": "long",
"type":"int",
"nickname":"get_down_endpoint_count",
"produces":[
"application/json"
@@ -126,7 +126,7 @@
{
"method":"GET",
"summary":"Get count up endpoint",
"type": "long",
"type":"int",
"nickname":"get_up_endpoint_count",
"produces":[
"application/json"
@@ -180,11 +180,11 @@
"description": "The endpoint address"
},
"generation": {
"type": "long",
"type": "int",
"description": "The heart beat generation"
},
"version": {
"type": "long",
"type": "int",
"description": "The heart beat version"
},
"update_time": {
@@ -209,7 +209,7 @@
"description": "Holds a version value for an application state",
"properties": {
"application_state": {
"type": "long",
"type": "int",
"description": "The application state enum index"
},
"value": {
@@ -217,7 +217,7 @@
"description": "The version value"
},
"version": {
"type": "long",
"type": "int",
"description": "The application state version"
}
}

View File

@@ -75,7 +75,7 @@
{
"method":"GET",
"summary":"Returns files which are pending for archival attempt. Does NOT include failed archive attempts",
"type": "long",
"type":"int",
"nickname":"get_current_generation_number",
"produces":[
"application/json"
@@ -99,7 +99,7 @@
{
"method":"GET",
"summary":"Get heart beat version for a node",
"type": "long",
"type":"int",
"nickname":"get_current_heart_beat_version",
"produces":[
"application/json"

View File

@@ -99,7 +99,7 @@
{
"method": "GET",
"summary": "Get create hint count",
"type": "long",
"type": "int",
"nickname": "get_create_hint_count",
"produces": [
"application/json"
@@ -123,7 +123,7 @@
{
"method": "GET",
"summary": "Get not stored hints count",
"type": "long",
"type": "int",
"nickname": "get_not_stored_hints_count",
"produces": [
"application/json"

View File

@@ -191,7 +191,7 @@
{
"method":"GET",
"summary":"Get the version number",
"type": "long",
"type":"int",
"nickname":"get_version",
"produces":[
"application/json"
@@ -249,7 +249,7 @@
"MIGRATION_REQUEST",
"PREPARE_MESSAGE",
"PREPARE_DONE_MESSAGE",
"UNUSED__STREAM_MUTATION",
"STREAM_MUTATION",
"STREAM_MUTATION_DONE",
"COMPLETE_MESSAGE",
"REPAIR_CHECKSUM_RANGE",

View File

@@ -68,7 +68,7 @@
"summary":"Get the hinted handoff enabled by dc",
"type":"array",
"items":{
"type":"array"
"type":"mapper_list"
},
"nickname":"get_hinted_handoff_enabled_by_dc",
"produces":[
@@ -105,7 +105,7 @@
{
"method":"GET",
"summary":"Get the max hint window",
"type": "long",
"type":"int",
"nickname":"get_max_hint_window",
"produces":[
"application/json"
@@ -128,7 +128,7 @@
"description":"max hint window in ms",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -141,7 +141,7 @@
{
"method":"GET",
"summary":"Get max hints in progress",
"type": "long",
"type":"int",
"nickname":"get_max_hints_in_progress",
"produces":[
"application/json"
@@ -164,7 +164,7 @@
"description":"max hints in progress",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -177,7 +177,7 @@
{
"method":"GET",
"summary":"get hints in progress",
"type": "long",
"type":"int",
"nickname":"get_hints_in_progress",
"produces":[
"application/json"
@@ -602,7 +602,7 @@
{
"method": "GET",
"summary": "Get cas write metrics",
"type": "long",
"type": "int",
"nickname": "get_cas_write_metrics_unfinished_commit",
"produces": [
"application/json"
@@ -632,7 +632,7 @@
{
"method": "GET",
"summary": "Get cas write metrics",
"type": "long",
"type": "int",
"nickname": "get_cas_write_metrics_condition_not_met",
"produces": [
"application/json"
@@ -641,28 +641,13 @@
}
]
},
{
"path": "/storage_proxy/metrics/cas_write/failed_read_round_optimization",
"operations": [
{
"method": "GET",
"summary": "Get cas write metrics",
"type": "long",
"nickname": "get_cas_write_metrics_failed_read_round_optimization",
"produces": [
"application/json"
],
"parameters": []
}
]
},
{
"path": "/storage_proxy/metrics/cas_read/unfinished_commit",
"operations": [
{
"method": "GET",
"summary": "Get cas read metrics",
"type": "long",
"type": "int",
"nickname": "get_cas_read_metrics_unfinished_commit",
"produces": [
"application/json"
@@ -686,13 +671,28 @@
}
]
},
{
"path": "/storage_proxy/metrics/cas_read/condition_not_met",
"operations": [
{
"method": "GET",
"summary": "Get cas read metrics",
"type": "int",
"nickname": "get_cas_read_metrics_condition_not_met",
"produces": [
"application/json"
],
"parameters": []
}
]
},
{
"path": "/storage_proxy/metrics/read/timeouts",
"operations": [
{
"method": "GET",
"summary": "Get read metrics",
"type": "long",
"type": "int",
"nickname": "get_read_metrics_timeouts",
"produces": [
"application/json"
@@ -707,7 +707,7 @@
{
"method": "GET",
"summary": "Get read metrics",
"type": "long",
"type": "int",
"nickname": "get_read_metrics_unavailables",
"produces": [
"application/json"
@@ -791,36 +791,6 @@
}
]
},
{
"path": "/storage_proxy/metrics/cas_read/moving_average_histogram",
"operations": [
{
"method": "GET",
"summary": "Get CAS read rate and latency histogram",
"$ref": "#/utils/rate_moving_average_and_histogram",
"nickname": "get_cas_read_metrics_latency_histogram",
"produces": [
"application/json"
],
"parameters": []
}
]
},
{
"path": "/storage_proxy/metrics/view_write/moving_average_histogram",
"operations": [
{
"method": "GET",
"summary": "Get view write rate and latency histogram",
"$ref": "#/utils/rate_moving_average_and_histogram",
"nickname": "get_view_write_metrics_latency_histogram",
"produces": [
"application/json"
],
"parameters": []
}
]
},
{
"path": "/storage_proxy/metrics/range/moving_average_histogram",
"operations": [
@@ -842,7 +812,7 @@
{
"method": "GET",
"summary": "Get range metrics",
"type": "long",
"type": "int",
"nickname": "get_range_metrics_timeouts",
"produces": [
"application/json"
@@ -857,7 +827,7 @@
{
"method": "GET",
"summary": "Get range metrics",
"type": "long",
"type": "int",
"nickname": "get_range_metrics_unavailables",
"produces": [
"application/json"
@@ -902,7 +872,7 @@
{
"method": "GET",
"summary": "Get write metrics",
"type": "long",
"type": "int",
"nickname": "get_write_metrics_timeouts",
"produces": [
"application/json"
@@ -917,7 +887,7 @@
{
"method": "GET",
"summary": "Get write metrics",
"type": "long",
"type": "int",
"nickname": "get_write_metrics_unavailables",
"produces": [
"application/json"
@@ -986,21 +956,6 @@
}
]
},
{
"path": "/storage_proxy/metrics/cas_write/moving_average_histogram",
"operations": [
{
"method": "GET",
"summary": "Get CAS write rate and latency histogram",
"$ref": "#/utils/rate_moving_average_and_histogram",
"nickname": "get_cas_write_metrics_latency_histogram",
"produces": [
"application/json"
],
"parameters": []
}
]
},
{
"path":"/storage_proxy/metrics/read/estimated_histogram/",
"operations":[
@@ -1023,7 +978,7 @@
{
"method":"GET",
"summary":"Get read latency",
"type": "long",
"type":"int",
"nickname":"get_read_latency",
"produces":[
"application/json"
@@ -1055,7 +1010,7 @@
{
"method":"GET",
"summary":"Get write latency",
"type": "long",
"type":"int",
"nickname":"get_write_latency",
"produces":[
"application/json"
@@ -1087,7 +1042,7 @@
{
"method":"GET",
"summary":"Get range latency",
"type": "long",
"type":"int",
"nickname":"get_range_latency",
"produces":[
"application/json"

View File

@@ -458,7 +458,7 @@
{
"method":"GET",
"summary":"Return the generation value for this node.",
"type": "long",
"type":"int",
"nickname":"get_current_generation_number",
"produces":[
"application/json"
@@ -511,21 +511,6 @@
}
]
},
{
"path":"/storage_service/cdc_streams_check_and_repair",
"operations":[
{
"method":"POST",
"summary":"Checks that CDC streams reflect current cluster topology and regenerates them if not.",
"type":"void",
"nickname":"cdc_streams_check_and_repair",
"produces":[
"application/json"
],
"parameters":[]
}
]
},
{
"path":"/storage_service/snapshots",
"operations":[
@@ -597,15 +582,7 @@
},
{
"name":"kn",
"description":"Comma seperated keyspaces name that their snapshot will be deleted",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
},
{
"name":"cf",
"description":"an optional table name that its snapshot will be deleted",
"description":"Comma seperated keyspaces name to snapshot",
"required":false,
"allowMultiple":false,
"type":"string",
@@ -669,7 +646,7 @@
{
"method":"POST",
"summary":"Trigger a cleanup of keys on a single keyspace",
"type": "long",
"type":"int",
"nickname":"force_keyspace_cleanup",
"produces":[
"application/json"
@@ -701,7 +678,7 @@
{
"method":"GET",
"summary":"Scrub (deserialize + reserialize at the latest version, skipping bad rows if any) the given keyspace. If columnFamilies array is empty, all CFs are scrubbed. Scrubbed CFs will be snapshotted first, if disableSnapshot is false",
"type": "long",
"type":"int",
"nickname":"scrub",
"produces":[
"application/json"
@@ -749,7 +726,7 @@
{
"method":"GET",
"summary":"Rewrite all sstables to the latest version. Unlike scrub, it doesn't skip bad rows and do not snapshot sstables first.",
"type": "long",
"type":"int",
"nickname":"upgrade_sstables",
"produces":[
"application/json"
@@ -823,7 +800,7 @@
"summary":"Return an array with the ids of the currently active repairs",
"type":"array",
"items":{
"type": "long"
"type":"int"
},
"nickname":"get_active_repair_async",
"produces":[
@@ -833,50 +810,13 @@
}
]
},
{
"path":"/storage_service/repair_status/",
"operations":[
{
"method":"GET",
"summary":"Query the repair status and return when the repair is finished or timeout",
"type":"string",
"enum":[
"RUNNING",
"SUCCESSFUL",
"FAILED"
],
"nickname":"repair_await_completion",
"produces":[
"application/json"
],
"parameters":[
{
"name":"id",
"description":"The repair ID to check for status",
"required":true,
"allowMultiple":false,
"type": "long",
"paramType":"query"
},
{
"name":"timeout",
"description":"Seconds to wait before the query returns even if the repair is not finished. The value -1 or not providing this parameter means no timeout",
"required":false,
"allowMultiple":false,
"type": "long",
"paramType":"query"
}
]
}
]
},
{
"path":"/storage_service/repair_async/{keyspace}",
"operations":[
{
"method":"POST",
"summary":"Invoke repair asynchronously. You can track repair progress by using the get supplying id",
"type": "long",
"type":"int",
"nickname":"repair_async",
"produces":[
"application/json"
@@ -1007,7 +947,7 @@
"description":"The repair ID to check for status",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -1337,18 +1277,18 @@
},
{
"name":"dynamic_update_interval",
"description":"interval in ms (default 100)",
"description":"integer, in ms (default 100)",
"required":false,
"allowMultiple":false,
"type":"long",
"type":"integer",
"paramType":"query"
},
{
"name":"dynamic_reset_interval",
"description":"interval in ms (default 600,000)",
"description":"integer, in ms (default 600,000)",
"required":false,
"allowMultiple":false,
"type":"long",
"type":"integer",
"paramType":"query"
},
{
@@ -1553,7 +1493,7 @@
"description":"Stream throughput",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -1561,7 +1501,7 @@
{
"method":"GET",
"summary":"Get stream throughput mb per sec",
"type": "long",
"type":"int",
"nickname":"get_stream_throughput_mb_per_sec",
"produces":[
"application/json"
@@ -1577,7 +1517,7 @@
{
"method":"GET",
"summary":"get compaction throughput mb per sec",
"type": "long",
"type":"int",
"nickname":"get_compaction_throughput_mb_per_sec",
"produces":[
"application/json"
@@ -1599,7 +1539,7 @@
"description":"compaction throughput",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -2003,7 +1943,7 @@
{
"method":"GET",
"summary":"Returns the threshold for warning of queries with many tombstones",
"type": "long",
"type":"int",
"nickname":"get_tombstone_warn_threshold",
"produces":[
"application/json"
@@ -2025,7 +1965,7 @@
"description":"tombstone debug threshold",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -2038,7 +1978,7 @@
{
"method":"GET",
"summary":"",
"type": "long",
"type":"int",
"nickname":"get_tombstone_failure_threshold",
"produces":[
"application/json"
@@ -2060,7 +2000,7 @@
"description":"tombstone debug threshold",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -2073,7 +2013,7 @@
{
"method":"GET",
"summary":"Returns the threshold for rejecting queries due to a large batch size",
"type": "long",
"type":"int",
"nickname":"get_batch_size_failure_threshold",
"produces":[
"application/json"
@@ -2095,7 +2035,7 @@
"description":"batch size debug threshold",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -2119,7 +2059,7 @@
"description":"throttle in kb",
"required":true,
"allowMultiple":false,
"type": "long",
"type":"int",
"paramType":"query"
}
]
@@ -2132,7 +2072,7 @@
{
"method":"GET",
"summary":"Get load",
"type": "long",
"type":"int",
"nickname":"get_metrics_load",
"produces":[
"application/json"
@@ -2148,7 +2088,7 @@
{
"method":"GET",
"summary":"Get exceptions",
"type": "long",
"type":"int",
"nickname":"get_exceptions",
"produces":[
"application/json"
@@ -2164,7 +2104,7 @@
{
"method":"GET",
"summary":"Get total hints in progress",
"type": "long",
"type":"int",
"nickname":"get_total_hints_in_progress",
"produces":[
"application/json"
@@ -2180,7 +2120,7 @@
{
"method":"GET",
"summary":"Get total hints",
"type": "long",
"type":"int",
"nickname":"get_total_hints",
"produces":[
"application/json"
@@ -2224,42 +2164,7 @@
]
}
]
},
{
"path":"/storage_service/sstable_info",
"operations":[
{
"method":"GET",
"summary":"SSTable information",
"type":"array",
"items":{
"type":"table_sstables"
},
"nickname":"sstable_info",
"produces":[
"application/json"
],
"parameters":[
{
"name":"keyspace",
"description":"The keyspace",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
},
{
"name":"cf",
"description":"column family name",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
}
]
}
]
}
}
],
"models":{
"mapper":{
@@ -2419,92 +2324,6 @@
"description":"The endpoint details"
}
}
},
"named_maps":{
"id":"named_maps",
"properties":{
"group":{
"type":"string"
},
"attributes":{
"type":"array",
"items":{
"type":"mapper"
}
}
}
},
"sstable":{
"id":"sstable",
"properties":{
"size":{
"type":"long",
"description":"Total size in bytes of sstable"
},
"data_size":{
"type":"long",
"description":"The size in bytes on disk of data"
},
"index_size":{
"type":"long",
"description":"The size in bytes on disk of index"
},
"filter_size":{
"type":"long",
"description":"The size in bytes on disk of filter"
},
"timestamp":{
"type":"datetime",
"description":"File creation time"
},
"generation":{
"type":"long",
"description":"SSTable generation"
},
"level":{
"type":"long",
"description":"SSTable level"
},
"version":{
"type":"string",
"enum":[
"ka", "la", "mc", "md"
],
"description":"SSTable version"
},
"properties":{
"type":"array",
"description":"SSTable attributes",
"items":{
"type":"mapper"
}
},
"extended_properties":{
"type":"array",
"description":"SSTable extended attributes",
"items":{
"type":"named_maps"
}
}
}
},
"table_sstables":{
"id":"table_sstables",
"description":"Per-table SSTable info and attributes",
"properties":{
"keyspace":{
"type":"string"
},
"table":{
"type":"string"
},
"sstables":{
"type":"array",
"items":{
"$ref":"sstable"
}
}
}
}
}
}

View File

@@ -32,7 +32,7 @@
{
"method":"GET",
"summary":"Get number of active outbound streams",
"type": "long",
"type":"int",
"nickname":"get_all_active_streams_outbound",
"produces":[
"application/json"
@@ -48,7 +48,7 @@
{
"method":"GET",
"summary":"Get total incoming bytes",
"type": "long",
"type":"int",
"nickname":"get_total_incoming_bytes",
"produces":[
"application/json"
@@ -72,7 +72,7 @@
{
"method":"GET",
"summary":"Get all total incoming bytes",
"type": "long",
"type":"int",
"nickname":"get_all_total_incoming_bytes",
"produces":[
"application/json"
@@ -88,7 +88,7 @@
{
"method":"GET",
"summary":"Get total outgoing bytes",
"type": "long",
"type":"int",
"nickname":"get_total_outgoing_bytes",
"produces":[
"application/json"
@@ -112,7 +112,7 @@
{
"method":"GET",
"summary":"Get all total outgoing bytes",
"type": "long",
"type":"int",
"nickname":"get_all_total_outgoing_bytes",
"produces":[
"application/json"
@@ -154,7 +154,7 @@
"description":"The peer"
},
"session_index":{
"type": "long",
"type":"int",
"description":"The session index"
},
"connecting":{
@@ -211,7 +211,7 @@
"description":"The ID"
},
"files":{
"type": "long",
"type":"int",
"description":"Number of files to transfer. Can be 0 if nothing to transfer for some streaming request."
},
"total_size":{
@@ -242,7 +242,7 @@
"description":"The peer address"
},
"session_index":{
"type": "long",
"type":"int",
"description":"The session index"
},
"file_name":{

View File

@@ -52,21 +52,6 @@
}
]
},
{
"path":"/system/uptime_ms",
"operations":[
{
"method":"GET",
"summary":"Get system uptime, in milliseconds",
"type":"long",
"nickname":"get_system_uptime",
"produces":[
"application/json"
],
"parameters":[]
}
]
},
{
"path":"/system/logger/{name}",
"operations":[

View File

@@ -20,9 +20,9 @@
*/
#include "api.hh"
#include <seastar/http/file_handler.hh>
#include <seastar/http/transformers.hh>
#include <seastar/http/api_docs.hh>
#include "http/file_handler.hh"
#include "http/transformers.hh"
#include "http/api_docs.hh"
#include "storage_service.hh"
#include "commitlog.hh"
#include "gossiper.hh"
@@ -36,14 +36,11 @@
#include "endpoint_snitch.hh"
#include "compaction_manager.hh"
#include "hinted_handoff.hh"
#include "error_injection.hh"
#include <seastar/http/exception.hh>
#include "http/exception.hh"
#include "stream_manager.hh"
#include "system.hh"
#include "api/config.hh"
logging::logger apilog("api");
namespace api {
static std::unique_ptr<reply> exception_reply(std::exception_ptr eptr) {
@@ -69,19 +66,13 @@ future<> set_server_init(http_context& ctx) {
rb->set_api_doc(r);
rb02->set_api_doc(r);
rb02->register_api_file(r, "swagger20_header");
set_config(rb02, ctx, r);
rb->register_function(r, "system",
"The system related API");
set_system(ctx, r);
});
}
future<> set_server_config(http_context& ctx) {
auto rb02 = std::make_shared < api_registry_builder20 > (ctx.api_doc, "/v2");
return ctx.http_server.set_routes([&ctx, rb02](routes& r) {
set_config(rb02, ctx, r);
});
}
static future<> register_api(http_context& ctx, const sstring& api_name,
const sstring api_desc,
std::function<void(http_context& ctx, routes& r)> f) {
@@ -93,42 +84,10 @@ static future<> register_api(http_context& ctx, const sstring& api_name,
});
}
future<> set_transport_controller(http_context& ctx, cql_transport::controller& ctl) {
return ctx.http_server.set_routes([&ctx, &ctl] (routes& r) { set_transport_controller(ctx, r, ctl); });
}
future<> unset_transport_controller(http_context& ctx) {
return ctx.http_server.set_routes([&ctx] (routes& r) { unset_transport_controller(ctx, r); });
}
future<> set_rpc_controller(http_context& ctx, thrift_controller& ctl) {
return ctx.http_server.set_routes([&ctx, &ctl] (routes& r) { set_rpc_controller(ctx, r, ctl); });
}
future<> unset_rpc_controller(http_context& ctx) {
return ctx.http_server.set_routes([&ctx] (routes& r) { unset_rpc_controller(ctx, r); });
}
future<> set_server_storage_service(http_context& ctx) {
return register_api(ctx, "storage_service", "The storage service API", set_storage_service);
}
future<> set_server_repair(http_context& ctx, sharded<netw::messaging_service>& ms) {
return ctx.http_server.set_routes([&ctx, &ms] (routes& r) { set_repair(ctx, r, ms); });
}
future<> unset_server_repair(http_context& ctx) {
return ctx.http_server.set_routes([&ctx] (routes& r) { unset_repair(ctx, r); });
}
future<> set_server_snapshot(http_context& ctx, sharded<db::snapshot_ctl>& snap_ctl) {
return ctx.http_server.set_routes([&ctx, &snap_ctl] (routes& r) { set_snapshot(ctx, r, snap_ctl); });
}
future<> unset_server_snapshot(http_context& ctx) {
return ctx.http_server.set_routes([&ctx] (routes& r) { unset_snapshot(ctx, r); });
}
future<> set_server_snitch(http_context& ctx) {
return register_api(ctx, "endpoint_snitch_info", "The endpoint snitch info API", set_endpoint_snitch);
}
@@ -143,14 +102,9 @@ future<> set_server_load_sstable(http_context& ctx) {
"The column family API", set_column_family);
}
future<> set_server_messaging_service(http_context& ctx, sharded<netw::messaging_service>& ms) {
future<> set_server_messaging_service(http_context& ctx) {
return register_api(ctx, "messaging_service",
"The messaging service API", [&ms] (http_context& ctx, routes& r) {
set_messaging_service(ctx, r, ms);
});
}
future<> unset_server_messaging_service(http_context& ctx) {
return ctx.http_server.set_routes([&ctx] (routes& r) { unset_messaging_service(ctx, r); });
"The messaging service API", set_messaging_service);
}
future<> set_server_storage_proxy(http_context& ctx) {
@@ -197,9 +151,6 @@ future<> set_server_done(http_context& ctx) {
rb->register_function(r, "collectd",
"The collectd API");
set_collectd(ctx, r);
rb->register_function(r, "error_injection",
"The error injection API");
set_error_injection(ctx, r);
});
}

View File

@@ -21,15 +21,13 @@
#pragma once
#include <seastar/json/json_elements.hh>
#include <type_traits>
#include "json/json_elements.hh"
#include <boost/lexical_cast.hpp>
#include <boost/algorithm/string/split.hpp>
#include <boost/algorithm/string/classification.hpp>
#include <boost/units/detail/utility.hpp>
#include "api/api-doc/utils.json.hh"
#include "utils/histogram.hh"
#include <seastar/http/exception.hh>
#include "http/exception.hh"
#include "api_init.hh"
#include "seastarx.hh"
@@ -218,44 +216,4 @@ std::vector<T> concat(std::vector<T> a, std::vector<T>&& b) {
return a;
}
template <class T, class Base = T>
class req_param {
public:
sstring name;
sstring param;
T value;
req_param(const request& req, sstring name, T default_val) : name(name) {
param = req.get_query_param(name);
if (param.empty()) {
value = default_val;
return;
}
try {
// boost::lexical_cast does not use boolalpha. Converting a
// true/false throws exceptions. We don't want that.
if constexpr (std::is_same_v<Base, bool>) {
// Cannot use boolalpha because we (probably) want to
// accept 1 and 0 as well as true and false. And True. And fAlse.
std::transform(param.begin(), param.end(), param.begin(), ::tolower);
if (param == "true" || param == "1") {
value = T(true);
} else if (param == "false" || param == "0") {
value = T(false);
} else {
throw boost::bad_lexical_cast{};
}
} else {
value = T{boost::lexical_cast<Base>(param)};
}
} catch (boost::bad_lexical_cast&) {
throw bad_param_exception(format("{} ({}): type error - should be {}", name, param, boost::units::detail::demangle(typeid(Base).name())));
}
}
operator T() const { return value; }
};
utils_json::estimated_histogram time_to_json_histogram(const utils::time_estimated_histogram& val);
}

View File

@@ -19,16 +19,9 @@
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include "database_fwd.hh"
#include "database.hh"
#include "service/storage_proxy.hh"
#include <seastar/http/httpd.hh>
namespace service { class load_meter; }
namespace locator { class shared_token_metadata; }
namespace cql_transport { class controller; }
class thrift_controller;
namespace db { class snapshot_ctl; }
namespace netw { class messaging_service; }
#include "http/httpd.hh"
namespace api {
@@ -38,34 +31,18 @@ struct http_context {
httpd::http_server_control http_server;
distributed<database>& db;
distributed<service::storage_proxy>& sp;
service::load_meter& lmeter;
const sharded<locator::shared_token_metadata>& shared_token_metadata;
http_context(distributed<database>& _db,
distributed<service::storage_proxy>& _sp,
service::load_meter& _lm, const sharded<locator::shared_token_metadata>& _stm)
: db(_db), sp(_sp), lmeter(_lm), shared_token_metadata(_stm) {
distributed<service::storage_proxy>& _sp)
: db(_db), sp(_sp) {
}
const locator::token_metadata& get_token_metadata();
};
future<> set_server_init(http_context& ctx);
future<> set_server_config(http_context& ctx);
future<> set_server_snitch(http_context& ctx);
future<> set_server_storage_service(http_context& ctx);
future<> set_server_repair(http_context& ctx, sharded<netw::messaging_service>& ms);
future<> unset_server_repair(http_context& ctx);
future<> set_transport_controller(http_context& ctx, cql_transport::controller& ctl);
future<> unset_transport_controller(http_context& ctx);
future<> set_rpc_controller(http_context& ctx, thrift_controller& ctl);
future<> unset_rpc_controller(http_context& ctx);
future<> set_server_snapshot(http_context& ctx, sharded<db::snapshot_ctl>& snap_ctl);
future<> unset_server_snapshot(http_context& ctx);
future<> set_server_gossip(http_context& ctx);
future<> set_server_load_sstable(http_context& ctx);
future<> set_server_messaging_service(http_context& ctx, sharded<netw::messaging_service>& ms);
future<> unset_server_messaging_service(http_context& ctx);
future<> set_server_messaging_service(http_context& ctx);
future<> set_server_storage_proxy(http_context& ctx);
future<> set_server_stream_manager(http_context& ctx);
future<> set_server_gossip_settle(http_context& ctx);

View File

@@ -208,11 +208,9 @@ void set_cache_service(http_context& ctx, routes& r) {
});
cs::get_row_capacity.set(r, [&ctx] (std::unique_ptr<request> req) {
return ctx.db.map_reduce0([](database& db) -> uint64_t {
return db.row_cache_tracker().region().occupancy().used_space();
}, uint64_t(0), std::plus<uint64_t>()).then([](const int64_t& res) {
return make_ready_future<json::json_return_type>(res);
});
return map_reduce_cf(ctx, uint64_t(0), [](const column_family& cf) {
return cf.get_row_cache().get_cache_tracker().region().occupancy().used_space();
}, std::plus<uint64_t>());
});
cs::get_row_hits.set(r, [&ctx] (std::unique_ptr<request> req) {
@@ -253,19 +251,15 @@ void set_cache_service(http_context& ctx, routes& r) {
cs::get_row_size.set(r, [&ctx] (std::unique_ptr<request> req) {
// In origin row size is the weighted size.
// We currently do not support weights, so we use num entries instead
return ctx.db.map_reduce0([](database& db) -> uint64_t {
return db.row_cache_tracker().partitions();
}, uint64_t(0), std::plus<uint64_t>()).then([](const int64_t& res) {
return make_ready_future<json::json_return_type>(res);
});
return map_reduce_cf(ctx, 0, [](const column_family& cf) {
return cf.get_row_cache().partitions();
}, std::plus<uint64_t>());
});
cs::get_row_entries.set(r, [&ctx] (std::unique_ptr<request> req) {
return ctx.db.map_reduce0([](database& db) -> uint64_t {
return db.row_cache_tracker().partitions();
}, uint64_t(0), std::plus<uint64_t>()).then([](const int64_t& res) {
return make_ready_future<json::json_return_type>(res);
});
return map_reduce_cf(ctx, 0, [](const column_family& cf) {
return cf.get_row_cache().partitions();
}, std::plus<uint64_t>());
});
cs::get_counter_capacity.set(r, [] (std::unique_ptr<request> req) {

View File

@@ -21,8 +21,8 @@
#include "collectd.hh"
#include "api/api-doc/collectd.json.hh"
#include <seastar/core/scollectd.hh>
#include <seastar/core/scollectd_api.hh>
#include "core/scollectd.hh"
#include "core/scollectd_api.hh"
#include "endian.h"
#include <boost/range/irange.hpp>
#include <regex>
@@ -64,7 +64,7 @@ static const char* str_to_regex(const sstring& v) {
void set_collectd(http_context& ctx, routes& r) {
cd::get_collectd.set(r, [&ctx](std::unique_ptr<request> req) {
auto id = ::make_shared<scollectd::type_instance_id>(req->param["pluginid"],
auto id = make_shared<scollectd::type_instance_id>(req->param["pluginid"],
req->get_query_param("instance"), req->get_query_param("type"),
req->get_query_param("type_instance"));

View File

@@ -22,14 +22,10 @@
#include "column_family.hh"
#include "api/api-doc/column_family.json.hh"
#include <vector>
#include <seastar/http/exception.hh>
#include "http/exception.hh"
#include "sstables/sstables.hh"
#include "utils/estimated_histogram.hh"
#include <algorithm>
#include "db/system_keyspace_view_types.hh"
#include "db/data_listeners.hh"
extern logging::logger apilog;
namespace api {
using namespace httpd;
@@ -38,7 +34,7 @@ using namespace std;
using namespace json;
namespace cf = httpd::column_family_json;
std::tuple<sstring, sstring> parse_fully_qualified_cf_name(sstring name) {
const utils::UUID& get_uuid(const sstring& name, const database& db) {
auto pos = name.find("%3A");
size_t end;
if (pos == sstring::npos) {
@@ -50,22 +46,14 @@ std::tuple<sstring, sstring> parse_fully_qualified_cf_name(sstring name) {
} else {
end = pos + 3;
}
return std::make_tuple(name.substr(0, pos), name.substr(end));
}
const utils::UUID& get_uuid(const sstring& ks, const sstring& cf, const database& db) {
try {
return db.find_uuid(ks, cf);
return db.find_uuid(name.substr(0, pos), name.substr(end));
} catch (std::out_of_range& e) {
throw bad_param_exception(format("Column family '{}:{}' not found", ks, cf));
throw bad_param_exception("Column family '" + name.substr(0, pos) + ":"
+ name.substr(end) + "' not found");
}
}
const utils::UUID& get_uuid(const sstring& name, const database& db) {
auto [ks, cf] = parse_fully_qualified_cf_name(name);
return get_uuid(ks, cf, db);
}
future<> foreach_column_family(http_context& ctx, const sstring& name, function<void(column_family&)> f) {
auto uuid = get_uuid(name, ctx.db.local());
@@ -75,28 +63,28 @@ future<> foreach_column_family(http_context& ctx, const sstring& name, function<
}
future<json::json_return_type> get_cf_stats(http_context& ctx, const sstring& name,
int64_t column_family_stats::*f) {
int64_t column_family::stats::*f) {
return map_reduce_cf(ctx, name, int64_t(0), [f](const column_family& cf) {
return cf.get_stats().*f;
}, std::plus<int64_t>());
}
future<json::json_return_type> get_cf_stats(http_context& ctx,
int64_t column_family_stats::*f) {
int64_t column_family::stats::*f) {
return map_reduce_cf(ctx, int64_t(0), [f](const column_family& cf) {
return cf.get_stats().*f;
}, std::plus<int64_t>());
}
static future<json::json_return_type> get_cf_stats_count(http_context& ctx, const sstring& name,
utils::timed_rate_moving_average_and_histogram column_family_stats::*f) {
utils::timed_rate_moving_average_and_histogram column_family::stats::*f) {
return map_reduce_cf(ctx, name, int64_t(0), [f](const column_family& cf) {
return (cf.get_stats().*f).hist.count;
}, std::plus<int64_t>());
}
static future<json::json_return_type> get_cf_stats_sum(http_context& ctx, const sstring& name,
utils::timed_rate_moving_average_and_histogram column_family_stats::*f) {
utils::timed_rate_moving_average_and_histogram column_family::stats::*f) {
auto uuid = get_uuid(name, ctx.db.local());
return ctx.db.map_reduce0([uuid, f](database& db) {
// Histograms information is sample of the actual load
@@ -112,14 +100,14 @@ static future<json::json_return_type> get_cf_stats_sum(http_context& ctx, const
static future<json::json_return_type> get_cf_stats_count(http_context& ctx,
utils::timed_rate_moving_average_and_histogram column_family_stats::*f) {
utils::timed_rate_moving_average_and_histogram column_family::stats::*f) {
return map_reduce_cf(ctx, int64_t(0), [f](const column_family& cf) {
return (cf.get_stats().*f).hist.count;
}, std::plus<int64_t>());
}
static future<json::json_return_type> get_cf_histogram(http_context& ctx, const sstring& name,
utils::timed_rate_moving_average_and_histogram column_family_stats::*f) {
utils::timed_rate_moving_average_and_histogram column_family::stats::*f) {
utils::UUID uuid = get_uuid(name, ctx.db.local());
return ctx.db.map_reduce0([f, uuid](const database& p) {
return (p.find_column_family(uuid).get_stats().*f).hist;},
@@ -130,7 +118,7 @@ static future<json::json_return_type> get_cf_histogram(http_context& ctx, const
});
}
static future<json::json_return_type> get_cf_histogram(http_context& ctx, utils::timed_rate_moving_average_and_histogram column_family_stats::*f) {
static future<json::json_return_type> get_cf_histogram(http_context& ctx, utils::timed_rate_moving_average_and_histogram column_family::stats::*f) {
std::function<utils::ihistogram(const database&)> fun = [f] (const database& db) {
utils::ihistogram res;
for (auto i : db.get_column_families()) {
@@ -146,7 +134,7 @@ static future<json::json_return_type> get_cf_histogram(http_context& ctx, utils:
}
static future<json::json_return_type> get_cf_rate_and_histogram(http_context& ctx, const sstring& name,
utils::timed_rate_moving_average_and_histogram column_family_stats::*f) {
utils::timed_rate_moving_average_and_histogram column_family::stats::*f) {
utils::UUID uuid = get_uuid(name, ctx.db.local());
return ctx.db.map_reduce0([f, uuid](const database& p) {
return (p.find_column_family(uuid).get_stats().*f).rate();},
@@ -157,7 +145,7 @@ static future<json::json_return_type> get_cf_rate_and_histogram(http_context& c
});
}
static future<json::json_return_type> get_cf_rate_and_histogram(http_context& ctx, utils::timed_rate_moving_average_and_histogram column_family_stats::*f) {
static future<json::json_return_type> get_cf_rate_and_histogram(http_context& ctx, utils::timed_rate_moving_average_and_histogram column_family::stats::*f) {
std::function<utils::rate_moving_average_and_histogram(const database&)> fun = [f] (const database& db) {
utils::rate_moving_average_and_histogram res;
for (auto i : db.get_column_families()) {
@@ -178,27 +166,27 @@ static future<json::json_return_type> get_cf_unleveled_sstables(http_context& ct
}, std::plus<int64_t>());
}
static int64_t min_partition_size(column_family& cf) {
static int64_t min_row_size(column_family& cf) {
int64_t res = INT64_MAX;
for (auto i: *cf.get_sstables() ) {
res = std::min(res, i->get_stats_metadata().estimated_partition_size.min());
res = std::min(res, i->get_stats_metadata().estimated_row_size.min());
}
return (res == INT64_MAX) ? 0 : res;
}
static int64_t max_partition_size(column_family& cf) {
static int64_t max_row_size(column_family& cf) {
int64_t res = 0;
for (auto i: *cf.get_sstables() ) {
res = std::max(i->get_stats_metadata().estimated_partition_size.max(), res);
res = std::max(i->get_stats_metadata().estimated_row_size.max(), res);
}
return res;
}
static integral_ratio_holder mean_partition_size(column_family& cf) {
static integral_ratio_holder mean_row_size(column_family& cf) {
integral_ratio_holder res;
for (auto i: *cf.get_sstables() ) {
auto c = i->get_stats_metadata().estimated_partition_size.count();
res.sub += i->get_stats_metadata().estimated_partition_size.mean() * c;
auto c = i->get_stats_metadata().estimated_row_size.count();
res.sub += i->get_stats_metadata().estimated_row_size.mean() * c;
res.total += c;
}
return res;
@@ -249,22 +237,17 @@ static future<json::json_return_type> sum_sstable(http_context& ctx, bool total)
});
}
future<json::json_return_type> map_reduce_cf_time_histogram(http_context& ctx, const sstring& name, std::function<utils::time_estimated_histogram(const column_family&)> f) {
return map_reduce_cf_raw(ctx, name, utils::time_estimated_histogram(), f, utils::time_estimated_histogram_merge).then([](const utils::time_estimated_histogram& res) {
return make_ready_future<json::json_return_type>(time_to_json_histogram(res));
});
}
template <typename T>
class sum_ratio {
uint64_t _n = 0;
T _total = 0;
public:
void operator()(T value) {
future<> operator()(T value) {
if (value > 0) {
_total += value;
_n++;
}
return make_ready_future<>();
}
// Returns average value of all registered ratios.
T get() && {
@@ -413,31 +396,29 @@ void set_column_family(http_context& ctx, routes& r) {
});
cf::get_memtable_switch_count.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_stats(ctx,req->param["name"] ,&column_family_stats::memtable_switch_count);
return get_cf_stats(ctx,req->param["name"] ,&column_family::stats::memtable_switch_count);
});
cf::get_all_memtable_switch_count.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_stats(ctx, &column_family_stats::memtable_switch_count);
return get_cf_stats(ctx, &column_family::stats::memtable_switch_count);
});
// FIXME: this refers to partitions, not rows.
cf::get_estimated_row_size_histogram.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf(ctx, req->param["name"], utils::estimated_histogram(0), [](column_family& cf) {
utils::estimated_histogram res(0);
for (auto i: *cf.get_sstables() ) {
res.merge(i->get_stats_metadata().estimated_partition_size);
res.merge(i->get_stats_metadata().estimated_row_size);
}
return res;
},
utils::estimated_histogram_merge, utils_json::estimated_histogram());
});
// FIXME: this refers to partitions, not rows.
cf::get_estimated_row_count.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf(ctx, req->param["name"], int64_t(0), [](column_family& cf) {
uint64_t res = 0;
for (auto i: *cf.get_sstables() ) {
res += i->get_stats_metadata().estimated_partition_size.count();
res += i->get_stats_metadata().estimated_row_size.count();
}
return res;
},
@@ -462,67 +443,67 @@ void set_column_family(http_context& ctx, routes& r) {
});
cf::get_pending_flushes.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_stats(ctx,req->param["name"] ,&column_family_stats::pending_flushes);
return get_cf_stats(ctx,req->param["name"] ,&column_family::stats::pending_flushes);
});
cf::get_all_pending_flushes.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_stats(ctx, &column_family_stats::pending_flushes);
return get_cf_stats(ctx, &column_family::stats::pending_flushes);
});
cf::get_read.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_stats_count(ctx,req->param["name"] ,&column_family_stats::reads);
return get_cf_stats_count(ctx,req->param["name"] ,&column_family::stats::reads);
});
cf::get_all_read.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_stats_count(ctx, &column_family_stats::reads);
return get_cf_stats_count(ctx, &column_family::stats::reads);
});
cf::get_write.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_stats_count(ctx, req->param["name"] ,&column_family_stats::writes);
return get_cf_stats_count(ctx, req->param["name"] ,&column_family::stats::writes);
});
cf::get_all_write.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_stats_count(ctx, &column_family_stats::writes);
return get_cf_stats_count(ctx, &column_family::stats::writes);
});
cf::get_read_latency_histogram_depricated.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_histogram(ctx, req->param["name"], &column_family_stats::reads);
return get_cf_histogram(ctx, req->param["name"], &column_family::stats::reads);
});
cf::get_read_latency_histogram.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_rate_and_histogram(ctx, req->param["name"], &column_family_stats::reads);
return get_cf_rate_and_histogram(ctx, req->param["name"], &column_family::stats::reads);
});
cf::get_read_latency.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_stats_sum(ctx,req->param["name"] ,&column_family_stats::reads);
return get_cf_stats_sum(ctx,req->param["name"] ,&column_family::stats::reads);
});
cf::get_write_latency.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_stats_sum(ctx, req->param["name"] ,&column_family_stats::writes);
return get_cf_stats_sum(ctx, req->param["name"] ,&column_family::stats::writes);
});
cf::get_all_read_latency_histogram_depricated.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_histogram(ctx, &column_family_stats::writes);
return get_cf_histogram(ctx, &column_family::stats::writes);
});
cf::get_all_read_latency_histogram.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_rate_and_histogram(ctx, &column_family_stats::writes);
return get_cf_rate_and_histogram(ctx, &column_family::stats::writes);
});
cf::get_write_latency_histogram_depricated.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_histogram(ctx, req->param["name"], &column_family_stats::writes);
return get_cf_histogram(ctx, req->param["name"], &column_family::stats::writes);
});
cf::get_write_latency_histogram.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_rate_and_histogram(ctx, req->param["name"], &column_family_stats::writes);
return get_cf_rate_and_histogram(ctx, req->param["name"], &column_family::stats::writes);
});
cf::get_all_write_latency_histogram_depricated.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_histogram(ctx, &column_family_stats::writes);
return get_cf_histogram(ctx, &column_family::stats::writes);
});
cf::get_all_write_latency_histogram.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_rate_and_histogram(ctx, &column_family_stats::writes);
return get_cf_rate_and_histogram(ctx, &column_family::stats::writes);
});
cf::get_pending_compactions.set(r, [&ctx] (std::unique_ptr<request> req) {
@@ -538,11 +519,11 @@ void set_column_family(http_context& ctx, routes& r) {
});
cf::get_live_ss_table_count.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_stats(ctx, req->param["name"], &column_family_stats::live_sstable_count);
return get_cf_stats(ctx, req->param["name"], &column_family::stats::live_sstable_count);
});
cf::get_all_live_ss_table_count.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_stats(ctx, &column_family_stats::live_sstable_count);
return get_cf_stats(ctx, &column_family::stats::live_sstable_count);
});
cf::get_unleveled_sstables.set(r, [&ctx] (std::unique_ptr<request> req) {
@@ -565,36 +546,30 @@ void set_column_family(http_context& ctx, routes& r) {
return sum_sstable(ctx, true);
});
// FIXME: this refers to partitions, not rows.
cf::get_min_row_size.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf(ctx, req->param["name"], INT64_MAX, min_partition_size, min_int64);
return map_reduce_cf(ctx, req->param["name"], INT64_MAX, min_row_size, min_int64);
});
// FIXME: this refers to partitions, not rows.
cf::get_all_min_row_size.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf(ctx, INT64_MAX, min_partition_size, min_int64);
return map_reduce_cf(ctx, INT64_MAX, min_row_size, min_int64);
});
// FIXME: this refers to partitions, not rows.
cf::get_max_row_size.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf(ctx, req->param["name"], int64_t(0), max_partition_size, max_int64);
return map_reduce_cf(ctx, req->param["name"], int64_t(0), max_row_size, max_int64);
});
// FIXME: this refers to partitions, not rows.
cf::get_all_max_row_size.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf(ctx, int64_t(0), max_partition_size, max_int64);
return map_reduce_cf(ctx, int64_t(0), max_row_size, max_int64);
});
// FIXME: this refers to partitions, not rows.
cf::get_mean_row_size.set(r, [&ctx] (std::unique_ptr<request> req) {
// Cassandra 3.x mean values are truncated as integrals.
return map_reduce_cf(ctx, req->param["name"], integral_ratio_holder(), mean_partition_size, std::plus<integral_ratio_holder>());
return map_reduce_cf(ctx, req->param["name"], integral_ratio_holder(), mean_row_size, std::plus<integral_ratio_holder>());
});
// FIXME: this refers to partitions, not rows.
cf::get_all_mean_row_size.set(r, [&ctx] (std::unique_ptr<request> req) {
// Cassandra 3.x mean values are truncated as integrals.
return map_reduce_cf(ctx, integral_ratio_holder(), mean_partition_size, std::plus<integral_ratio_holder>());
return map_reduce_cf(ctx, integral_ratio_holder(), mean_row_size, std::plus<integral_ratio_holder>());
});
cf::get_bloom_filter_false_positives.set(r, [&ctx] (std::unique_ptr<request> req) {
@@ -801,22 +776,25 @@ void set_column_family(http_context& ctx, routes& r) {
});
cf::get_cas_prepare.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf_time_histogram(ctx, req->param["name"], [](const column_family& cf) {
return cf.get_stats().estimated_cas_prepare;
});
cf::get_cas_prepare.set(r, [] (std::unique_ptr<request> req) {
//TBD
unimplemented();
//auto id = get_uuid(req->param["name"], ctx.db.local());
return make_ready_future<json::json_return_type>(0);
});
cf::get_cas_propose.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf_time_histogram(ctx, req->param["name"], [](const column_family& cf) {
return cf.get_stats().estimated_cas_accept;
});
cf::get_cas_propose.set(r, [] (std::unique_ptr<request> req) {
//TBD
unimplemented();
//auto id = get_uuid(req->param["name"], ctx.db.local());
return make_ready_future<json::json_return_type>(0);
});
cf::get_cas_commit.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf_time_histogram(ctx, req->param["name"], [](const column_family& cf) {
return cf.get_stats().estimated_cas_learn;
});
cf::get_cas_commit.set(r, [] (std::unique_ptr<request> req) {
//TBD
unimplemented();
//auto id = get_uuid(req->param["name"], ctx.db.local());
return make_ready_future<json::json_return_type>(0);
});
cf::get_sstables_per_read_histogram.set(r, [&ctx] (std::unique_ptr<request> req) {
@@ -827,11 +805,11 @@ void set_column_family(http_context& ctx, routes& r) {
});
cf::get_tombstone_scanned_histogram.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_histogram(ctx, req->param["name"], &column_family_stats::tombstone_scanned);
return get_cf_histogram(ctx, req->param["name"], &column_family::stats::tombstone_scanned);
});
cf::get_live_scanned_histogram.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cf_histogram(ctx, req->param["name"], &column_family_stats::live_scanned);
return get_cf_histogram(ctx, req->param["name"], &column_family::stats::live_scanned);
});
cf::get_col_update_time_delta_histogram.set(r, [] (std::unique_ptr<request> req) {
@@ -842,51 +820,19 @@ void set_column_family(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(res);
});
cf::get_auto_compaction.set(r, [&ctx] (const_req req) {
const utils::UUID& uuid = get_uuid(req.param["name"], ctx.db.local());
column_family& cf = ctx.db.local().find_column_family(uuid);
return !cf.is_auto_compaction_disabled_by_user();
cf::is_auto_compaction_disabled.set(r, [] (const_req req) {
// FIXME
// currently auto compaction is disable
// it should be changed when it would have an API
return true;
});
cf::enable_auto_compaction.set(r, [&ctx](std::unique_ptr<request> req) {
return foreach_column_family(ctx, req->param["name"], [](column_family &cf) {
cf.enable_auto_compaction();
}).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
cf::get_built_indexes.set(r, [](const_req) {
// FIXME
// Currently there are no index support
return std::vector<sstring>();
});
cf::disable_auto_compaction.set(r, [&ctx](std::unique_ptr<request> req) {
return foreach_column_family(ctx, req->param["name"], [](column_family &cf) {
cf.disable_auto_compaction();
}).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
cf::get_built_indexes.set(r, [&ctx](std::unique_ptr<request> req) {
auto ks_cf = parse_fully_qualified_cf_name(req->param["name"]);
auto&& ks = std::get<0>(ks_cf);
auto&& cf_name = std::get<1>(ks_cf);
return db::system_keyspace::load_view_build_progress().then([ks, cf_name, &ctx](const std::vector<db::system_keyspace::view_build_progress>& vb) mutable {
std::set<sstring> vp;
for (auto b : vb) {
if (b.view.first == ks) {
vp.insert(b.view.second);
}
}
std::vector<sstring> res;
auto uuid = get_uuid(ks, cf_name, ctx.db.local());
column_family& cf = ctx.db.local().find_column_family(uuid);
res.reserve(cf.get_index_manager().list_indexes().size());
for (auto&& i : cf.get_index_manager().list_indexes()) {
if (!vp.contains(secondary_index::index_table_name(i.metadata().name()))) {
res.emplace_back(i.metadata().name());
}
}
return make_ready_future<json::json_return_type>(res);
});
});
cf::get_compression_metadata_off_heap_memory_used.set(r, [](const_req) {
// FIXME
@@ -914,15 +860,17 @@ void set_column_family(http_context& ctx, routes& r) {
});
cf::get_read_latency_estimated_histogram.set(r, [&ctx](std::unique_ptr<request> req) {
return map_reduce_cf_time_histogram(ctx, req->param["name"], [](const column_family& cf) {
return map_reduce_cf(ctx, req->param["name"], utils::estimated_histogram(0), [](column_family& cf) {
return cf.get_stats().estimated_read;
});
},
utils::estimated_histogram_merge, utils_json::estimated_histogram());
});
cf::get_write_latency_estimated_histogram.set(r, [&ctx](std::unique_ptr<request> req) {
return map_reduce_cf_time_histogram(ctx, req->param["name"], [](const column_family& cf) {
return map_reduce_cf(ctx, req->param["name"], utils::estimated_histogram(0), [](column_family& cf) {
return cf.get_stats().estimated_write;
});
},
utils::estimated_histogram_merge, utils_json::estimated_histogram());
});
cf::set_compaction_strategy_class.set(r, [&ctx](std::unique_ptr<request> req) {
@@ -972,55 +920,5 @@ void set_column_family(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(container_to_vec(res));
});
});
cf::toppartitions.set(r, [&ctx] (std::unique_ptr<request> req) {
auto name_param = req->param["name"];
auto [ks, cf] = parse_fully_qualified_cf_name(name_param);
api::req_param<std::chrono::milliseconds, unsigned> duration{*req, "duration", 1000ms};
api::req_param<unsigned> capacity(*req, "capacity", 256);
api::req_param<unsigned> list_size(*req, "list_size", 10);
apilog.info("toppartitions query: name={} duration={} list_size={} capacity={}",
name_param, duration.param, list_size.param, capacity.param);
return seastar::do_with(db::toppartitions_query(ctx.db, ks, cf, duration.value, list_size, capacity), [&ctx](auto& q) {
return q.scatter().then([&q] {
return sleep(q.duration()).then([&q] {
return q.gather(q.capacity()).then([&q] (auto topk_results) {
apilog.debug("toppartitions query: processing results");
cf::toppartitions_query_results results;
for (auto& d: topk_results.read.top(q.list_size())) {
cf::toppartitions_record r;
r.partition = sstring(d.item);
r.count = d.count;
r.error = d.error;
results.read.push(r);
}
for (auto& d: topk_results.write.top(q.list_size())) {
cf::toppartitions_record r;
r.partition = sstring(d.item);
r.count = d.count;
r.error = d.error;
results.write.push(r);
}
return make_ready_future<json::json_return_type>(results);
});
});
});
});
});
cf::force_major_compaction.set(r, [&ctx](std::unique_ptr<request> req) {
if (req->get_query_param("split_output") != "") {
fail(unimplemented::cause::API);
}
return foreach_column_family(ctx, req->param["name"], [](column_family &cf) {
return cf.compact_all_sstables();
}).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
}
}

View File

@@ -24,7 +24,6 @@
#include "api.hh"
#include "api/api-doc/column_family.json.hh"
#include "database.hh"
#include <seastar/core/future-util.hh>
#include <any>
namespace api {
@@ -39,14 +38,14 @@ template<class Mapper, class I, class Reducer>
future<I> map_reduce_cf_raw(http_context& ctx, const sstring& name, I init,
Mapper mapper, Reducer reducer) {
auto uuid = get_uuid(name, ctx.db.local());
using mapper_type = std::function<std::unique_ptr<std::any>(database&)>;
using reducer_type = std::function<std::unique_ptr<std::any>(std::unique_ptr<std::any>, std::unique_ptr<std::any>)>;
using mapper_type = std::function<std::any (database&)>;
using reducer_type = std::function<std::any (std::any, std::any)>;
return ctx.db.map_reduce0(mapper_type([mapper, uuid](database& db) {
return std::make_unique<std::any>(I(mapper(db.find_column_family(uuid))));
}), std::make_unique<std::any>(std::move(init)), reducer_type([reducer = std::move(reducer)] (std::unique_ptr<std::any> a, std::unique_ptr<std::any> b) mutable {
return std::make_unique<std::any>(I(reducer(std::any_cast<I>(std::move(*a)), std::any_cast<I>(std::move(*b)))));
})).then([] (std::unique_ptr<std::any> r) {
return std::any_cast<I>(std::move(*r));
return I(mapper(db.find_column_family(uuid)));
}), std::any(std::move(init)), reducer_type([reducer = std::move(reducer)] (std::any a, std::any b) mutable {
return I(reducer(std::any_cast<I>(std::move(a)), std::any_cast<I>(std::move(b))));
})).then([] (std::any r) {
return std::any_cast<I>(std::move(r));
});
}
@@ -68,36 +67,32 @@ future<json::json_return_type> map_reduce_cf(http_context& ctx, const sstring& n
});
}
future<json::json_return_type> map_reduce_cf_time_histogram(http_context& ctx, const sstring& name, std::function<utils::time_estimated_histogram(const column_family&)> f);
struct map_reduce_column_families_locally {
std::any init;
std::function<std::unique_ptr<std::any>(column_family&)> mapper;
std::function<std::unique_ptr<std::any>(std::unique_ptr<std::any>, std::unique_ptr<std::any>)> reducer;
future<std::unique_ptr<std::any>> operator()(database& db) const {
auto res = seastar::make_lw_shared<std::unique_ptr<std::any>>(std::make_unique<std::any>(init));
return do_for_each(db.get_column_families(), [res, this](const std::pair<utils::UUID, seastar::lw_shared_ptr<table>>& i) {
*res = std::move(reducer(std::move(*res), mapper(*i.second.get())));
}).then([res] {
return std::move(*res);
});
std::function<std::any (column_family&)> mapper;
std::function<std::any (std::any, std::any)> reducer;
std::any operator()(database& db) const {
auto res = init;
for (auto i : db.get_column_families()) {
res = reducer(res, mapper(*i.second.get()));
}
return res;
}
};
template<class Mapper, class I, class Reducer>
future<I> map_reduce_cf_raw(http_context& ctx, I init,
Mapper mapper, Reducer reducer) {
using mapper_type = std::function<std::unique_ptr<std::any>(column_family&)>;
using reducer_type = std::function<std::unique_ptr<std::any>(std::unique_ptr<std::any>, std::unique_ptr<std::any>)>;
using mapper_type = std::function<std::any (column_family&)>;
using reducer_type = std::function<std::any (std::any, std::any)>;
auto wrapped_mapper = mapper_type([mapper = std::move(mapper)] (column_family& cf) mutable {
return std::make_unique<std::any>(I(mapper(cf)));
return I(mapper(cf));
});
auto wrapped_reducer = reducer_type([reducer = std::move(reducer)] (std::unique_ptr<std::any> a, std::unique_ptr<std::any> b) mutable {
return std::make_unique<std::any>(I(reducer(std::any_cast<I>(std::move(*a)), std::any_cast<I>(std::move(*b)))));
auto wrapped_reducer = reducer_type([reducer = std::move(reducer)] (std::any a, std::any b) mutable {
return I(reducer(std::any_cast<I>(std::move(a)), std::any_cast<I>(std::move(b))));
});
return ctx.db.map_reduce0(map_reduce_column_families_locally{init,
std::move(wrapped_mapper), wrapped_reducer}, std::make_unique<std::any>(init), wrapped_reducer).then([] (std::unique_ptr<std::any> res) {
return std::any_cast<I>(std::move(*res));
return ctx.db.map_reduce0(map_reduce_column_families_locally{init, std::move(wrapped_mapper), wrapped_reducer}, std::any(init), wrapped_reducer).then([] (std::any res) {
return std::any_cast<I>(std::move(res));
});
}
@@ -111,9 +106,9 @@ future<json::json_return_type> map_reduce_cf(http_context& ctx, I init,
}
future<json::json_return_type> get_cf_stats(http_context& ctx, const sstring& name,
int64_t column_family_stats::*f);
int64_t column_family::stats::*f);
future<json::json_return_type> get_cf_stats(http_context& ctx,
int64_t column_family_stats::*f);
int64_t column_family::stats::*f);
}

View File

@@ -20,18 +20,17 @@
*/
#include "commitlog.hh"
#include "db/commitlog/commitlog.hh"
#include <db/commitlog/commitlog.hh>
#include "api/api-doc/commitlog.json.hh"
#include "database.hh"
#include <vector>
namespace api {
template<typename T>
static auto acquire_cl_metric(http_context& ctx, std::function<T (db::commitlog*)> func) {
typedef T ret_type;
template<typename Func>
static auto acquire_cl_metric(http_context& ctx, Func&& func) {
typedef std::result_of_t<Func(db::commitlog *)> ret_type;
return ctx.db.map_reduce0([func = std::move(func)](database& db) {
return ctx.db.map_reduce0([func = std::forward<Func>(func)](database& db) {
if (db.commitlog() == nullptr) {
return make_ready_future<ret_type>();
}
@@ -64,15 +63,15 @@ void set_commitlog(http_context& ctx, routes& r) {
});
httpd::commitlog_json::get_completed_tasks.set(r, [&ctx](std::unique_ptr<request> req) {
return acquire_cl_metric<uint64_t>(ctx, std::bind(&db::commitlog::get_completed_tasks, std::placeholders::_1));
return acquire_cl_metric(ctx, std::bind(&db::commitlog::get_completed_tasks, std::placeholders::_1));
});
httpd::commitlog_json::get_pending_tasks.set(r, [&ctx](std::unique_ptr<request> req) {
return acquire_cl_metric<uint64_t>(ctx, std::bind(&db::commitlog::get_pending_tasks, std::placeholders::_1));
return acquire_cl_metric(ctx, std::bind(&db::commitlog::get_pending_tasks, std::placeholders::_1));
});
httpd::commitlog_json::get_total_commit_log_size.set(r, [&ctx](std::unique_ptr<request> req) {
return acquire_cl_metric<uint64_t>(ctx, std::bind(&db::commitlog::get_total_size, std::placeholders::_1));
return acquire_cl_metric(ctx, std::bind(&db::commitlog::get_total_size, std::placeholders::_1));
});
}

View File

@@ -24,7 +24,6 @@
#include "api/api-doc/compaction_manager.json.hh"
#include "db/system_keyspace.hh"
#include "column_family.hh"
#include <utility>
namespace api {
@@ -39,16 +38,6 @@ static future<json::json_return_type> get_cm_stats(http_context& ctx,
return make_ready_future<json::json_return_type>(res);
});
}
static std::unordered_map<std::pair<sstring, sstring>, uint64_t, utils::tuple_hash> sum_pending_tasks(std::unordered_map<std::pair<sstring, sstring>, uint64_t, utils::tuple_hash>&& a,
const std::unordered_map<std::pair<sstring, sstring>, uint64_t, utils::tuple_hash>& b) {
for (auto&& i : b) {
if (i.second) {
a[i.first] += i.second;
}
}
return std::move(a);
}
void set_compaction_manager(http_context& ctx, routes& r) {
cm::get_compactions.set(r, [&ctx] (std::unique_ptr<request> req) {
@@ -58,8 +47,8 @@ void set_compaction_manager(http_context& ctx, routes& r) {
for (const auto& c : cm.get_compactions()) {
cm::summary s;
s.ks = c->ks_name;
s.cf = c->cf_name;
s.ks = c->ks;
s.cf = c->cf;
s.unit = "keys";
s.task_type = sstables::compaction_name(c->type);
s.completed = c->total_keys_written;
@@ -72,32 +61,6 @@ void set_compaction_manager(http_context& ctx, routes& r) {
});
});
cm::get_pending_tasks_by_table.set(r, [&ctx] (std::unique_ptr<request> req) {
return ctx.db.map_reduce0([&ctx](database& db) {
return do_with(std::unordered_map<std::pair<sstring, sstring>, uint64_t, utils::tuple_hash>(), [&ctx, &db](std::unordered_map<std::pair<sstring, sstring>, uint64_t, utils::tuple_hash>& tasks) {
return do_for_each(db.get_column_families(), [&tasks](const std::pair<utils::UUID, seastar::lw_shared_ptr<table>>& i) {
table& cf = *i.second.get();
tasks[std::make_pair(cf.schema()->ks_name(), cf.schema()->cf_name())] = cf.get_compaction_strategy().estimated_pending_compactions(cf);
return make_ready_future<>();
}).then([&tasks] {
return std::move(tasks);
});
});
}, std::unordered_map<std::pair<sstring, sstring>, uint64_t, utils::tuple_hash>(), sum_pending_tasks).then(
[](const std::unordered_map<std::pair<sstring, sstring>, uint64_t, utils::tuple_hash>& task_map) {
std::vector<cm::pending_compaction> res;
res.reserve(task_map.size());
for (auto i : task_map) {
cm::pending_compaction task;
task.ks = i.first.first;
task.cf = i.first.second;
task.task = i.second;
res.emplace_back(std::move(task));
}
return make_ready_future<json::json_return_type>(res);
});
});
cm::force_user_defined_compaction.set(r, [] (std::unique_ptr<request> req) {
//TBD
// FIXME
@@ -140,37 +103,29 @@ void set_compaction_manager(http_context& ctx, routes& r) {
});
cm::get_compaction_history.set(r, [] (std::unique_ptr<request> req) {
std::function<future<>(output_stream<char>&&)> f = [](output_stream<char>&& s) {
return do_with(output_stream<char>(std::move(s)), true, [] (output_stream<char>& s, bool& first){
return s.write("[").then([&s, &first] {
return db::system_keyspace::get_compaction_history([&s, &first](const db::system_keyspace::compaction_history_entry& entry) mutable {
cm::history h;
h.id = entry.id.to_sstring();
h.ks = std::move(entry.ks);
h.cf = std::move(entry.cf);
h.compacted_at = entry.compacted_at;
h.bytes_in = entry.bytes_in;
h.bytes_out = entry.bytes_out;
for (auto it : entry.rows_merged) {
httpd::compaction_manager_json::row_merged e;
e.key = it.first;
e.value = it.second;
h.rows_merged.push(std::move(e));
}
auto fut = first ? make_ready_future<>() : s.write(", ");
first = false;
return fut.then([&s, h = std::move(h)] {
return formatter::write(s, h);
});
}).then([&s] {
return s.write("]").then([&s] {
return s.close();
});
});
});
});
};
return make_ready_future<json::json_return_type>(std::move(f));
return db::system_keyspace::get_compaction_history().then([] (std::vector<db::system_keyspace::compaction_history_entry> history) {
std::vector<cm::history> res;
res.reserve(history.size());
for (auto& entry : history) {
cm::history h;
h.id = entry.id.to_sstring();
h.ks = std::move(entry.ks);
h.cf = std::move(entry.cf);
h.compacted_at = entry.compacted_at;
h.bytes_in = entry.bytes_in;
h.bytes_out = entry.bytes_out;
for (auto it : entry.rows_merged) {
httpd::compaction_manager_json::row_merged e;
e.key = it.first;
e.value = it.second;
h.rows_merged.push(std::move(e));
}
res.push_back(std::move(h));
}
return make_ready_future<json::json_return_type>(res);
});
});
cm::get_compaction_info.set(r, [] (std::unique_ptr<request> req) {

View File

@@ -22,7 +22,6 @@
#include "api/config.hh"
#include "api/api-doc/config.json.hh"
#include "db/config.hh"
#include "database.hh"
#include <sstream>
#include <boost/algorithm/string/replace.hpp>
@@ -44,14 +43,14 @@ json::json_return_type get_json_return_type(const db::seed_provider_type& val) {
return json::json_return_type(val.class_name);
}
std::string_view format_type(std::string_view type) {
std::string format_type(const std::string& type) {
if (type == "int") {
return "integer";
}
return type;
}
future<> get_config_swagger_entry(std::string_view name, const std::string& description, std::string_view type, bool& first, output_stream<char>& os) {
future<> get_config_swagger_entry(const std::string& name, const std::string& description, const std::string& type, bool& first, output_stream<char>& os) {
std::stringstream ss;
if (first) {
first=false;
@@ -88,29 +87,23 @@ future<> get_config_swagger_entry(std::string_view name, const std::string& desc
}
namespace cs = httpd::config_json;
#define _get_config_value(name, type, deflt, status, desc, ...) if (id == #name) {return get_json_return_type(ctx.db.local().get_config().name());}
#define _get_config_description(name, type, deflt, status, desc, ...) f = f.then([&os, &first] {return get_config_swagger_entry(#name, desc, #type, first, os);});
void set_config(std::shared_ptr < api_registry_builder20 > rb, http_context& ctx, routes& r) {
rb->register_function(r, [&ctx] (output_stream<char>& os) {
return do_with(true, [&os, &ctx] (bool& first) {
rb->register_function(r, [] (output_stream<char>& os) {
return do_with(true, [&os] (bool& first) {
auto f = make_ready_future();
for (auto&& cfg_ref : ctx.db.local().get_config().values()) {
auto&& cfg = cfg_ref.get();
f = f.then([&os, &first, &cfg] {
return get_config_swagger_entry(cfg.name(), std::string(cfg.desc()), cfg.type_name(), first, os);
});
}
_make_config_values(_get_config_description)
return f;
});
});
cs::find_config_id.set(r, [&ctx] (const_req r) {
auto id = r.param["id"];
for (auto&& cfg_ref : ctx.db.local().get_config().values()) {
auto&& cfg = cfg_ref.get();
if (id == cfg.name()) {
return cfg.value_as_json();
}
}
_make_config_values(_get_config_value)
throw bad_param_exception(sstring("No such config entry: ") + id);
});
}

View File

@@ -1,69 +0,0 @@
/*
* Copyright (C) 2020 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#include "api/api-doc/error_injection.json.hh"
#include "api/api.hh"
#include <seastar/http/exception.hh>
#include "log.hh"
#include "utils/error_injection.hh"
#include "seastar/core/future-util.hh"
namespace api {
namespace hf = httpd::error_injection_json;
void set_error_injection(http_context& ctx, routes& r) {
hf::enable_injection.set(r, [](std::unique_ptr<request> req) {
sstring injection = req->param["injection"];
bool one_shot = req->get_query_param("one_shot") == "True";
auto& errinj = utils::get_local_injector();
return errinj.enable_on_all(injection, one_shot).then([] {
return make_ready_future<json::json_return_type>(json::json_void());
});
});
hf::get_enabled_injections_on_all.set(r, [](std::unique_ptr<request> req) {
auto& errinj = utils::get_local_injector();
auto ret = errinj.enabled_injections_on_all();
return make_ready_future<json::json_return_type>(ret);
});
hf::disable_injection.set(r, [](std::unique_ptr<request> req) {
sstring injection = req->param["injection"];
auto& errinj = utils::get_local_injector();
return errinj.disable_on_all(injection).then([] {
return make_ready_future<json::json_return_type>(json::json_void());
});
});
hf::disable_on_all.set(r, [](std::unique_ptr<request> req) {
auto& errinj = utils::get_local_injector();
return errinj.disable_on_all().then([] {
return make_ready_future<json::json_return_type>(json::json_void());
});
});
}
} // namespace api

View File

@@ -1,30 +0,0 @@
/*
* Copyright (C) 2019 ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#pragma once
#include "api.hh"
namespace api {
void set_error_injection(http_context& ctx, routes& r);
}

View File

@@ -21,7 +21,7 @@
#include "gossiper.hh"
#include "api/api-doc/gossiper.json.hh"
#include "gms/gossiper.hh"
#include <gms/gossiper.hh>
namespace api {
using namespace json;

View File

@@ -23,7 +23,7 @@
#include "api/lsa.hh"
#include "api/api.hh"
#include <seastar/http/exception.hh>
#include "http/exception.hh"
#include "utils/logalloc.hh"
#include "log.hh"

View File

@@ -21,7 +21,7 @@
#include "messaging_service.hh"
#include "message/messaging_service.hh"
#include <seastar/rpc/rpc_types.hh>
#include "rpc/rpc_types.hh"
#include "api/api-doc/messaging_service.json.hh"
#include <iostream>
#include <sstream>
@@ -53,8 +53,8 @@ std::vector<message_counter> map_to_message_counters(
* according to a function that it gets as a parameter.
*
*/
future_json_function get_client_getter(sharded<netw::messaging_service>& ms, std::function<uint64_t(const shard_info&)> f) {
return [&ms, f](std::unique_ptr<request> req) {
future_json_function get_client_getter(std::function<uint64_t(const shard_info&)> f) {
return [f](std::unique_ptr<request> req) {
using map_type = std::unordered_map<gms::inet_address, uint64_t>;
auto get_shard_map = [f](messaging_service& ms) {
std::unordered_map<gms::inet_address, unsigned long> map;
@@ -63,70 +63,70 @@ future_json_function get_client_getter(sharded<netw::messaging_service>& ms, std
});
return map;
};
return ms.map_reduce0(get_shard_map, map_type(), map_sum<map_type>).
return get_messaging_service().map_reduce0(get_shard_map, map_type(), map_sum<map_type>).
then([](map_type&& map) {
return make_ready_future<json::json_return_type>(map_to_message_counters(map));
});
};
}
future_json_function get_server_getter(sharded<netw::messaging_service>& ms, std::function<uint64_t(const rpc::stats&)> f) {
return [&ms, f](std::unique_ptr<request> req) {
future_json_function get_server_getter(std::function<uint64_t(const rpc::stats&)> f) {
return [f](std::unique_ptr<request> req) {
using map_type = std::unordered_map<gms::inet_address, uint64_t>;
auto get_shard_map = [f](messaging_service& ms) {
std::unordered_map<gms::inet_address, unsigned long> map;
ms.foreach_server_connection_stats([&map, f] (const rpc::client_info& info, const rpc::stats& stats) mutable {
map[gms::inet_address(info.addr.addr())] = f(stats);
map[gms::inet_address(net::ipv4_address(info.addr))] = f(stats);
});
return map;
};
return ms.map_reduce0(get_shard_map, map_type(), map_sum<map_type>).
return get_messaging_service().map_reduce0(get_shard_map, map_type(), map_sum<map_type>).
then([](map_type&& map) {
return make_ready_future<json::json_return_type>(map_to_message_counters(map));
});
};
}
void set_messaging_service(http_context& ctx, routes& r, sharded<netw::messaging_service>& ms) {
get_timeout_messages.set(r, get_client_getter(ms, [](const shard_info& c) {
void set_messaging_service(http_context& ctx, routes& r) {
get_timeout_messages.set(r, get_client_getter([](const shard_info& c) {
return c.get_stats().timeout;
}));
get_sent_messages.set(r, get_client_getter(ms, [](const shard_info& c) {
get_sent_messages.set(r, get_client_getter([](const shard_info& c) {
return c.get_stats().sent_messages;
}));
get_dropped_messages.set(r, get_client_getter(ms, [](const shard_info& c) {
get_dropped_messages.set(r, get_client_getter([](const shard_info& c) {
// We don't have the same drop message mechanism
// as origin has.
// hence we can always return 0
return 0;
}));
get_exception_messages.set(r, get_client_getter(ms, [](const shard_info& c) {
get_exception_messages.set(r, get_client_getter([](const shard_info& c) {
return c.get_stats().exception_received;
}));
get_pending_messages.set(r, get_client_getter(ms, [](const shard_info& c) {
get_pending_messages.set(r, get_client_getter([](const shard_info& c) {
return c.get_stats().pending;
}));
get_respond_pending_messages.set(r, get_server_getter(ms, [](const rpc::stats& c) {
get_respond_pending_messages.set(r, get_server_getter([](const rpc::stats& c) {
return c.pending;
}));
get_respond_completed_messages.set(r, get_server_getter(ms, [](const rpc::stats& c) {
get_respond_completed_messages.set(r, get_server_getter([](const rpc::stats& c) {
return c.sent_messages;
}));
get_version.set(r, [&ms](const_req req) {
return ms.local().get_raw_version(req.get_query_param("addr"));
get_version.set(r, [](const_req req) {
return netw::get_local_messaging_service().get_raw_version(req.get_query_param("addr"));
});
get_dropped_messages_by_ver.set(r, [&ms](std::unique_ptr<request> req) {
get_dropped_messages_by_ver.set(r, [](std::unique_ptr<request> req) {
shared_ptr<std::vector<uint64_t>> map = make_shared<std::vector<uint64_t>>(num_verb);
return ms.map_reduce([map](const uint64_t* local_map) mutable {
return netw::get_messaging_service().map_reduce([map](const uint64_t* local_map) mutable {
for (auto i = 0; i < num_verb; i++) {
(*map)[i]+= local_map[i];
}
@@ -139,7 +139,7 @@ void set_messaging_service(http_context& ctx, routes& r, sharded<netw::messaging
messaging_verb v = i; // for type safety we use messaging_verb values
auto idx = static_cast<uint32_t>(v);
if (idx >= map->size()) {
throw std::runtime_error(format("verb index out of bounds: {:d}, map size: {:d}", idx, map->size()));
throw std::runtime_error(sprint("verb index out of bounds: %lu, map size: %lu", idx, map->size()));
}
if ((*map)[idx] > 0) {
c.count = (*map)[idx];
@@ -151,18 +151,5 @@ void set_messaging_service(http_context& ctx, routes& r, sharded<netw::messaging
});
});
}
void unset_messaging_service(http_context& ctx, routes& r) {
get_timeout_messages.unset(r);
get_sent_messages.unset(r);
get_dropped_messages.unset(r);
get_exception_messages.unset(r);
get_pending_messages.unset(r);
get_respond_pending_messages.unset(r);
get_respond_completed_messages.unset(r);
get_version.unset(r);
get_dropped_messages_by_ver.unset(r);
}
}

View File

@@ -23,11 +23,8 @@
#include "api.hh"
namespace netw { class messaging_service; }
namespace api {
void set_messaging_service(http_context& ctx, routes& r, sharded<netw::messaging_service>& ms);
void unset_messaging_service(http_context& ctx, routes& r);
void set_messaging_service(http_context& ctx, routes& r);
}

View File

@@ -26,8 +26,6 @@
#include "service/storage_service.hh"
#include "db/config.hh"
#include "utils/histogram.hh"
#include "database.hh"
#include "seastar/core/scheduling_specific.hh"
namespace api {
@@ -35,70 +33,12 @@ namespace sp = httpd::storage_proxy_json;
using proxy = service::storage_proxy;
using namespace json;
/**
* This function implement a two dimentional map reduce where
* the first level is a distributed storage_proxy class and the
* second level is the stats per scheduling group class.
* @param d - a reference to the storage_proxy distributed class.
* @param mapper - the internal mapper that is used to map the internal
* stat class into a value of type `V`.
* @param reducer - the reducer that is used in both outer and inner
* aggregations.
* @param initial_value - the initial value to use for both aggregations
* @return A future that resolves to the result of the aggregation.
*/
template<typename V, typename Reducer, typename InnerMapper>
future<V> two_dimensional_map_reduce(distributed<service::storage_proxy>& d,
InnerMapper mapper, Reducer reducer, V initial_value) {
return d.map_reduce0( [mapper, reducer, initial_value] (const service::storage_proxy& sp) {
return map_reduce_scheduling_group_specific<service::storage_proxy_stats::stats>(
mapper, reducer, initial_value, sp.get_stats_key());
}, initial_value, reducer);
static future<utils::rate_moving_average> sum_timed_rate(distributed<proxy>& d, utils::timed_rate_moving_average proxy::stats::*f) {
return d.map_reduce0([f](const proxy& p) {return (p.get_stats().*f).rate();}, utils::rate_moving_average(),
std::plus<utils::rate_moving_average>());
}
/**
* This function implement a two dimentional map reduce where
* the first level is a distributed storage_proxy class and the
* second level is the stats per scheduling group class.
* @param d - a reference to the storage_proxy distributed class.
* @param f - a field pointer which is the implicit internal reducer.
* @param reducer - the reducer that is used in both outer and inner
* aggregations.
* @param initial_value - the initial value to use for both aggregations* @return
* @return A future that resolves to the result of the aggregation.
*/
template<typename V, typename Reducer, typename F>
future<V> two_dimensional_map_reduce(distributed<service::storage_proxy>& d,
V F::*f, Reducer reducer, V initial_value) {
return two_dimensional_map_reduce(d, [f] (F& stats) {
return stats.*f;
}, reducer, initial_value);
}
/**
* A partial Specialization of sum_stats for the storage proxy
* case where the get stats function doesn't return a
* stats object with fields but a per scheduling group
* stats object, the name was also changed since functions
* partial specialization is not supported in C++.
*
*/
template<typename V, typename F>
future<json::json_return_type> sum_stats_storage_proxy(distributed<proxy>& d, V F::*f) {
return two_dimensional_map_reduce(d, [f] (F& stats) { return stats.*f; }, std::plus<V>(), V(0)).then([] (V val) {
return make_ready_future<json::json_return_type>(val);
});
}
static future<utils::rate_moving_average> sum_timed_rate(distributed<proxy>& d, utils::timed_rate_moving_average service::storage_proxy_stats::stats::*f) {
return two_dimensional_map_reduce(d, [f] (service::storage_proxy_stats::stats& stats) {
return (stats.*f).rate();
}, std::plus<utils::rate_moving_average>(), utils::rate_moving_average());
}
static future<json::json_return_type> sum_timed_rate_as_obj(distributed<proxy>& d, utils::timed_rate_moving_average service::storage_proxy_stats::stats::*f) {
static future<json::json_return_type> sum_timed_rate_as_obj(distributed<proxy>& d, utils::timed_rate_moving_average proxy::stats::*f) {
return sum_timed_rate(d, f).then([](const utils::rate_moving_average& val) {
httpd::utils_json::rate_moving_average m;
m = val;
@@ -106,93 +46,29 @@ static future<json::json_return_type> sum_timed_rate_as_obj(distributed<proxy>&
});
}
httpd::utils_json::rate_moving_average_and_histogram get_empty_moving_average() {
return timer_to_json(utils::rate_moving_average_and_histogram());
}
static future<json::json_return_type> sum_timed_rate_as_long(distributed<proxy>& d, utils::timed_rate_moving_average service::storage_proxy_stats::stats::*f) {
static future<json::json_return_type> sum_timed_rate_as_long(distributed<proxy>& d, utils::timed_rate_moving_average proxy::stats::*f) {
return sum_timed_rate(d, f).then([](const utils::rate_moving_average& val) {
return make_ready_future<json::json_return_type>(val.count);
});
}
utils_json::estimated_histogram time_to_json_histogram(const utils::time_estimated_histogram& val) {
utils_json::estimated_histogram res;
for (size_t i = 0; i < val.size(); i++) {
res.buckets.push(val.get(i));
res.bucket_offsets.push(val.get_bucket_lower_limit(i));
}
return res;
}
static future<json::json_return_type> sum_estimated_histogram(http_context& ctx, utils::time_estimated_histogram service::storage_proxy_stats::stats::*f) {
return two_dimensional_map_reduce(ctx.sp, f, utils::time_estimated_histogram_merge,
utils::time_estimated_histogram()).then([](const utils::time_estimated_histogram& val) {
return make_ready_future<json::json_return_type>(time_to_json_histogram(val));
});
}
static future<json::json_return_type> sum_estimated_histogram(http_context& ctx, utils::estimated_histogram service::storage_proxy_stats::stats::*f) {
return two_dimensional_map_reduce(ctx.sp, f, utils::estimated_histogram_merge,
utils::estimated_histogram()).then([](const utils::estimated_histogram& val) {
static future<json::json_return_type> sum_estimated_histogram(http_context& ctx, utils::estimated_histogram proxy::stats::*f) {
return ctx.sp.map_reduce0([f](const proxy& p) {return p.get_stats().*f;}, utils::estimated_histogram(),
utils::estimated_histogram_merge).then([](const utils::estimated_histogram& val) {
utils_json::estimated_histogram res;
res = val;
return make_ready_future<json::json_return_type>(res);
});
}
static future<json::json_return_type> total_latency(http_context& ctx, utils::timed_rate_moving_average_and_histogram service::storage_proxy_stats::stats::*f) {
return two_dimensional_map_reduce(ctx.sp, [f] (service::storage_proxy_stats::stats& stats) {
return (stats.*f).hist.mean * (stats.*f).hist.count;
}, std::plus<double>(), 0.0).then([](double val) {
static future<json::json_return_type> total_latency(http_context& ctx, utils::timed_rate_moving_average_and_histogram proxy::stats::*f) {
return ctx.sp.map_reduce0([f](const proxy& p) {return (p.get_stats().*f).hist.mean * (p.get_stats().*f).hist.count;}, 0.0,
std::plus<double>()).then([](double val) {
int64_t res = val;
return make_ready_future<json::json_return_type>(res);
});
}
/**
* A partial Specialization of sum_histogram_stats
* for the storage proxy case where the get stats
* function doesn't return a stats object with
* fields but a per scheduling group stats object,
* the name was also changed since function partial
* specialization is not supported in C++.
*/
template<typename F>
future<json::json_return_type>
sum_histogram_stats_storage_proxy(distributed<proxy>& d,
utils::timed_rate_moving_average_and_histogram F::*f) {
return two_dimensional_map_reduce(d, [f] (service::storage_proxy_stats::stats& stats) {
return (stats.*f).hist;
}, std::plus<utils::ihistogram>(), utils::ihistogram()).
then([](const utils::ihistogram& val) {
return make_ready_future<json::json_return_type>(to_json(val));
});
}
/**
* A partial Specialization of sum_timer_stats for the
* storage proxy case where the get stats function
* doesn't return a stats object with fields but a
* per scheduling group stats object, the name
* was also changed since partial function specialization
* is not supported in C++.
*/
template<typename F>
future<json::json_return_type>
sum_timer_stats_storage_proxy(distributed<proxy>& d,
utils::timed_rate_moving_average_and_histogram F::*f) {
return two_dimensional_map_reduce(d, [f] (service::storage_proxy_stats::stats& stats) {
return (stats.*f).rate();
}, std::plus<utils::rate_moving_average_and_histogram>(),
utils::rate_moving_average_and_histogram()).then([](const utils::rate_moving_average_and_histogram& val) {
return make_ready_future<json::json_return_type>(timer_to_json(val));
});
}
void set_storage_proxy(http_context& ctx, routes& r) {
sp::get_total_hints.set(r, [](std::unique_ptr<request> req) {
//TBD
@@ -200,40 +76,33 @@ void set_storage_proxy(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(0);
});
sp::get_hinted_handoff_enabled.set(r, [&ctx](std::unique_ptr<request> req) {
const auto& filter = service::get_storage_proxy().local().get_hints_host_filter();
return make_ready_future<json::json_return_type>(!filter.is_disabled_for_all());
sp::get_hinted_handoff_enabled.set(r, [](std::unique_ptr<request> req) {
//TBD
// FIXME
// hinted handoff is not supported currently,
// so we should return false
return make_ready_future<json::json_return_type>(false);
});
sp::set_hinted_handoff_enabled.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto enable = req->get_query_param("enable");
auto filter = (enable == "true" || enable == "1")
? db::hints::host_filter(db::hints::host_filter::enabled_for_all_tag {})
: db::hints::host_filter(db::hints::host_filter::disabled_for_all_tag {});
return service::get_storage_proxy().invoke_on_all([filter = std::move(filter)] (service::storage_proxy& sp) {
return sp.change_hints_host_filter(filter);
}).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
return make_ready_future<json::json_return_type>(json_void());
});
sp::get_hinted_handoff_enabled_by_dc.set(r, [](std::unique_ptr<request> req) {
std::vector<sstring> res;
const auto& filter = service::get_storage_proxy().local().get_hints_host_filter();
const auto& dcs = filter.get_dcs();
res.reserve(res.size());
std::copy(dcs.begin(), dcs.end(), std::back_inserter(res));
//TBD
unimplemented();
std::vector<sp::mapper_list> res;
return make_ready_future<json::json_return_type>(res);
});
sp::set_hinted_handoff_enabled_by_dc_list.set(r, [](std::unique_ptr<request> req) {
auto dcs = req->get_query_param("dcs");
auto filter = db::hints::host_filter::parse_from_dc_list(std::move(dcs));
return service::get_storage_proxy().invoke_on_all([filter = std::move(filter)] (service::storage_proxy& sp) {
return sp.change_hints_host_filter(filter);
}).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
//TBD
unimplemented();
auto enable = req->get_query_param("dcs");
return make_ready_future<json::json_return_type>(json_void());
});
sp::get_max_hint_window.set(r, [](std::unique_ptr<request> req) {
@@ -352,15 +221,15 @@ void set_storage_proxy(http_context& ctx, routes& r) {
});
sp::get_read_repair_attempted.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_stats_storage_proxy(ctx.sp, &service::storage_proxy_stats::stats::read_repair_attempts);
return sum_stats(ctx.sp, &proxy::stats::read_repair_attempts);
});
sp::get_read_repair_repaired_blocking.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_stats_storage_proxy(ctx.sp, &service::storage_proxy_stats::stats::read_repair_repaired_blocking);
return sum_stats(ctx.sp, &proxy::stats::read_repair_repaired_blocking);
});
sp::get_read_repair_repaired_background.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_stats_storage_proxy(ctx.sp, &service::storage_proxy_stats::stats::read_repair_repaired_background);
return sum_stats(ctx.sp, &proxy::stats::read_repair_repaired_background);
});
sp::get_schema_versions.set(r, [](std::unique_ptr<request> req) {
@@ -376,154 +245,163 @@ void set_storage_proxy(http_context& ctx, routes& r) {
});
});
sp::get_cas_read_timeouts.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_long(ctx.sp, &proxy::stats::cas_read_timeouts);
sp::get_cas_read_timeouts.set(r, [](std::unique_ptr<request> req) {
//TBD
// FIXME
// cas is not supported yet, so just return 0
return make_ready_future<json::json_return_type>(0);
});
sp::get_cas_read_unavailables.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_long(ctx.sp, &proxy::stats::cas_read_unavailables);
sp::get_cas_read_unavailables.set(r, [](std::unique_ptr<request> req) {
//TBD
// FIXME
// cas is not supported yet, so just return 0
return make_ready_future<json::json_return_type>(0);
});
sp::get_cas_write_timeouts.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_long(ctx.sp, &proxy::stats::cas_write_timeouts);
sp::get_cas_write_timeouts.set(r, [](std::unique_ptr<request> req) {
//TBD
// FIXME
// cas is not supported yet, so just return 0
return make_ready_future<json::json_return_type>(0);
});
sp::get_cas_write_unavailables.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_long(ctx.sp, &proxy::stats::cas_write_unavailables);
sp::get_cas_write_unavailables.set(r, [](std::unique_ptr<request> req) {
//TBD
// FIXME
// cas is not supported yet, so just return 0
return make_ready_future<json::json_return_type>(0);
});
sp::get_cas_write_metrics_unfinished_commit.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_stats(ctx.sp, &proxy::stats::cas_write_unfinished_commit);
sp::get_cas_write_metrics_unfinished_commit.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
return make_ready_future<json::json_return_type>(0);
});
sp::get_cas_write_metrics_contention.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_estimated_histogram(ctx, &proxy::stats::cas_write_contention);
sp::get_cas_write_metrics_contention.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
return make_ready_future<json::json_return_type>(0);
});
sp::get_cas_write_metrics_condition_not_met.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_stats(ctx.sp, &proxy::stats::cas_write_condition_not_met);
sp::get_cas_write_metrics_condition_not_met.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
return make_ready_future<json::json_return_type>(0);
});
sp::get_cas_write_metrics_failed_read_round_optimization.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_stats(ctx.sp, &proxy::stats::cas_failed_read_round_optimization);
sp::get_cas_read_metrics_unfinished_commit.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
return make_ready_future<json::json_return_type>(0);
});
sp::get_cas_read_metrics_unfinished_commit.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_stats(ctx.sp, &proxy::stats::cas_read_unfinished_commit);
sp::get_cas_read_metrics_contention.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
return make_ready_future<json::json_return_type>(0);
});
sp::get_cas_read_metrics_contention.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_estimated_histogram(ctx, &proxy::stats::cas_read_contention);
sp::get_cas_read_metrics_condition_not_met.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
return make_ready_future<json::json_return_type>(0);
});
sp::get_read_metrics_timeouts.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_long(ctx.sp, &service::storage_proxy_stats::stats::read_timeouts);
return sum_timed_rate_as_long(ctx.sp, &proxy::stats::read_timeouts);
});
sp::get_read_metrics_unavailables.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_long(ctx.sp, &service::storage_proxy_stats::stats::read_unavailables);
return sum_timed_rate_as_long(ctx.sp, &proxy::stats::read_unavailables);
});
sp::get_range_metrics_timeouts.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_long(ctx.sp, &service::storage_proxy_stats::stats::range_slice_timeouts);
return sum_timed_rate_as_long(ctx.sp, &proxy::stats::range_slice_timeouts);
});
sp::get_range_metrics_unavailables.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_long(ctx.sp, &service::storage_proxy_stats::stats::range_slice_unavailables);
return sum_timed_rate_as_long(ctx.sp, &proxy::stats::range_slice_unavailables);
});
sp::get_write_metrics_timeouts.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_long(ctx.sp, &service::storage_proxy_stats::stats::write_timeouts);
return sum_timed_rate_as_long(ctx.sp, &proxy::stats::write_timeouts);
});
sp::get_write_metrics_unavailables.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_long(ctx.sp, &service::storage_proxy_stats::stats::write_unavailables);
return sum_timed_rate_as_long(ctx.sp, &proxy::stats::write_unavailables);
});
sp::get_read_metrics_timeouts_rates.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_obj(ctx.sp, &service::storage_proxy_stats::stats::read_timeouts);
return sum_timed_rate_as_obj(ctx.sp, &proxy::stats::read_timeouts);
});
sp::get_read_metrics_unavailables_rates.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_obj(ctx.sp, &service::storage_proxy_stats::stats::read_unavailables);
return sum_timed_rate_as_obj(ctx.sp, &proxy::stats::read_unavailables);
});
sp::get_range_metrics_timeouts_rates.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_obj(ctx.sp, &service::storage_proxy_stats::stats::range_slice_timeouts);
return sum_timed_rate_as_obj(ctx.sp, &proxy::stats::range_slice_timeouts);
});
sp::get_range_metrics_unavailables_rates.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_obj(ctx.sp, &service::storage_proxy_stats::stats::range_slice_unavailables);
return sum_timed_rate_as_obj(ctx.sp, &proxy::stats::range_slice_unavailables);
});
sp::get_write_metrics_timeouts_rates.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_obj(ctx.sp, &service::storage_proxy_stats::stats::write_timeouts);
return sum_timed_rate_as_obj(ctx.sp, &proxy::stats::write_timeouts);
});
sp::get_write_metrics_unavailables_rates.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timed_rate_as_obj(ctx.sp, &service::storage_proxy_stats::stats::write_unavailables);
return sum_timed_rate_as_obj(ctx.sp, &proxy::stats::write_unavailables);
});
sp::get_range_metrics_latency_histogram_depricated.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_histogram_stats_storage_proxy(ctx.sp, &service::storage_proxy_stats::stats::range);
return sum_histogram_stats(ctx.sp, &proxy::stats::range);
});
sp::get_write_metrics_latency_histogram_depricated.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_histogram_stats_storage_proxy(ctx.sp, &service::storage_proxy_stats::stats::write);
return sum_histogram_stats(ctx.sp, &proxy::stats::write);
});
sp::get_read_metrics_latency_histogram_depricated.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_histogram_stats_storage_proxy(ctx.sp, &service::storage_proxy_stats::stats::read);
return sum_histogram_stats(ctx.sp, &proxy::stats::read);
});
sp::get_range_metrics_latency_histogram.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timer_stats_storage_proxy(ctx.sp, &service::storage_proxy_stats::stats::range);
return sum_timer_stats(ctx.sp, &proxy::stats::range);
});
sp::get_write_metrics_latency_histogram.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timer_stats_storage_proxy(ctx.sp, &service::storage_proxy_stats::stats::write);
});
sp::get_cas_write_metrics_latency_histogram.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timer_stats(ctx.sp, &proxy::stats::cas_write);
});
sp::get_cas_read_metrics_latency_histogram.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timer_stats(ctx.sp, &proxy::stats::cas_read);
});
sp::get_view_write_metrics_latency_histogram.set(r, [&ctx](std::unique_ptr<request> req) {
//TBD
// FIXME
// No View metrics are available, so just return empty moving average
return make_ready_future<json::json_return_type>(get_empty_moving_average());
return sum_timer_stats(ctx.sp, &proxy::stats::write);
});
sp::get_read_metrics_latency_histogram.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timer_stats_storage_proxy(ctx.sp, &service::storage_proxy_stats::stats::read);
return sum_timer_stats(ctx.sp, &proxy::stats::read);
});
sp::get_read_estimated_histogram.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_estimated_histogram(ctx, &service::storage_proxy_stats::stats::estimated_read);
return sum_estimated_histogram(ctx, &proxy::stats::estimated_read);
});
sp::get_read_latency.set(r, [&ctx](std::unique_ptr<request> req) {
return total_latency(ctx, &service::storage_proxy_stats::stats::read);
return total_latency(ctx, &proxy::stats::read);
});
sp::get_write_estimated_histogram.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_estimated_histogram(ctx, &service::storage_proxy_stats::stats::estimated_write);
return sum_estimated_histogram(ctx, &proxy::stats::estimated_write);
});
sp::get_write_latency.set(r, [&ctx](std::unique_ptr<request> req) {
return total_latency(ctx, &service::storage_proxy_stats::stats::write);
return total_latency(ctx, &proxy::stats::write);
});
sp::get_range_estimated_histogram.set(r, [&ctx](std::unique_ptr<request> req) {
return sum_timer_stats_storage_proxy(ctx.sp, &service::storage_proxy_stats::stats::range);
return sum_timer_stats(ctx.sp, &proxy::stats::range);
});
sp::get_range_latency.set(r, [&ctx](std::unique_ptr<request> req) {
return total_latency(ctx, &service::storage_proxy_stats::stats::range);
return total_latency(ctx, &proxy::stats::range);
});
}

View File

@@ -22,37 +22,22 @@
#include "storage_service.hh"
#include "api/api-doc/storage_service.json.hh"
#include "db/config.hh"
#include "db/schema_tables.hh"
#include <optional>
#include <time.h>
#include <boost/range/adaptor/map.hpp>
#include <boost/range/adaptor/filtered.hpp>
#include "service/storage_service.hh"
#include "service/load_meter.hh"
#include "db/commitlog/commitlog.hh"
#include "gms/gossiper.hh"
#include "db/system_keyspace.hh"
#include "seastar/http/exception.hh"
#include <service/storage_service.hh>
#include <db/commitlog/commitlog.hh>
#include <gms/gossiper.hh>
#include <db/system_keyspace.hh>
#include "http/exception.hh"
#include "repair/repair.hh"
#include "locator/snitch_base.hh"
#include "column_family.hh"
#include "log.hh"
#include "release.hh"
#include "sstables/compaction_manager.hh"
#include "sstables/sstables.hh"
#include "database.hh"
#include "db/extensions.hh"
#include "db/snapshot-ctl.hh"
#include "transport/controller.hh"
#include "thrift/controller.hh"
#include "locator/token_metadata.hh"
namespace api {
const locator::token_metadata& http_context::get_token_metadata() {
return *shared_token_metadata.local().get();
}
namespace ss = httpd::storage_service_json;
using namespace json;
@@ -63,195 +48,27 @@ static sstring validate_keyspace(http_context& ctx, const parameters& param) {
throw bad_param_exception("Keyspace " + param["keyspace"] + " Does not exist");
}
static ss::token_range token_range_endpoints_to_json(const dht::token_range_endpoints& d) {
ss::token_range r;
r.start_token = d._start_token;
r.end_token = d._end_token;
r.endpoints = d._endpoints;
r.rpc_endpoints = d._rpc_endpoints;
for (auto det : d._endpoint_details) {
ss::endpoint_detail ed;
ed.host = det._host;
ed.datacenter = det._datacenter;
if (det._rack != "") {
ed.rack = det._rack;
static std::vector<ss::token_range> describe_ring(const sstring& keyspace) {
std::vector<ss::token_range> res;
for (auto d : service::get_local_storage_service().describe_ring(keyspace)) {
ss::token_range r;
r.start_token = d._start_token;
r.end_token = d._end_token;
r.endpoints = d._endpoints;
r.rpc_endpoints = d._rpc_endpoints;
for (auto det : d._endpoint_details) {
ss::endpoint_detail ed;
ed.host = det._host;
ed.datacenter = det._datacenter;
if (det._rack != "") {
ed.rack = det._rack;
}
r.endpoint_details.push(ed);
}
r.endpoint_details.push(ed);
res.push_back(r);
}
return r;
}
using ks_cf_func = std::function<future<json::json_return_type>(http_context&, std::unique_ptr<request>, sstring, std::vector<sstring>)>;
static auto wrap_ks_cf(http_context &ctx, ks_cf_func f) {
return [&ctx, f = std::move(f)](std::unique_ptr<request> req) {
auto keyspace = validate_keyspace(ctx, req->param);
auto column_families = split_cf(req->get_query_param("cf"));
if (column_families.empty()) {
column_families = map_keys(ctx.db.local().find_keyspace(keyspace).metadata().get()->cf_meta_data());
}
return f(ctx, std::move(req), std::move(keyspace), std::move(column_families));
};
}
future<json::json_return_type> set_tables_autocompaction(http_context& ctx, const sstring &keyspace, std::vector<sstring> tables, bool enabled) {
if (tables.empty()) {
tables = map_keys(ctx.db.local().find_keyspace(keyspace).metadata().get()->cf_meta_data());
}
return service::get_local_storage_service().set_tables_autocompaction(keyspace, tables, enabled).then([]{
return make_ready_future<json::json_return_type>(json_void());
});
}
void set_transport_controller(http_context& ctx, routes& r, cql_transport::controller& ctl) {
ss::start_native_transport.set(r, [&ctl](std::unique_ptr<request> req) {
return ctl.start_server().then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::stop_native_transport.set(r, [&ctl](std::unique_ptr<request> req) {
return ctl.stop_server().then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::is_native_transport_running.set(r, [&ctl] (std::unique_ptr<request> req) {
return ctl.is_server_running().then([] (bool running) {
return make_ready_future<json::json_return_type>(running);
});
});
}
void unset_transport_controller(http_context& ctx, routes& r) {
ss::start_native_transport.unset(r);
ss::stop_native_transport.unset(r);
ss::is_native_transport_running.unset(r);
}
void set_rpc_controller(http_context& ctx, routes& r, thrift_controller& ctl) {
ss::stop_rpc_server.set(r, [&ctl](std::unique_ptr<request> req) {
return ctl.stop_server().then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::start_rpc_server.set(r, [&ctl](std::unique_ptr<request> req) {
return ctl.start_server().then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::is_rpc_server_running.set(r, [&ctl] (std::unique_ptr<request> req) {
return ctl.is_server_running().then([] (bool running) {
return make_ready_future<json::json_return_type>(running);
});
});
}
void unset_rpc_controller(http_context& ctx, routes& r) {
ss::stop_rpc_server.unset(r);
ss::start_rpc_server.unset(r);
ss::is_rpc_server_running.unset(r);
}
void set_repair(http_context& ctx, routes& r, sharded<netw::messaging_service>& ms) {
ss::repair_async.set(r, [&ctx, &ms](std::unique_ptr<request> req) {
static std::vector<sstring> options = {"primaryRange", "parallelism", "incremental",
"jobThreads", "ranges", "columnFamilies", "dataCenters", "hosts", "trace",
"startToken", "endToken" };
std::unordered_map<sstring, sstring> options_map;
for (auto o : options) {
auto s = req->get_query_param(o);
if (s != "") {
options_map[o] = s;
}
}
// The repair process is asynchronous: repair_start only starts it and
// returns immediately, not waiting for the repair to finish. The user
// then has other mechanisms to track the ongoing repair's progress,
// or stop it.
return repair_start(ctx.db, ms, validate_keyspace(ctx, req->param),
options_map).then([] (int i) {
return make_ready_future<json::json_return_type>(i);
});
});
ss::get_active_repair_async.set(r, [&ctx](std::unique_ptr<request> req) {
return get_active_repairs(ctx.db).then([] (std::vector<int> res){
return make_ready_future<json::json_return_type>(res);
});
});
ss::repair_async_status.set(r, [&ctx](std::unique_ptr<request> req) {
return repair_get_status(ctx.db, boost::lexical_cast<int>( req->get_query_param("id")))
.then_wrapped([] (future<repair_status>&& fut) {
ss::ns_repair_async_status::return_type_wrapper res;
try {
res = fut.get0();
} catch(std::runtime_error& e) {
throw httpd::bad_param_exception(e.what());
}
return make_ready_future<json::json_return_type>(json::json_return_type(res));
});
});
ss::repair_await_completion.set(r, [&ctx](std::unique_ptr<request> req) {
int id;
using clock = std::chrono::steady_clock;
clock::time_point expire;
try {
id = boost::lexical_cast<int>(req->get_query_param("id"));
// If timeout is not provided, it means no timeout.
sstring s = req->get_query_param("timeout");
int64_t timeout = s.empty() ? int64_t(-1) : boost::lexical_cast<int64_t>(s);
if (timeout < 0 && timeout != -1) {
return make_exception_future<json::json_return_type>(
httpd::bad_param_exception("timeout can only be -1 (means no timeout) or non negative integer"));
}
if (timeout < 0) {
expire = clock::time_point::max();
} else {
expire = clock::now() + std::chrono::seconds(timeout);
}
} catch (std::exception& e) {
return make_exception_future<json::json_return_type>(httpd::bad_param_exception(e.what()));
}
return repair_await_completion(ctx.db, id, expire)
.then_wrapped([] (future<repair_status>&& fut) {
ss::ns_repair_async_status::return_type_wrapper res;
try {
res = fut.get0();
} catch (std::exception& e) {
return make_exception_future<json::json_return_type>(httpd::server_error_exception(e.what()));
}
return make_ready_future<json::json_return_type>(json::json_return_type(res));
});
});
ss::force_terminate_all_repair_sessions.set(r, [](std::unique_ptr<request> req) {
return repair_abort_all(service::get_local_storage_service().db()).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::force_terminate_all_repair_sessions_new.set(r, [](std::unique_ptr<request> req) {
return repair_abort_all(service::get_local_storage_service().db()).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
}
void unset_repair(http_context& ctx, routes& r) {
ss::repair_async.unset(r);
ss::get_active_repair_async.unset(r);
ss::repair_async_status.unset(r);
ss::repair_await_completion.unset(r);
ss::force_terminate_all_repair_sessions.unset(r);
ss::force_terminate_all_repair_sessions_new.unset(r);
return res;
}
void set_storage_service(http_context& ctx, routes& r) {
@@ -261,15 +78,15 @@ void set_storage_service(http_context& ctx, routes& r) {
});
});
ss::get_tokens.set(r, [&ctx] (std::unique_ptr<request> req) {
return make_ready_future<json::json_return_type>(stream_range_as_array(ctx.get_token_metadata().sorted_tokens(), [](const dht::token& i) {
ss::get_tokens.set(r, [] (std::unique_ptr<request> req) {
return make_ready_future<json::json_return_type>(stream_range_as_array(service::get_local_storage_service().get_token_metadata().sorted_tokens(), [](const dht::token& i) {
return boost::lexical_cast<std::string>(i);
}));
});
ss::get_node_tokens.set(r, [&ctx] (std::unique_ptr<request> req) {
ss::get_node_tokens.set(r, [] (std::unique_ptr<request> req) {
gms::inet_address addr(req->param["endpoint"]);
return make_ready_future<json::json_return_type>(stream_range_as_array(ctx.get_token_metadata().get_tokens(addr), [](const dht::token& i) {
return make_ready_future<json::json_return_type>(stream_range_as_array(service::get_local_storage_service().get_token_metadata().get_tokens(addr), [](const dht::token& i) {
return boost::lexical_cast<std::string>(i);
}));
});
@@ -287,8 +104,8 @@ void set_storage_service(http_context& ctx, routes& r) {
}));
});
ss::get_leaving_nodes.set(r, [&ctx](const_req req) {
return container_to_vec(ctx.get_token_metadata().get_leaving_endpoints());
ss::get_leaving_nodes.set(r, [](const_req req) {
return container_to_vec(service::get_local_storage_service().get_token_metadata().get_leaving_endpoints());
});
ss::get_moving_nodes.set(r, [](const_req req) {
@@ -296,8 +113,8 @@ void set_storage_service(http_context& ctx, routes& r) {
return container_to_vec(addr);
});
ss::get_joining_nodes.set(r, [&ctx](const_req req) {
auto points = ctx.get_token_metadata().get_bootstrap_tokens();
ss::get_joining_nodes.set(r, [](const_req req) {
auto points = service::get_local_storage_service().get_token_metadata().get_bootstrap_tokens();
std::unordered_set<sstring> addr;
for (auto i: points) {
addr.insert(boost::lexical_cast<std::string>(i.second));
@@ -325,26 +142,11 @@ void set_storage_service(http_context& ctx, routes& r) {
});
ss::get_range_to_endpoint_map.set(r, [&ctx](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto keyspace = validate_keyspace(ctx, req->param);
std::vector<ss::maplist_mapper> res;
return make_ready_future<json::json_return_type>(stream_range_as_array(service::get_local_storage_service().get_range_to_address_map(keyspace),
[](const std::pair<dht::token_range, std::vector<gms::inet_address>>& entry){
ss::maplist_mapper m;
if (entry.first.start()) {
m.key.push(entry.first.start().value().value().to_sstring());
} else {
m.key.push("");
}
if (entry.first.end()) {
m.key.push(entry.first.end().value().value().to_sstring());
} else {
m.key.push("");
}
for (const gms::inet_address& address : entry.second) {
m.value.push(address.to_sstring());
}
return m;
}));
return make_ready_future<json::json_return_type>(res);
});
ss::get_pending_range_to_endpoint_map.set(r, [&ctx](std::unique_ptr<request> req) {
@@ -355,26 +157,27 @@ void set_storage_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(res);
});
ss::describe_any_ring.set(r, [&ctx](std::unique_ptr<request> req) {
return make_ready_future<json::json_return_type>(stream_range_as_array(service::get_local_storage_service().describe_ring(""), token_range_endpoints_to_json));
ss::describe_any_ring.set(r, [&ctx](const_req req) {
return describe_ring("");
});
ss::describe_ring.set(r, [&ctx](std::unique_ptr<request> req) {
auto keyspace = validate_keyspace(ctx, req->param);
return make_ready_future<json::json_return_type>(stream_range_as_array(service::get_local_storage_service().describe_ring(keyspace), token_range_endpoints_to_json));
ss::describe_ring.set(r, [&ctx](const_req req) {
auto keyspace = validate_keyspace(ctx, req.param);
return describe_ring(keyspace);
});
ss::get_host_id_map.set(r, [&ctx](const_req req) {
ss::get_host_id_map.set(r, [](const_req req) {
std::vector<ss::mapper> res;
return map_to_key_value(ctx.get_token_metadata().get_endpoint_to_host_id_map_for_reading(), res);
return map_to_key_value(service::get_local_storage_service().
get_token_metadata().get_endpoint_to_host_id_map_for_reading(), res);
});
ss::get_load.set(r, [&ctx](std::unique_ptr<request> req) {
return get_cf_stats(ctx, &column_family_stats::live_disk_space_used);
return get_cf_stats(ctx, &column_family::stats::live_disk_space_used);
});
ss::get_load_map.set(r, [&ctx] (std::unique_ptr<request> req) {
return ctx.lmeter.get_load_map().then([] (auto&& load_map) {
ss::get_load_map.set(r, [] (std::unique_ptr<request> req) {
return service::get_local_storage_service().get_load_map().then([] (auto&& load_map) {
std::vector<ss::map_string_double> res;
for (auto i : load_map) {
ss::map_string_double val;
@@ -399,12 +202,64 @@ void set_storage_service(http_context& ctx, routes& r) {
req.get_query_param("key")));
});
ss::cdc_streams_check_and_repair.set(r, [&ctx] (std::unique_ptr<request> req) {
return service::get_local_storage_service().check_and_repair_cdc_streams().then([] {
ss::get_snapshot_details.set(r, [](std::unique_ptr<request> req) {
return service::get_local_storage_service().get_snapshot_details().then([] (auto result) {
std::vector<ss::snapshots> res;
for (auto& map: result) {
ss::snapshots all_snapshots;
all_snapshots.key = map.first;
std::vector<ss::snapshot> snapshot;
for (auto& cf: map.second) {
ss::snapshot s;
s.ks = cf.ks;
s.cf = cf.cf;
s.live = cf.live;
s.total = cf.total;
snapshot.push_back(std::move(s));
}
all_snapshots.value = std::move(snapshot);
res.push_back(std::move(all_snapshots));
}
return make_ready_future<json::json_return_type>(std::move(res));
});
});
ss::take_snapshot.set(r, [](std::unique_ptr<request> req) {
auto tag = req->get_query_param("tag");
auto column_family = req->get_query_param("cf");
std::vector<sstring> keynames = split(req->get_query_param("kn"), ",");
auto resp = make_ready_future<>();
if (column_family.empty()) {
resp = service::get_local_storage_service().take_snapshot(tag, keynames);
} else {
if (keynames.size() > 1) {
throw httpd::bad_param_exception("Only one keyspace allowed when specifying a column family");
}
resp = service::get_local_storage_service().take_column_family_snapshot(keynames[0], column_family, tag);
}
return resp.then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::del_snapshot.set(r, [](std::unique_ptr<request> req) {
auto tag = req->get_query_param("tag");
std::vector<sstring> keynames = split(req->get_query_param("kn"), ",");
return service::get_local_storage_service().clear_snapshot(tag, keynames).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::true_snapshots_size.set(r, [](std::unique_ptr<request> req) {
return service::get_local_storage_service().true_snapshots_size().then([] (int64_t size) {
return make_ready_future<json::json_return_type>(size);
});
});
ss::force_keyspace_compaction.set(r, [&ctx](std::unique_ptr<request> req) {
auto keyspace = validate_keyspace(ctx, req->param);
auto column_families = split_cf(req->get_query_param("cf"));
@@ -430,40 +285,38 @@ void set_storage_service(http_context& ctx, routes& r) {
if (column_families.empty()) {
column_families = map_keys(ctx.db.local().find_keyspace(keyspace).metadata().get()->cf_meta_data());
}
return service::get_local_storage_service().is_cleanup_allowed(keyspace).then([&ctx, keyspace,
column_families = std::move(column_families)] (bool is_cleanup_allowed) mutable {
if (!is_cleanup_allowed) {
return make_exception_future<json::json_return_type>(
std::runtime_error("Can not perform cleanup operation when topology changes"));
return ctx.db.invoke_on_all([keyspace, column_families] (database& db) {
std::vector<column_family*> column_families_vec;
auto& cm = db.get_compaction_manager();
for (auto cf : column_families) {
column_families_vec.push_back(&db.find_column_family(keyspace, cf));
}
return ctx.db.invoke_on_all([keyspace, column_families] (database& db) {
std::vector<column_family*> column_families_vec;
auto& cm = db.get_compaction_manager();
for (auto cf : column_families) {
column_families_vec.push_back(&db.find_column_family(keyspace, cf));
}
return parallel_for_each(column_families_vec, [&cm, &db] (column_family* cf) {
return cm.perform_cleanup(db, cf);
});
}).then([]{
return make_ready_future<json::json_return_type>(0);
});
});
});
ss::upgrade_sstables.set(r, wrap_ks_cf(ctx, [] (http_context& ctx, std::unique_ptr<request> req, sstring keyspace, std::vector<sstring> column_families) {
bool exclude_current_version = req_param<bool>(*req, "exclude_current_version", false);
return ctx.db.invoke_on_all([=] (database& db) {
return do_for_each(column_families, [=, &db](sstring cfname) {
auto& cm = db.get_compaction_manager();
auto& cf = db.find_column_family(keyspace, cfname);
return cm.perform_sstable_upgrade(db, &cf, exclude_current_version);
return parallel_for_each(column_families_vec, [&cm] (column_family* cf) {
return cm.perform_cleanup(cf);
});
}).then([]{
return make_ready_future<json::json_return_type>(0);
});
}));
});
ss::scrub.set(r, [&ctx](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto keyspace = validate_keyspace(ctx, req->param);
auto column_family = req->get_query_param("cf");
auto disable_snapshot = req->get_query_param("disable_snapshot");
auto skip_corrupted = req->get_query_param("skip_corrupted");
return make_ready_future<json::json_return_type>(json_void());
});
ss::upgrade_sstables.set(r, [&ctx](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto keyspace = validate_keyspace(ctx, req->param);
auto column_family = req->get_query_param("cf");
auto exclude_current_version = req->get_query_param("exclude_current_version");
return make_ready_future<json::json_return_type>(json_void());
});
ss::force_keyspace_flush.set(r, [&ctx](std::unique_ptr<request> req) {
auto keyspace = validate_keyspace(ctx, req->param);
@@ -481,6 +334,59 @@ void set_storage_service(http_context& ctx, routes& r) {
});
ss::repair_async.set(r, [&ctx](std::unique_ptr<request> req) {
static std::vector<sstring> options = {"primaryRange", "parallelism", "incremental",
"jobThreads", "ranges", "columnFamilies", "dataCenters", "hosts", "trace",
"startToken", "endToken" };
std::unordered_map<sstring, sstring> options_map;
for (auto o : options) {
auto s = req->get_query_param(o);
if (s != "") {
options_map[o] = s;
}
}
// The repair process is asynchronous: repair_start only starts it and
// returns immediately, not waiting for the repair to finish. The user
// then has other mechanisms to track the ongoing repair's progress,
// or stop it.
return repair_start(ctx.db, validate_keyspace(ctx, req->param),
options_map).then([] (int i) {
return make_ready_future<json::json_return_type>(i);
});
});
ss::get_active_repair_async.set(r, [&ctx](std::unique_ptr<request> req) {
return get_active_repairs(ctx.db).then([] (std::vector<int> res){
return make_ready_future<json::json_return_type>(res);
});
});
ss::repair_async_status.set(r, [&ctx](std::unique_ptr<request> req) {
return repair_get_status(ctx.db, boost::lexical_cast<int>( req->get_query_param("id")))
.then_wrapped([] (future<repair_status>&& fut) {
ss::ns_repair_async_status::return_type_wrapper res;
try {
res = fut.get0();
} catch(std::runtime_error& e) {
throw httpd::bad_param_exception(e.what());
}
return make_ready_future<json::json_return_type>(json::json_return_type(res));
});
});
ss::force_terminate_all_repair_sessions.set(r, [](std::unique_ptr<request> req) {
return repair_abort_all(service::get_local_storage_service().db()).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::force_terminate_all_repair_sessions_new.set(r, [](std::unique_ptr<request> req) {
return repair_abort_all(service::get_local_storage_service().db()).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::decommission.set(r, [](std::unique_ptr<request> req) {
return service::get_local_storage_service().decommission().then([] {
return make_ready_future<json::json_return_type>(json_void());
@@ -548,7 +454,7 @@ void set_storage_service(http_context& ctx, routes& r) {
return service::get_storage_service().map_reduce(adder<service::storage_service::drain_progress>(), [] (auto& ss) {
return ss.get_drain_progress();
}).then([] (auto&& progress) {
auto progress_str = format("Drained {}/{} ColumnFamilies", progress.remaining_cfs, progress.total_cfs);
auto progress_str = sprint("Drained %s/%s ColumnFamilies", progress.remaining_cfs, progress.total_cfs);
return make_ready_future<json::json_return_type>(std::move(progress_str));
});
});
@@ -616,8 +522,46 @@ void set_storage_service(http_context& ctx, routes& r) {
});
});
ss::stop_rpc_server.set(r, [](std::unique_ptr<request> req) {
return service::get_local_storage_service().stop_rpc_server().then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::start_rpc_server.set(r, [](std::unique_ptr<request> req) {
return service::get_local_storage_service().start_rpc_server().then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::is_rpc_server_running.set(r, [] (std::unique_ptr<request> req) {
return service::get_local_storage_service().is_rpc_server_running().then([] (bool running) {
return make_ready_future<json::json_return_type>(running);
});
});
ss::start_native_transport.set(r, [](std::unique_ptr<request> req) {
return service::get_local_storage_service().start_native_transport().then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::stop_native_transport.set(r, [](std::unique_ptr<request> req) {
return service::get_local_storage_service().stop_native_transport().then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::is_native_transport_running.set(r, [] (std::unique_ptr<request> req) {
return service::get_local_storage_service().is_native_transport_running().then([] (bool running) {
return make_ready_future<json::json_return_type>(running);
});
});
ss::join_ring.set(r, [](std::unique_ptr<request> req) {
return make_ready_future<json::json_return_type>(json_void());
return service::get_local_storage_service().join_ring().then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::is_joined.set(r, [] (std::unique_ptr<request> req) {
@@ -721,11 +665,7 @@ void set_storage_service(http_context& ctx, routes& r) {
auto coordinator = std::hash<sstring>()(cf) % smp::count;
return service::get_storage_service().invoke_on(coordinator, [ks = std::move(ks), cf = std::move(cf)] (service::storage_service& s) {
return s.load_new_sstables(ks, cf);
}).then_wrapped([] (auto&& f) {
if (f.failed()) {
auto msg = fmt::format("Failed to load new sstables: {}", f.get_exception());
return make_exception_future<json::json_return_type>(httpd::server_error_exception(msg));
}
}).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
@@ -738,17 +678,14 @@ void set_storage_service(http_context& ctx, routes& r) {
});
ss::reset_local_schema.set(r, [](std::unique_ptr<request> req) {
// FIXME: We should truncate schema tables if more than one node in the cluster.
auto& sp = service::get_storage_proxy();
auto& fs = service::get_local_storage_service().features();
return db::schema_tables::recalculate_schema_version(sp, fs).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
//TBD
unimplemented();
return make_ready_future<json::json_return_type>(json_void());
});
ss::set_trace_probability.set(r, [](std::unique_ptr<request> req) {
auto probability = req->get_query_param("probability");
return futurize_invoke([probability] {
return futurize<json::json_return_type>::apply([probability] {
double real_prob = std::stod(probability.c_str());
return tracing::tracing::tracing_instance().invoke_on_all([real_prob] (auto& local_tracing) {
local_tracing.set_trace_probability(real_prob);
@@ -762,7 +699,7 @@ void set_storage_service(http_context& ctx, routes& r) {
} catch (std::out_of_range& e) {
throw httpd::bad_param_exception(e.what());
} catch (std::invalid_argument&){
throw httpd::bad_param_exception(format("Bad format in a probability value: \"{}\"", probability.c_str()));
throw httpd::bad_param_exception(sprint("Bad format in a probability value: \"%s\"", probability.c_str()));
}
});
});
@@ -798,22 +735,24 @@ void set_storage_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(json_void());
});
} catch (...) {
throw httpd::bad_param_exception(format("Bad format value: "));
throw httpd::bad_param_exception(sprint("Bad format value: "));
}
});
ss::enable_auto_compaction.set(r, [&ctx](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto keyspace = validate_keyspace(ctx, req->param);
auto tables = split_cf(req->get_query_param("cf"));
return set_tables_autocompaction(ctx, keyspace, tables, true);
auto column_family = req->get_query_param("cf");
return make_ready_future<json::json_return_type>(json_void());
});
ss::disable_auto_compaction.set(r, [&ctx](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto keyspace = validate_keyspace(ctx, req->param);
auto tables = split_cf(req->get_query_param("cf"));
return set_tables_autocompaction(ctx, keyspace, tables, false);
auto column_family = req->get_query_param("cf");
return make_ready_future<json::json_return_type>(json_void());
});
ss::deliver_hints.set(r, [](std::unique_ptr<request> req) {
@@ -878,7 +817,7 @@ void set_storage_service(http_context& ctx, routes& r) {
});
ss::get_metrics_load.set(r, [&ctx](std::unique_ptr<request> req) {
return get_cf_stats(ctx, &column_family_stats::live_disk_space_used);
return get_cf_stats(ctx, &column_family::stats::live_disk_space_used);
});
ss::get_exceptions.set(r, [](const_req req) {
@@ -920,243 +859,6 @@ void set_storage_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(map_to_key_value(std::move(status), res));
});
});
ss::sstable_info.set(r, [&ctx] (std::unique_ptr<request> req) {
auto ks = api::req_param<sstring>(*req, "keyspace", {}).value;
auto cf = api::req_param<sstring>(*req, "cf", {}).value;
// The size of this vector is bound by ks::cf. I.e. it is as most Nks + Ncf long
// which is not small, but not huge either.
using table_sstables_list = std::vector<ss::table_sstables>;
return do_with(table_sstables_list{}, [ks, cf, &ctx](table_sstables_list& dst) {
return service::get_local_storage_service().db().map_reduce([&dst](table_sstables_list&& res) {
for (auto&& t : res) {
auto i = std::find_if(dst.begin(), dst.end(), [&t](const ss::table_sstables& t2) {
return t.keyspace() == t2.keyspace() && t.table() == t2.table();
});
if (i == dst.end()) {
dst.emplace_back(std::move(t));
continue;
}
auto& ssd = i->sstables;
for (auto&& sd : t.sstables._elements) {
auto j = std::find_if(ssd._elements.begin(), ssd._elements.end(), [&sd](const ss::sstable& s) {
return s.generation() == sd.generation();
});
if (j == ssd._elements.end()) {
i->sstables.push(std::move(sd));
}
}
}
}, [ks, cf](const database& db) {
// see above
table_sstables_list res;
auto& ext = db.get_config().extensions();
for (auto& t : db.get_column_families() | boost::adaptors::map_values) {
auto& schema = t->schema();
if ((ks.empty() || ks == schema->ks_name()) && (cf.empty() || cf == schema->cf_name())) {
// at most Nsstables long
ss::table_sstables tst;
tst.keyspace = schema->ks_name();
tst.table = schema->cf_name();
for (auto sstable : *t->get_sstables_including_compacted_undeleted()) {
auto ts = db_clock::to_time_t(sstable->data_file_write_time());
::tm t;
::gmtime_r(&ts, &t);
ss::sstable info;
info.timestamp = t;
info.generation = sstable->generation();
info.level = sstable->get_sstable_level();
info.size = sstable->bytes_on_disk();
info.data_size = sstable->ondisk_data_size();
info.index_size = sstable->index_size();
info.filter_size = sstable->filter_size();
info.version = sstable->get_version();
if (sstable->has_component(sstables::component_type::CompressionInfo)) {
auto& c = sstable->get_compression();
auto cp = sstables::get_sstable_compressor(c);
ss::named_maps nm;
nm.group = "compression_parameters";
for (auto& p : cp->options()) {
ss::mapper e;
e.key = p.first;
e.value = p.second;
nm.attributes.push(std::move(e));
}
if (!cp->options().contains(compression_parameters::SSTABLE_COMPRESSION)) {
ss::mapper e;
e.key = compression_parameters::SSTABLE_COMPRESSION;
e.value = cp->name();
nm.attributes.push(std::move(e));
}
info.extended_properties.push(std::move(nm));
}
sstables::file_io_extension::attr_value_map map;
for (auto* ep : ext.sstable_file_io_extensions()) {
map.merge(ep->get_attributes(*sstable));
}
for (auto& p : map) {
struct {
const sstring& key;
ss::sstable& info;
void operator()(const std::map<sstring, sstring>& map) const {
ss::named_maps nm;
nm.group = key;
for (auto& p : map) {
ss::mapper e;
e.key = p.first;
e.value = p.second;
nm.attributes.push(std::move(e));
}
info.extended_properties.push(std::move(nm));
}
void operator()(const sstring& value) const {
ss::mapper e;
e.key = key;
e.value = value;
info.properties.push(std::move(e));
}
} v{p.first, info};
std::visit(v, p.second);
}
tst.sstables.push(std::move(info));
}
res.emplace_back(std::move(tst));
}
}
std::sort(res.begin(), res.end(), [](const ss::table_sstables& t1, const ss::table_sstables& t2) {
return t1.keyspace() < t2.keyspace() || (t1.keyspace() == t2.keyspace() && t1.table() < t2.table());
});
return res;
}).then([&dst] {
return make_ready_future<json::json_return_type>(stream_object(dst));
});
});
});
}
void set_snapshot(http_context& ctx, routes& r, sharded<db::snapshot_ctl>& snap_ctl) {
ss::get_snapshot_details.set(r, [&snap_ctl](std::unique_ptr<request> req) {
return snap_ctl.local().get_snapshot_details().then([] (std::unordered_map<sstring, std::vector<db::snapshot_ctl::snapshot_details>>&& result) {
std::function<future<>(output_stream<char>&&)> f = [result = std::move(result)](output_stream<char>&& s) {
return do_with(output_stream<char>(std::move(s)), true, [&result] (output_stream<char>& s, bool& first){
return s.write("[").then([&s, &first, &result] {
return do_for_each(result, [&s, &first](std::tuple<sstring, std::vector<db::snapshot_ctl::snapshot_details>>&& map){
return do_with(ss::snapshots(), [&s, &first, &map](ss::snapshots& all_snapshots) {
all_snapshots.key = std::get<0>(map);
future<> f = first ? make_ready_future<>() : s.write(", ");
first = false;
std::vector<ss::snapshot> snapshot;
for (auto& cf: std::get<1>(map)) {
ss::snapshot snp;
snp.ks = cf.ks;
snp.cf = cf.cf;
snp.live = cf.live;
snp.total = cf.total;
snapshot.push_back(std::move(snp));
}
all_snapshots.value = std::move(snapshot);
return f.then([&s, &all_snapshots] {
return all_snapshots.write(s);
});
});
});
}).then([&s] {
return s.write("]").then([&s] {
return s.close();
});
});
});
};
return make_ready_future<json::json_return_type>(std::move(f));
});
});
ss::take_snapshot.set(r, [&snap_ctl](std::unique_ptr<request> req) {
auto tag = req->get_query_param("tag");
auto column_families = split(req->get_query_param("cf"), ",");
std::vector<sstring> keynames = split(req->get_query_param("kn"), ",");
auto resp = make_ready_future<>();
if (column_families.empty()) {
resp = snap_ctl.local().take_snapshot(tag, keynames);
} else {
if (keynames.empty()) {
throw httpd::bad_param_exception("The keyspace of column families must be specified");
}
if (keynames.size() > 1) {
throw httpd::bad_param_exception("Only one keyspace allowed when specifying a column family");
}
resp = snap_ctl.local().take_column_family_snapshot(keynames[0], column_families, tag);
}
return resp.then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::del_snapshot.set(r, [&snap_ctl](std::unique_ptr<request> req) {
auto tag = req->get_query_param("tag");
auto column_family = req->get_query_param("cf");
std::vector<sstring> keynames = split(req->get_query_param("kn"), ",");
return snap_ctl.local().clear_snapshot(tag, keynames, column_family).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
ss::true_snapshots_size.set(r, [&snap_ctl](std::unique_ptr<request> req) {
return snap_ctl.local().true_snapshots_size().then([] (int64_t size) {
return make_ready_future<json::json_return_type>(size);
});
});
ss::scrub.set(r, wrap_ks_cf(ctx, [&snap_ctl] (http_context& ctx, std::unique_ptr<request> req, sstring keyspace, std::vector<sstring> column_families) {
const auto skip_corrupted = req_param<bool>(*req, "skip_corrupted", false);
auto f = make_ready_future<>();
if (!req_param<bool>(*req, "disable_snapshot", false)) {
auto tag = format("pre-scrub-{:d}", db_clock::now().time_since_epoch().count());
f = parallel_for_each(column_families, [&snap_ctl, keyspace, tag](sstring cf) {
return snap_ctl.local().take_column_family_snapshot(keyspace, cf, tag);
});
}
return f.then([&ctx, keyspace, column_families, skip_corrupted] {
return ctx.db.invoke_on_all([=] (database& db) {
return do_for_each(column_families, [=, &db](sstring cfname) {
auto& cm = db.get_compaction_manager();
auto& cf = db.find_column_family(keyspace, cfname);
return cm.perform_sstable_scrub(&cf, skip_corrupted);
});
});
}).then([]{
return make_ready_future<json::json_return_type>(0);
});
}));
}
void unset_snapshot(http_context& ctx, routes& r) {
ss::get_snapshot_details.unset(r);
ss::take_snapshot.unset(r);
ss::del_snapshot.unset(r);
ss::true_snapshots_size.unset(r);
ss::scrub.unset(r);
}
}

View File

@@ -21,24 +21,10 @@
#pragma once
#include <seastar/core/sharded.hh>
#include "api.hh"
namespace cql_transport { class controller; }
class thrift_controller;
namespace db { class snapshot_ctl; }
namespace netw { class messaging_service; }
namespace api {
void set_storage_service(http_context& ctx, routes& r);
void set_repair(http_context& ctx, routes& r, sharded<netw::messaging_service>& ms);
void unset_repair(http_context& ctx, routes& r);
void set_transport_controller(http_context& ctx, routes& r, cql_transport::controller& ctl);
void unset_transport_controller(http_context& ctx, routes& r);
void set_rpc_controller(http_context& ctx, routes& r, thrift_controller& ctl);
void unset_rpc_controller(http_context& ctx, routes& r);
void set_snapshot(http_context& ctx, routes& r, sharded<db::snapshot_ctl>& snap_ctl);
void unset_snapshot(http_context& ctx, routes& r);
}

View File

@@ -22,8 +22,7 @@
#include "api/api-doc/system.json.hh"
#include "api/api.hh"
#include <seastar/core/reactor.hh>
#include <seastar/http/exception.hh>
#include "http/exception.hh"
#include "log.hh"
namespace api {
@@ -31,10 +30,6 @@ namespace api {
namespace hs = httpd::system_json;
void set_system(http_context& ctx, routes& r) {
hs::get_system_uptime.set(r, [](const_req req) {
return std::chrono::duration_cast<std::chrono::milliseconds>(engine().uptime()).count();
});
hs::get_all_logger_names.set(r, [](const_req req) {
return logging::logger_registry().get_all_logger_names();
});

View File

@@ -21,7 +21,6 @@
#include "atomic_cell.hh"
#include "atomic_cell_or_collection.hh"
#include "counters.hh"
#include "types.hh"
/// LSA mirator for cells with irrelevant type
@@ -148,6 +147,35 @@ atomic_cell_or_collection::atomic_cell_or_collection(const abstract_type& type,
{
}
static collection_mutation_view get_collection_mutation_view(const uint8_t* ptr)
{
auto f = data::cell::structure::get_member<data::cell::tags::flags>(ptr);
auto ti = data::type_info::make_collection();
data::cell::context ctx(f, ti);
auto view = data::cell::structure::get_member<data::cell::tags::cell>(ptr).as<data::cell::tags::collection>(ctx);
auto dv = data::cell::variable_value::make_view(view, f.get<data::cell::tags::external_data>());
return collection_mutation_view { dv };
}
collection_mutation_view atomic_cell_or_collection::as_collection_mutation() const {
return get_collection_mutation_view(_data.get());
}
collection_mutation::collection_mutation(const collection_type_impl& type, collection_mutation_view v)
: _data(imr_object_type::make(data::cell::make_collection(v.data), &type.imr_state().lsa_migrator()))
{
}
collection_mutation::collection_mutation(const collection_type_impl& type, bytes_view v)
: _data(imr_object_type::make(data::cell::make_collection(v), &type.imr_state().lsa_migrator()))
{
}
collection_mutation::operator collection_mutation_view() const
{
return get_collection_mutation_view(_data.get());
}
bool atomic_cell_or_collection::equals(const abstract_type& type, const atomic_cell_or_collection& other) const
{
auto ptr_a = _data.get();
@@ -163,20 +191,20 @@ bool atomic_cell_or_collection::equals(const abstract_type& type, const atomic_c
if (a.timestamp() != b.timestamp()) {
return false;
}
if (a.is_live() != b.is_live()) {
return false;
}
if (a.is_live()) {
if (a.is_counter_update() != b.is_counter_update()) {
if (!b.is_live()) {
return false;
}
if (a.is_counter_update()) {
if (!b.is_counter_update()) {
return false;
}
return a.counter_update_value() == b.counter_update_value();
}
if (a.is_live_and_has_ttl() != b.is_live_and_has_ttl()) {
return false;
}
if (a.is_live_and_has_ttl()) {
if (!b.is_live_and_has_ttl()) {
return false;
}
if (a.ttl() != b.ttl() || a.expiry() != b.expiry()) {
return false;
}
@@ -202,86 +230,29 @@ size_t atomic_cell_or_collection::external_memory_usage(const abstract_type& t)
size_t external_value_size = 0;
if (flags.get<data::cell::tags::external_data>()) {
if (flags.get<data::cell::tags::collection>()) {
external_value_size = as_collection_mutation().data.size_bytes();
external_value_size = get_collection_mutation_view(_data.get()).data.size_bytes();
} else {
auto cell_view = data::cell::atomic_cell_view(t.imr_state().type_info(), view);
external_value_size = cell_view.value_size();
}
// Add overhead of chunk headers. The last one is a special case.
external_value_size += (external_value_size - 1) / data::cell::effective_external_chunk_length * data::cell::external_chunk_overhead;
external_value_size += (external_value_size - 1) / data::cell::maximum_external_chunk_length * data::cell::external_chunk_overhead;
external_value_size += data::cell::external_last_chunk_overhead;
}
return data::cell::structure::serialized_object_size(_data.get(), ctx)
+ imr_object_type::size_overhead + external_value_size;
}
std::ostream&
operator<<(std::ostream& os, const atomic_cell_view& acv) {
if (acv.is_live()) {
return fmt_print(os, "atomic_cell{{{},ts={:d},expiry={:d},ttl={:d}}}",
acv.is_counter_update()
? "counter_update_value=" + to_sstring(acv.counter_update_value())
: to_hex(acv.value().linearize()),
acv.timestamp(),
acv.is_live_and_has_ttl() ? acv.expiry().time_since_epoch().count() : -1,
acv.is_live_and_has_ttl() ? acv.ttl().count() : 0);
} else {
return fmt_print(os, "atomic_cell{{DEAD,ts={:d},deletion_time={:d}}}",
acv.timestamp(), acv.deletion_time().time_since_epoch().count());
}
}
std::ostream&
operator<<(std::ostream& os, const atomic_cell& ac) {
return os << atomic_cell_view(ac);
}
std::ostream&
operator<<(std::ostream& os, const atomic_cell_view::printer& acvp) {
auto& type = acvp._type;
auto& acv = acvp._cell;
if (acv.is_live()) {
std::ostringstream cell_value_string_builder;
if (type.is_counter()) {
if (acv.is_counter_update()) {
cell_value_string_builder << "counter_update_value=" << acv.counter_update_value();
} else {
cell_value_string_builder << "shards: ";
counter_cell_view::with_linearized(acv, [&cell_value_string_builder] (counter_cell_view& ccv) {
cell_value_string_builder << ::join(", ", ccv.shards());
});
}
} else {
cell_value_string_builder << type.to_string(acv.value().linearize());
}
return fmt_print(os, "atomic_cell{{{},ts={:d},expiry={:d},ttl={:d}}}",
cell_value_string_builder.str(),
acv.timestamp(),
acv.is_live_and_has_ttl() ? acv.expiry().time_since_epoch().count() : -1,
acv.is_live_and_has_ttl() ? acv.ttl().count() : 0);
} else {
return fmt_print(os, "atomic_cell{{DEAD,ts={:d},deletion_time={:d}}}",
acv.timestamp(), acv.deletion_time().time_since_epoch().count());
}
}
std::ostream&
operator<<(std::ostream& os, const atomic_cell::printer& acp) {
return operator<<(os, static_cast<const atomic_cell_view::printer&>(acp));
}
std::ostream& operator<<(std::ostream& os, const atomic_cell_or_collection::printer& p) {
if (!p._cell._data.get()) {
std::ostream& operator<<(std::ostream& os, const atomic_cell_or_collection& c) {
if (!c._data.get()) {
return os << "{ null atomic_cell_or_collection }";
}
using dc = data::cell;
os << "{ ";
if (dc::structure::get_member<dc::tags::flags>(p._cell._data.get()).get<dc::tags::collection>()) {
os << "collection ";
auto cmv = p._cell.as_collection_mutation();
os << collection_mutation_view::printer(*p._cdef.type, cmv);
if (dc::structure::get_member<dc::tags::flags>(c._data.get()).get<dc::tags::collection>()) {
os << "collection";
} else {
os << atomic_cell_view::printer(*p._cdef.type, p._cell.as_atomic_cell(p._cdef));
os << "atomic cell";
}
return os << " }";
return os << " @" << static_cast<const void*>(c._data.get()) << " }";
}

View File

@@ -26,9 +26,10 @@
#include "tombstone.hh"
#include "gc_clock.hh"
#include "utils/managed_bytes.hh"
#include <seastar/net//byteorder.hh>
#include "net/byteorder.hh"
#include <cstdint>
#include <iosfwd>
#include <seastar/util/gcc6-concepts.hh>
#include "data/cell.hh"
#include "data/schema_info.hh"
#include "imr/utils.hh"
@@ -38,7 +39,6 @@
class abstract_type;
class collection_type_impl;
class atomic_cell_or_collection;
using atomic_cell_value_view = data::value_view;
using atomic_cell_value_mutable_view = data::value_mutable_view;
@@ -153,14 +153,6 @@ public:
}
friend std::ostream& operator<<(std::ostream& os, const atomic_cell_view& acv);
class printer {
const abstract_type& _type;
const atomic_cell_view& _cell;
public:
printer(const abstract_type& type, const atomic_cell_view& cell) : _type(type), _cell(cell) {}
friend std::ostream& operator<<(std::ostream& os, const printer& acvp);
};
};
class atomic_cell_mutable_view final : public basic_atomic_cell_view<mutable_view::yes> {
@@ -227,12 +219,30 @@ public:
static atomic_cell make_live_uninitialized(const abstract_type& type, api::timestamp_type timestamp, size_t size);
friend class atomic_cell_or_collection;
friend std::ostream& operator<<(std::ostream& os, const atomic_cell& ac);
};
class printer : atomic_cell_view::printer {
public:
printer(const abstract_type& type, const atomic_cell_view& cell) : atomic_cell_view::printer(type, cell) {}
friend std::ostream& operator<<(std::ostream& os, const printer& acvp);
};
class collection_mutation_view;
// Represents a mutation of a collection. Actual format is determined by collection type,
// and is:
// set: list of atomic_cell
// map: list of pair<atomic_cell, bytes> (for key/value)
// list: tbd, probably ugly
class collection_mutation {
public:
using imr_object_type = imr::utils::object<data::cell::structure>;
imr_object_type _data;
collection_mutation() {}
collection_mutation(const collection_type_impl&, collection_mutation_view v);
collection_mutation(const collection_type_impl&, bytes_view bv);
operator collection_mutation_view() const;
};
class collection_mutation_view {
public:
atomic_cell_value_view data;
};
class column_definition;

View File

@@ -24,7 +24,6 @@
// Not part of atomic_cell.hh to avoid cyclic dependency between types.hh and atomic_cell.hh
#include "types.hh"
#include "types/collection.hh"
#include "atomic_cell.hh"
#include "atomic_cell_or_collection.hh"
#include "hashing.hh"
@@ -34,12 +33,14 @@ template<>
struct appending_hash<collection_mutation_view> {
template<typename Hasher>
void operator()(Hasher& h, collection_mutation_view cell, const column_definition& cdef) const {
cell.with_deserialized(*cdef.type, [&] (collection_mutation_view_description m_view) {
::feed_hash(h, m_view.tomb);
for (auto&& key_and_value : m_view.cells) {
::feed_hash(h, key_and_value.first);
::feed_hash(h, key_and_value.second, cdef);
}
cell.data.with_linearized([&] (bytes_view cell_bv) {
auto ctype = static_pointer_cast<const collection_type_impl>(cdef.type);
auto m_view = ctype->deserialize_mutation_form(cell_bv);
::feed_hash(h, m_view.tomb);
for (auto&& key_and_value : m_view.cells) {
::feed_hash(h, key_and_value.first);
::feed_hash(h, key_and_value.second, cdef);
}
});
}
};

View File

@@ -22,7 +22,6 @@
#pragma once
#include "atomic_cell.hh"
#include "collection_mutation.hh"
#include "schema.hh"
#include "hashing.hh"
@@ -68,19 +67,7 @@ public:
bytes_view serialize() const;
bool equals(const abstract_type& type, const atomic_cell_or_collection& other) const;
size_t external_memory_usage(const abstract_type&) const;
class printer {
const column_definition& _cdef;
const atomic_cell_or_collection& _cell;
public:
printer(const column_definition& cdef, const atomic_cell_or_collection& cell)
: _cdef(cdef), _cell(cell) { }
printer(const printer&) = delete;
printer(printer&&) = delete;
friend std::ostream& operator<<(std::ostream&, const printer&);
};
friend std::ostream& operator<<(std::ostream&, const printer&);
friend std::ostream& operator<<(std::ostream&, const atomic_cell_or_collection&);
};
namespace std {

View File

@@ -26,7 +26,10 @@
namespace auth {
constexpr std::string_view allow_all_authenticator_name("org.apache.cassandra.auth.AllowAllAuthenticator");
const sstring& allow_all_authenticator_name() {
static const sstring name = meta::AUTH_PACKAGE_NAME + "AllowAllAuthenticator";
return name;
}
// To ensure correct initialization order, we unfortunately need to use a string literal.
static const class_registrator<

View File

@@ -37,7 +37,7 @@ class migration_manager;
namespace auth {
extern const std::string_view allow_all_authenticator_name;
const sstring& allow_all_authenticator_name();
class allow_all_authenticator final : public authenticator {
public:
@@ -52,8 +52,8 @@ public:
return make_ready_future<>();
}
virtual std::string_view qualified_java_name() const override {
return allow_all_authenticator_name;
virtual const sstring& qualified_java_name() const override {
return allow_all_authenticator_name();
}
virtual bool require_authentication() const override {
@@ -72,19 +72,19 @@ public:
return make_ready_future<authenticated_user>(anonymous_user());
}
virtual future<> create(std::string_view, const authentication_options& options) const override {
virtual future<> create(stdx::string_view, const authentication_options& options) const override {
return make_ready_future();
}
virtual future<> alter(std::string_view, const authentication_options& options) const override {
virtual future<> alter(stdx::string_view, const authentication_options& options) const override {
return make_ready_future();
}
virtual future<> drop(std::string_view) const override {
virtual future<> drop(stdx::string_view) const override {
return make_ready_future();
}
virtual future<custom_options> query_custom_options(std::string_view role_name) const override {
virtual future<custom_options> query_custom_options(stdx::string_view role_name) const override {
return make_ready_future<custom_options>();
}

View File

@@ -26,7 +26,10 @@
namespace auth {
constexpr std::string_view allow_all_authorizer_name("org.apache.cassandra.auth.AllowAllAuthorizer");
const sstring& allow_all_authorizer_name() {
static const sstring name = meta::AUTH_PACKAGE_NAME + "AllowAllAuthorizer";
return name;
}
// To ensure correct initialization order, we unfortunately need to use a string literal.
static const class_registrator<

View File

@@ -23,6 +23,7 @@
#include "auth/authorizer.hh"
#include "exceptions/exceptions.hh"
#include "stdx.hh"
namespace cql3 {
class query_processor;
@@ -34,7 +35,7 @@ class migration_manager;
namespace auth {
extern const std::string_view allow_all_authorizer_name;
const sstring& allow_all_authorizer_name();
class allow_all_authorizer final : public authorizer {
public:
@@ -49,20 +50,20 @@ public:
return make_ready_future<>();
}
virtual std::string_view qualified_java_name() const override {
return allow_all_authorizer_name;
virtual const sstring& qualified_java_name() const override {
return allow_all_authorizer_name();
}
virtual future<permission_set> authorize(const role_or_anonymous&, const resource&) const override {
return make_ready_future<permission_set>(permissions::ALL);
}
virtual future<> grant(std::string_view, permission_set, const resource&) const override {
virtual future<> grant(stdx::string_view, permission_set, const resource&) const override {
return make_exception_future<>(
unsupported_authorization_operation("GRANT operation is not supported by AllowAllAuthorizer"));
}
virtual future<> revoke(std::string_view, permission_set, const resource&) const override {
virtual future<> revoke(stdx::string_view, permission_set, const resource&) const override {
return make_exception_future<>(
unsupported_authorization_operation("REVOKE operation is not supported by AllowAllAuthorizer"));
}
@@ -73,7 +74,7 @@ public:
"LIST PERMISSIONS operation is not supported by AllowAllAuthorizer"));
}
virtual future<> revoke_all(std::string_view) const override {
virtual future<> revoke_all(stdx::string_view) const override {
return make_exception_future(
unsupported_authorization_operation("REVOKE operation is not supported by AllowAllAuthorizer"));
}

View File

@@ -45,7 +45,7 @@
namespace auth {
authenticated_user::authenticated_user(std::string_view name)
authenticated_user::authenticated_user(stdx::string_view name)
: name(sstring(name)) {
}

View File

@@ -41,7 +41,7 @@
#pragma once
#include <string_view>
#include <experimental/string_view>
#include <functional>
#include <iosfwd>
#include <optional>
@@ -49,6 +49,7 @@
#include <seastar/core/sstring.hh>
#include "seastarx.hh"
#include "stdx.hh"
namespace auth {
@@ -66,7 +67,7 @@ public:
/// An anonymous user.
///
authenticated_user() = default;
explicit authenticated_user(std::string_view name);
explicit authenticated_user(stdx::string_view name);
};
///

View File

@@ -57,7 +57,7 @@ inline bool any_authentication_options(const authentication_options& aos) noexce
class unsupported_authentication_option : public std::invalid_argument {
public:
explicit unsupported_authentication_option(authentication_option k)
: std::invalid_argument(format("The {} option is not supported.", k)) {
: std::invalid_argument(sprint("The %s option is not supported.", k)) {
}
};

View File

@@ -45,6 +45,7 @@
#include "auth/common.hh"
#include "auth/password_authenticator.hh"
#include "cql3/query_processor.hh"
#include "db/config.hh"
#include "utils/class_registrator.hh"
const sstring auth::authenticator::USERNAME_KEY("username");

View File

@@ -41,7 +41,7 @@
#pragma once
#include <string_view>
#include <experimental/string_view>
#include <memory>
#include <set>
#include <stdexcept>
@@ -55,10 +55,10 @@
#include "auth/authentication_options.hh"
#include "auth/resource.hh"
#include "auth/sasl_challenge.hh"
#include "bytes.hh"
#include "enum_set.hh"
#include "exceptions/exceptions.hh"
#include "stdx.hh"
namespace db {
class config;
@@ -96,7 +96,7 @@ public:
///
/// A fully-qualified (class with package) Java-like name for this implementation.
///
virtual std::string_view qualified_java_name() const = 0;
virtual const sstring& qualified_java_name() const = 0;
virtual bool require_authentication() const = 0;
@@ -122,7 +122,7 @@ public:
///
/// The options provided must be a subset of `supported_options()`.
///
virtual future<> create(std::string_view role_name, const authentication_options& options) const = 0;
virtual future<> create(stdx::string_view role_name, const authentication_options& options) const = 0;
///
/// Alter the authentication record of an existing user.
@@ -131,25 +131,39 @@ public:
///
/// Callers must ensure that the specification of `alterable_options()` is adhered to.
///
virtual future<> alter(std::string_view role_name, const authentication_options& options) const = 0;
virtual future<> alter(stdx::string_view role_name, const authentication_options& options) const = 0;
///
/// Delete the authentication record for a user. This will disallow the user from logging in.
///
virtual future<> drop(std::string_view role_name) const = 0;
virtual future<> drop(stdx::string_view role_name) const = 0;
///
/// Query for custom options (those corresponding to \ref authentication_options::options).
///
/// If no options are set the result is an empty container.
///
virtual future<custom_options> query_custom_options(std::string_view role_name) const = 0;
virtual future<custom_options> query_custom_options(stdx::string_view role_name) const = 0;
///
/// System resources used internally as part of the implementation. These are made inaccessible to users.
///
virtual const resource_set& protected_resources() const = 0;
///
/// A stateful SASL challenge which supports many authentication schemes (depending on the implementation).
///
class sasl_challenge {
public:
virtual ~sasl_challenge() = default;
virtual bytes evaluate_response(bytes_view client_response) = 0;
virtual bool is_complete() const = 0;
virtual future<authenticated_user> get_authenticated_user() const = 0;
};
virtual ::shared_ptr<sasl_challenge> new_sasl_challenge() const = 0;
};

View File

@@ -41,7 +41,7 @@
#pragma once
#include <string_view>
#include <experimental/string_view>
#include <functional>
#include <optional>
#include <stdexcept>
@@ -54,6 +54,7 @@
#include "auth/permission.hh"
#include "auth/resource.hh"
#include "seastarx.hh"
#include "stdx.hh"
namespace auth {
@@ -100,7 +101,7 @@ public:
///
/// A fully-qualified (class with package) Java-like name for this implementation.
///
virtual std::string_view qualified_java_name() const = 0;
virtual const sstring& qualified_java_name() const = 0;
///
/// Query for the permissions granted directly to a role for a particular \ref resource (and not any of its
@@ -116,14 +117,14 @@ public:
///
/// \throws \ref unsupported_authorization_operation if granting permissions is not supported.
///
virtual future<> grant(std::string_view role_name, permission_set, const resource&) const = 0;
virtual future<> grant(stdx::string_view role_name, permission_set, const resource&) const = 0;
///
/// Revoke a set of permissions from a role for a particular \ref resource.
///
/// \throws \ref unsupported_authorization_operation if revoking permissions is not supported.
///
virtual future<> revoke(std::string_view role_name, permission_set, const resource&) const = 0;
virtual future<> revoke(stdx::string_view role_name, permission_set, const resource&) const = 0;
///
/// Query for all directly granted permissions.
@@ -137,7 +138,7 @@ public:
///
/// \throws \ref unsupported_authorization_operation if revoking permissions is not supported.
///
virtual future<> revoke_all(std::string_view role_name) const = 0;
virtual future<> revoke_all(stdx::string_view role_name) const = 0;
///
/// Revoke all permissions granted to any role for a particular resource.

View File

@@ -34,9 +34,10 @@ namespace auth {
namespace meta {
constexpr std::string_view AUTH_KS("system_auth");
constexpr std::string_view USERS_CF("users");
constexpr std::string_view AUTH_PACKAGE_NAME("org.apache.cassandra.auth.");
const sstring DEFAULT_SUPERUSER_NAME("cassandra");
const sstring AUTH_KS("system_auth");
const sstring USERS_CF("users");
const sstring AUTH_PACKAGE_NAME("org.apache.cassandra.auth.");
}
@@ -47,9 +48,9 @@ future<> do_after_system_ready(seastar::abort_source& as, seastar::noncopyable_f
struct empty_state { };
return delay_until_system_ready(as).then([&as, func = std::move(func)] () mutable {
return exponential_backoff_retry::do_until_value(1s, 1min, as, [func = std::move(func)] {
return func().then_wrapped([] (auto&& f) -> std::optional<empty_state> {
return func().then_wrapped([] (auto&& f) -> stdx::optional<empty_state> {
if (f.failed()) {
auth_log.debug("Auth task failed with error, rescheduling: {}", f.get_exception());
auth_log.info("Auth task failed with error, rescheduling: {}", f.get_exception());
return { };
}
return { empty_state() };
@@ -58,40 +59,32 @@ future<> do_after_system_ready(seastar::abort_source& as, seastar::noncopyable_f
}).discard_result();
}
static future<> create_metadata_table_if_missing_impl(
std::string_view table_name,
future<> create_metadata_table_if_missing(
stdx::string_view table_name,
cql3::query_processor& qp,
std::string_view cql,
stdx::string_view cql,
::service::migration_manager& mm) {
static auto ignore_existing = [] (seastar::noncopyable_function<future<>()> func) {
return futurize_invoke(std::move(func)).handle_exception_type([] (exceptions::already_exists_exception& ignored) { });
};
auto& db = qp.db();
auto parsed_statement = cql3::query_processor::parse_statement(cql);
auto& parsed_cf_statement = static_cast<cql3::statements::raw::cf_statement&>(*parsed_statement);
auto& db = qp.db().local();
parsed_cf_statement.prepare_keyspace(meta::AUTH_KS);
if (db.has_schema(meta::AUTH_KS, sstring(table_name))) {
return make_ready_future<>();
}
auto parsed_statement = static_pointer_cast<cql3::statements::raw::cf_statement>(
cql3::query_processor::parse_statement(cql));
parsed_statement->prepare_keyspace(meta::AUTH_KS);
auto statement = static_pointer_cast<cql3::statements::create_table_statement>(
parsed_cf_statement.prepare(db, qp.get_cql_stats())->statement);
parsed_statement->prepare(db, qp.get_cql_stats())->statement);
const auto schema = statement->get_cf_meta_data(qp.db());
const auto schema = statement->get_cf_meta_data(qp.db().local());
const auto uuid = generate_legacy_id(schema->ks_name(), schema->cf_name());
schema_builder b(schema);
b.set_uuid(uuid);
schema_ptr table = b.build();
return ignore_existing([&mm, table = std::move(table)] () {
return mm.announce_new_column_family(table, false);
});
}
future<> create_metadata_table_if_missing(
std::string_view table_name,
cql3::query_processor& qp,
std::string_view cql,
::service::migration_manager& mm) noexcept {
return futurize_invoke(create_metadata_table_if_missing_impl, table_name, qp, cql, mm);
return mm.announce_new_column_family(b.build(), false);
}
future<> wait_for_schema_agreement(::service::migration_manager& mm, const database& db, seastar::abort_source& as) {
@@ -108,17 +101,10 @@ future<> wait_for_schema_agreement(::service::migration_manager& mm, const datab
});
}
::service::query_state& internal_distributed_query_state() noexcept {
#ifdef DEBUG
// Give the much slower debug tests more headroom for completing auth queries.
static const auto t = 30s;
#else
const timeout_config& internal_distributed_timeout_config() noexcept {
static const auto t = 5s;
#endif
static const timeout_config tc{t, t, t, t, t, t, t};
static thread_local ::service::client_state cs(::service::client_state::internal_tag{}, tc);
static thread_local ::service::query_state qs(cs, empty_service_permit());
return qs;
return tc;
}
}

View File

@@ -22,20 +22,18 @@
#pragma once
#include <chrono>
#include <string_view>
#include <experimental/string_view>
#include <seastar/core/future.hh>
#include <seastar/core/abort_source.hh>
#include <seastar/util/noncopyable_function.hh>
#include <seastar/core/seastar.hh>
#include <seastar/core/reactor.hh>
#include <seastar/core/resource.hh>
#include <seastar/core/sstring.hh>
#include <seastar/core/smp.hh>
#include "log.hh"
#include "seastarx.hh"
#include "utils/exponential_backoff_retry.hh"
#include "service/query_state.hh"
using namespace std::chrono_literals;
@@ -54,16 +52,16 @@ namespace auth {
namespace meta {
constexpr std::string_view DEFAULT_SUPERUSER_NAME("cassandra");
extern const std::string_view AUTH_KS;
extern const std::string_view USERS_CF;
extern const std::string_view AUTH_PACKAGE_NAME;
extern const sstring DEFAULT_SUPERUSER_NAME;
extern const sstring AUTH_KS;
extern const sstring USERS_CF;
extern const sstring AUTH_PACKAGE_NAME;
}
template <class Task>
future<> once_among_shards(Task&& f) {
if (this_shard_id() == 0u) {
if (engine().cpu_id() == 0u) {
return f();
}
@@ -78,16 +76,16 @@ inline future<> delay_until_system_ready(seastar::abort_source& as) {
future<> do_after_system_ready(seastar::abort_source& as, seastar::noncopyable_function<future<>()> func);
future<> create_metadata_table_if_missing(
std::string_view table_name,
stdx::string_view table_name,
cql3::query_processor&,
std::string_view cql,
::service::migration_manager&) noexcept;
stdx::string_view cql,
::service::migration_manager&);
future<> wait_for_schema_agreement(::service::migration_manager&, const database&, seastar::abort_source&);
///
/// Time-outs for internal, non-local CQL queries.
///
::service::query_state& internal_distributed_query_state() noexcept;
const timeout_config& internal_distributed_timeout_config() noexcept;
}

View File

@@ -51,7 +51,7 @@ extern "C" {
#include <boost/algorithm/string/join.hpp>
#include <boost/range.hpp>
#include <seastar/core/seastar.hh>
#include <seastar/core/reactor.hh>
#include "auth/authenticated_user.hh"
#include "auth/common.hh"
@@ -61,18 +61,18 @@ extern "C" {
#include "cql3/untyped_result_set.hh"
#include "exceptions/exceptions.hh"
#include "log.hh"
#include "database.hh"
namespace auth {
std::string_view default_authorizer::qualified_java_name() const {
return "org.apache.cassandra.auth.CassandraAuthorizer";
const sstring& default_authorizer_name() {
static const sstring name = meta::AUTH_PACKAGE_NAME + "CassandraAuthorizer";
return name;
}
static constexpr std::string_view ROLE_NAME = "role";
static constexpr std::string_view RESOURCE_NAME = "resource";
static constexpr std::string_view PERMISSIONS_NAME = "permissions";
static constexpr std::string_view PERMISSIONS_CF = "role_permissions";
static const sstring ROLE_NAME = "role";
static const sstring RESOURCE_NAME = "resource";
static const sstring PERMISSIONS_NAME = "permissions";
static const sstring PERMISSIONS_CF = "role_permissions";
static logging::logger alogger("default_authorizer");
@@ -94,15 +94,16 @@ default_authorizer::~default_authorizer() {
static const sstring legacy_table_name{"permissions"};
bool default_authorizer::legacy_metadata_exists() const {
return _qp.db().has_schema(meta::AUTH_KS, legacy_table_name);
return _qp.db().local().has_schema(meta::AUTH_KS, legacy_table_name);
}
future<bool> default_authorizer::any_granted() const {
static const sstring query = format("SELECT * FROM {}.{} LIMIT 1", meta::AUTH_KS, PERMISSIONS_CF);
static const sstring query = sprint("SELECT * FROM %s.%s LIMIT 1", meta::AUTH_KS, PERMISSIONS_CF);
return _qp.execute_internal(
return _qp.process(
query,
db::consistency_level::LOCAL_ONE,
infinite_timeout_config,
{},
true).then([this](::shared_ptr<cql3::untyped_result_set> results) {
return !results->empty();
@@ -111,11 +112,12 @@ future<bool> default_authorizer::any_granted() const {
future<> default_authorizer::migrate_legacy_metadata() const {
alogger.info("Starting migration of legacy permissions metadata.");
static const sstring query = format("SELECT * FROM {}.{}", meta::AUTH_KS, legacy_table_name);
static const sstring query = sprint("SELECT * FROM %s.%s", meta::AUTH_KS, legacy_table_name);
return _qp.execute_internal(
return _qp.process(
query,
db::consistency_level::LOCAL_ONE).then([this](::shared_ptr<cql3::untyped_result_set> results) {
db::consistency_level::LOCAL_ONE,
infinite_timeout_config).then([this](::shared_ptr<cql3::untyped_result_set> results) {
return do_for_each(*results, [this](const cql3::untyped_result_set_row& row) {
return do_with(
row.get_as<sstring>("username"),
@@ -158,7 +160,7 @@ future<> default_authorizer::start() {
_migration_manager).then([this] {
_finished = do_after_system_ready(_as, [this] {
return async([this] {
wait_for_schema_agreement(_migration_manager, _qp.db(), _as).get0();
wait_for_schema_agreement(_migration_manager, _qp.db().local(), _as).get0();
if (legacy_metadata_exists()) {
if (!any_granted().get0()) {
@@ -185,16 +187,18 @@ default_authorizer::authorize(const role_or_anonymous& maybe_role, const resourc
return make_ready_future<permission_set>(permissions::NONE);
}
static const sstring query = format("SELECT {} FROM {}.{} WHERE {} = ? AND {} = ?",
static const sstring query = sprint(
"SELECT %s FROM %s.%s WHERE %s = ? AND %s = ?",
PERMISSIONS_NAME,
meta::AUTH_KS,
PERMISSIONS_CF,
ROLE_NAME,
RESOURCE_NAME);
return _qp.execute_internal(
return _qp.process(
query,
db::consistency_level::LOCAL_ONE,
infinite_timeout_config,
{*maybe_role.name, r.name()}).then([](::shared_ptr<cql3::untyped_result_set> results) {
if (results->empty()) {
return permissions::NONE;
@@ -206,12 +210,13 @@ default_authorizer::authorize(const role_or_anonymous& maybe_role, const resourc
future<>
default_authorizer::modify(
std::string_view role_name,
stdx::string_view role_name,
permission_set set,
const resource& resource,
std::string_view op) const {
stdx::string_view op) const {
return do_with(
format("UPDATE {}.{} SET {} = {} {} ? WHERE {} = ? AND {} = ?",
sprint(
"UPDATE %s.%s SET %s = %s %s ? WHERE %s = ? AND %s = ?",
meta::AUTH_KS,
PERMISSIONS_CF,
PERMISSIONS_NAME,
@@ -220,35 +225,36 @@ default_authorizer::modify(
ROLE_NAME,
RESOURCE_NAME),
[this, &role_name, set, &resource](const auto& query) {
return _qp.execute_internal(
return _qp.process(
query,
db::consistency_level::ONE,
internal_distributed_query_state(),
internal_distributed_timeout_config(),
{permissions::to_strings(set), sstring(role_name), resource.name()}).discard_result();
});
}
future<> default_authorizer::grant(std::string_view role_name, permission_set set, const resource& resource) const {
future<> default_authorizer::grant(stdx::string_view role_name, permission_set set, const resource& resource) const {
return modify(role_name, std::move(set), resource, "+");
}
future<> default_authorizer::revoke(std::string_view role_name, permission_set set, const resource& resource) const {
future<> default_authorizer::revoke(stdx::string_view role_name, permission_set set, const resource& resource) const {
return modify(role_name, std::move(set), resource, "-");
}
future<std::vector<permission_details>> default_authorizer::list_all() const {
static const sstring query = format("SELECT {}, {}, {} FROM {}.{}",
static const sstring query = sprint(
"SELECT %s, %s, %s FROM %s.%s",
ROLE_NAME,
RESOURCE_NAME,
PERMISSIONS_NAME,
meta::AUTH_KS,
PERMISSIONS_CF);
return _qp.execute_internal(
return _qp.process(
query,
db::consistency_level::ONE,
internal_distributed_query_state(),
internal_distributed_timeout_config(),
{},
true).then([](::shared_ptr<cql3::untyped_result_set> results) {
std::vector<permission_details> all_details;
@@ -266,16 +272,17 @@ future<std::vector<permission_details>> default_authorizer::list_all() const {
});
}
future<> default_authorizer::revoke_all(std::string_view role_name) const {
static const sstring query = format("DELETE FROM {}.{} WHERE {} = ?",
future<> default_authorizer::revoke_all(stdx::string_view role_name) const {
static const sstring query = sprint(
"DELETE FROM %s.%s WHERE %s = ?",
meta::AUTH_KS,
PERMISSIONS_CF,
ROLE_NAME);
return _qp.execute_internal(
return _qp.process(
query,
db::consistency_level::ONE,
internal_distributed_query_state(),
internal_distributed_timeout_config(),
{sstring(role_name)}).discard_result().handle_exception([role_name](auto ep) {
try {
std::rethrow_exception(ep);
@@ -286,15 +293,17 @@ future<> default_authorizer::revoke_all(std::string_view role_name) const {
}
future<> default_authorizer::revoke_all(const resource& resource) const {
static const sstring query = format("SELECT {} FROM {}.{} WHERE {} = ? ALLOW FILTERING",
static const sstring query = sprint(
"SELECT %s FROM %s.%s WHERE %s = ? ALLOW FILTERING",
ROLE_NAME,
meta::AUTH_KS,
PERMISSIONS_CF,
RESOURCE_NAME);
return _qp.execute_internal(
return _qp.process(
query,
db::consistency_level::LOCAL_ONE,
infinite_timeout_config,
{resource.name()}).then_wrapped([this, resource](future<::shared_ptr<cql3::untyped_result_set>> f) {
try {
auto res = f.get0();
@@ -302,15 +311,17 @@ future<> default_authorizer::revoke_all(const resource& resource) const {
res->begin(),
res->end(),
[this, res, resource](const cql3::untyped_result_set::row& r) {
static const sstring query = format("DELETE FROM {}.{} WHERE {} = ? AND {} = ?",
static const sstring query = sprint(
"DELETE FROM %s.%s WHERE %s = ? AND %s = ?",
meta::AUTH_KS,
PERMISSIONS_CF,
ROLE_NAME,
RESOURCE_NAME);
return _qp.execute_internal(
return _qp.process(
query,
db::consistency_level::LOCAL_ONE,
infinite_timeout_config,
{r.get_as<sstring>(ROLE_NAME), resource.name()}).discard_result().handle_exception(
[resource](auto ep) {
try {

View File

@@ -51,6 +51,8 @@
namespace auth {
const sstring& default_authorizer_name();
class default_authorizer : public authorizer {
cql3::query_processor& _qp;
@@ -69,17 +71,19 @@ public:
virtual future<> stop() override;
virtual std::string_view qualified_java_name() const override;
virtual const sstring& qualified_java_name() const override {
return default_authorizer_name();
}
virtual future<permission_set> authorize(const role_or_anonymous&, const resource&) const override;
virtual future<> grant(std::string_view, permission_set, const resource&) const override;
virtual future<> grant(stdx::string_view, permission_set, const resource&) const override;
virtual future<> revoke( std::string_view, permission_set, const resource&) const override;
virtual future<> revoke( stdx::string_view, permission_set, const resource&) const override;
virtual future<std::vector<permission_details>> list_all() const override;
virtual future<> revoke_all(std::string_view) const override;
virtual future<> revoke_all(stdx::string_view) const override;
virtual future<> revoke_all(const resource&) const override;
@@ -92,7 +96,7 @@ private:
future<> migrate_legacy_metadata() const;
future<> modify(std::string_view, permission_set, const resource&, std::string_view) const;
future<> modify(stdx::string_view, permission_set, const resource&, stdx::string_view) const;
};
} /* namespace auth */

View File

@@ -44,11 +44,9 @@
#include <algorithm>
#include <chrono>
#include <random>
#include <string_view>
#include <optional>
#include <boost/algorithm/cxx11/all_of.hpp>
#include <seastar/core/seastar.hh>
#include <seastar/core/reactor.hh>
#include "auth/authenticated_user.hh"
#include "auth/common.hh"
@@ -58,17 +56,18 @@
#include "log.hh"
#include "service/migration_manager.hh"
#include "utils/class_registrator.hh"
#include "database.hh"
namespace auth {
constexpr std::string_view password_authenticator_name("org.apache.cassandra.auth.PasswordAuthenticator");
const sstring& password_authenticator_name() {
static const sstring name = meta::AUTH_PACKAGE_NAME + "PasswordAuthenticator";
return name;
}
// name of the hash column.
static constexpr std::string_view SALTED_HASH = "salted_hash";
static constexpr std::string_view OPTIONS = "options";
static constexpr std::string_view DEFAULT_USER_NAME = meta::DEFAULT_SUPERUSER_NAME;
static const sstring DEFAULT_USER_PASSWORD = sstring(meta::DEFAULT_SUPERUSER_NAME);
static const sstring SALTED_HASH = "salted_hash";
static const sstring DEFAULT_USER_NAME = meta::DEFAULT_SUPERUSER_NAME;
static const sstring DEFAULT_USER_PASSWORD = meta::DEFAULT_SUPERUSER_NAME;
static logging::logger plogger("password_authenticator");
@@ -94,36 +93,34 @@ static bool has_salted_hash(const cql3::untyped_result_set_row& row) {
return !row.get_or<sstring>(SALTED_HASH, "").empty();
}
static const sstring& update_row_query() {
static const sstring update_row_query = format("UPDATE {} SET {} = ? WHERE {} = ?",
meta::roles_table::qualified_name,
SALTED_HASH,
meta::roles_table::role_col_name);
return update_row_query;
}
static const sstring update_row_query = sprint(
"UPDATE %s SET %s = ? WHERE %s = ?",
meta::roles_table::qualified_name(),
SALTED_HASH,
meta::roles_table::role_col_name);
static const sstring legacy_table_name{"credentials"};
bool password_authenticator::legacy_metadata_exists() const {
return _qp.db().has_schema(meta::AUTH_KS, legacy_table_name);
return _qp.db().local().has_schema(meta::AUTH_KS, legacy_table_name);
}
future<> password_authenticator::migrate_legacy_metadata() const {
plogger.info("Starting migration of legacy authentication metadata.");
static const sstring query = format("SELECT * FROM {}.{}", meta::AUTH_KS, legacy_table_name);
static const sstring query = sprint("SELECT * FROM %s.%s", meta::AUTH_KS, legacy_table_name);
return _qp.execute_internal(
return _qp.process(
query,
db::consistency_level::QUORUM,
internal_distributed_query_state()).then([this](::shared_ptr<cql3::untyped_result_set> results) {
internal_distributed_timeout_config()).then([this](::shared_ptr<cql3::untyped_result_set> results) {
return do_for_each(*results, [this](const cql3::untyped_result_set_row& row) {
auto username = row.get_as<sstring>("username");
auto salted_hash = row.get_as<sstring>(SALTED_HASH);
return _qp.execute_internal(
update_row_query(),
return _qp.process(
update_row_query,
consistency_for_user(username),
internal_distributed_query_state(),
internal_distributed_timeout_config(),
{std::move(salted_hash), username}).discard_result();
}).finally([results] {});
}).then([] {
@@ -137,10 +134,10 @@ future<> password_authenticator::migrate_legacy_metadata() const {
future<> password_authenticator::create_default_if_missing() const {
return default_role_row_satisfies(_qp, &has_salted_hash).then([this](bool exists) {
if (!exists) {
return _qp.execute_internal(
update_row_query(),
return _qp.process(
update_row_query,
db::consistency_level::QUORUM,
internal_distributed_query_state(),
internal_distributed_timeout_config(),
{passwords::hash(DEFAULT_USER_PASSWORD, rng_for_salt), DEFAULT_USER_NAME}).then([](auto&&) {
plogger.info("Created default superuser authentication record.");
});
@@ -160,7 +157,7 @@ future<> password_authenticator::start() {
_stopped = do_after_system_ready(_as, [this] {
return async([this] {
wait_for_schema_agreement(_migration_manager, _qp.db(), _as).get0();
wait_for_schema_agreement(_migration_manager, _qp.db().local(), _as).get0();
if (any_nondefault_role_row_satisfies(_qp, &has_salted_hash).get0()) {
if (legacy_metadata_exists()) {
@@ -188,15 +185,15 @@ future<> password_authenticator::stop() {
return _stopped.handle_exception_type([] (const sleep_aborted&) { }).handle_exception_type([](const abort_requested_exception&) {});
}
db::consistency_level password_authenticator::consistency_for_user(std::string_view role_name) {
db::consistency_level password_authenticator::consistency_for_user(stdx::string_view role_name) {
if (role_name == DEFAULT_USER_NAME) {
return db::consistency_level::QUORUM;
}
return db::consistency_level::LOCAL_ONE;
}
std::string_view password_authenticator::qualified_java_name() const {
return password_authenticator_name;
const sstring& password_authenticator::qualified_java_name() const {
return password_authenticator_name();
}
bool password_authenticator::require_authentication() const {
@@ -204,20 +201,20 @@ bool password_authenticator::require_authentication() const {
}
authentication_option_set password_authenticator::supported_options() const {
return authentication_option_set{authentication_option::password, authentication_option::options};
return authentication_option_set{authentication_option::password};
}
authentication_option_set password_authenticator::alterable_options() const {
return authentication_option_set{authentication_option::password, authentication_option::options};
return authentication_option_set{authentication_option::password};
}
future<authenticated_user> password_authenticator::authenticate(
const credentials_map& credentials) const {
if (!credentials.contains(USERNAME_KEY)) {
throw exceptions::authentication_exception(format("Required key '{}' is missing", USERNAME_KEY));
if (!credentials.count(USERNAME_KEY)) {
throw exceptions::authentication_exception(sprint("Required key '%s' is missing", USERNAME_KEY));
}
if (!credentials.contains(PASSWORD_KEY)) {
throw exceptions::authentication_exception(format("Required key '{}' is missing", PASSWORD_KEY));
if (!credentials.count(PASSWORD_KEY)) {
throw exceptions::authentication_exception(sprint("Required key '%s' is missing", PASSWORD_KEY));
}
auto& username = credentials.at(USERNAME_KEY);
@@ -228,22 +225,23 @@ future<authenticated_user> password_authenticator::authenticate(
// obsolete prepared statements pretty quickly.
// Rely on query processing caching statements instead, and lets assume
// that a map lookup string->statement is not gonna kill us much.
return futurize_invoke([this, username, password] {
static const sstring query = format("SELECT {} FROM {} WHERE {} = ?",
return futurize_apply([this, username, password] {
static const sstring query = sprint(
"SELECT %s FROM %s WHERE %s = ?",
SALTED_HASH,
meta::roles_table::qualified_name,
meta::roles_table::qualified_name(),
meta::roles_table::role_col_name);
return _qp.execute_internal(
return _qp.process(
query,
consistency_for_user(username),
internal_distributed_query_state(),
internal_distributed_timeout_config(),
{username},
true);
}).then_wrapped([=](future<::shared_ptr<cql3::untyped_result_set>> f) {
try {
auto res = f.get0();
auto salted_hash = std::optional<sstring>();
auto salted_hash = std::experimental::optional<sstring>();
if (!res->empty()) {
salted_hash = res->one().get_opt<sstring>(SALTED_HASH);
}
@@ -255,99 +253,57 @@ future<authenticated_user> password_authenticator::authenticate(
std::throw_with_nested(exceptions::authentication_exception("Could not verify password"));
} catch (exceptions::request_execution_exception& e) {
std::throw_with_nested(exceptions::authentication_exception(e.what()));
} catch (exceptions::authentication_exception& e) {
std::throw_with_nested(e);
} catch (...) {
std::throw_with_nested(exceptions::authentication_exception("authentication failed"));
}
});
}
future<> password_authenticator::maybe_update_custom_options(std::string_view role_name, const authentication_options& options) const {
static const sstring query = format("UPDATE {} SET {} = ? WHERE {} = ?",
meta::roles_table::qualified_name,
OPTIONS,
meta::roles_table::role_col_name);
if (!options.options) {
future<> password_authenticator::create(stdx::string_view role_name, const authentication_options& options) const {
if (!options.password) {
return make_ready_future<>();
}
std::vector<std::pair<data_value, data_value>> entries;
for (const auto& entry : *options.options) {
entries.push_back({data_value(entry.first), data_value(entry.second)});
}
auto map_value = make_map_value(map_type_impl::get_instance(utf8_type, utf8_type, false), entries);
return _qp.execute_internal(
query,
return _qp.process(
update_row_query,
consistency_for_user(role_name),
internal_distributed_query_state(),
{std::move(map_value), sstring(role_name)}).discard_result();
internal_distributed_timeout_config(),
{passwords::hash(*options.password, rng_for_salt), sstring(role_name)}).discard_result();
}
future<> password_authenticator::create(std::string_view role_name, const authentication_options& options) const {
future<> password_authenticator::alter(stdx::string_view role_name, const authentication_options& options) const {
if (!options.password) {
return maybe_update_custom_options(role_name, options);
return make_ready_future<>();
}
return _qp.execute_internal(
update_row_query(),
consistency_for_user(role_name),
internal_distributed_query_state(),
{passwords::hash(*options.password, rng_for_salt), sstring(role_name)}).discard_result().then([this, role_name, &options] {
return maybe_update_custom_options(role_name, options);
});
}
future<> password_authenticator::alter(std::string_view role_name, const authentication_options& options) const {
if (!options.password) {
return maybe_update_custom_options(role_name, options);
}
static const sstring query = format("UPDATE {} SET {} = ? WHERE {} = ?",
meta::roles_table::qualified_name,
static const sstring query = sprint(
"UPDATE %s SET %s = ? WHERE %s = ?",
meta::roles_table::qualified_name(),
SALTED_HASH,
meta::roles_table::role_col_name);
return _qp.execute_internal(
return _qp.process(
query,
consistency_for_user(role_name),
internal_distributed_query_state(),
{passwords::hash(*options.password, rng_for_salt), sstring(role_name)}).discard_result().then([this, role_name, &options] {
return maybe_update_custom_options(role_name, options);
}).discard_result();
internal_distributed_timeout_config(),
{passwords::hash(*options.password, rng_for_salt), sstring(role_name)}).discard_result();
}
future<> password_authenticator::drop(std::string_view name) const {
static const sstring query = format("DELETE {} FROM {} WHERE {} = ?",
future<> password_authenticator::drop(stdx::string_view name) const {
static const sstring query = sprint(
"DELETE %s FROM %s WHERE %s = ?",
SALTED_HASH,
meta::roles_table::qualified_name,
meta::roles_table::qualified_name(),
meta::roles_table::role_col_name);
return _qp.execute_internal(
return _qp.process(
query, consistency_for_user(name),
internal_distributed_query_state(),
internal_distributed_timeout_config(),
{sstring(name)}).discard_result();
}
future<custom_options> password_authenticator::query_custom_options(std::string_view role_name) const {
static const sstring query = format("SELECT {} FROM {} WHERE {} = ?",
OPTIONS,
meta::roles_table::qualified_name,
meta::roles_table::role_col_name);
return _qp.execute_internal(
query, consistency_for_user(role_name),
internal_distributed_query_state(),
{sstring(role_name)}).then([](::shared_ptr<cql3::untyped_result_set> rs) {
custom_options opts;
const auto& row = rs->one();
if (row.has(OPTIONS)) {
row.get_map_data<sstring, sstring>(OPTIONS, std::inserter(opts, opts.end()), utf8_type, utf8_type);
}
return opts;
});
future<custom_options> password_authenticator::query_custom_options(stdx::string_view role_name) const {
return make_ready_future<custom_options>();
}
const resource_set& password_authenticator::protected_resources() const {
@@ -355,13 +311,75 @@ const resource_set& password_authenticator::protected_resources() const {
return resources;
}
::shared_ptr<sasl_challenge> password_authenticator::new_sasl_challenge() const {
return ::make_shared<plain_sasl_challenge>([this](std::string_view username, std::string_view password) {
credentials_map credentials{};
credentials[USERNAME_KEY] = sstring(username);
credentials[PASSWORD_KEY] = sstring(password);
return this->authenticate(credentials);
});
::shared_ptr<authenticator::sasl_challenge> password_authenticator::new_sasl_challenge() const {
class plain_text_password_challenge : public sasl_challenge {
const password_authenticator& _self;
public:
plain_text_password_challenge(const password_authenticator& self) : _self(self) {
}
/**
* SASL PLAIN mechanism specifies that credentials are encoded in a
* sequence of UTF-8 bytes, delimited by 0 (US-ASCII NUL).
* The form is : {code}authzId<NUL>authnId<NUL>password<NUL>{code}
* authzId is optional, and in fact we don't care about it here as we'll
* set the authzId to match the authnId (that is, there is no concept of
* a user being authorized to act on behalf of another).
*
* @param bytes encoded credentials string sent by the client
* @return map containing the username/password pairs in the form an IAuthenticator
* would expect
* @throws javax.security.sasl.SaslException
*/
bytes evaluate_response(bytes_view client_response) override {
plogger.debug("Decoding credentials from client token");
sstring username, password;
auto b = client_response.crbegin();
auto e = client_response.crend();
auto i = b;
while (i != e) {
if (*i == 0) {
sstring tmp(i.base(), b.base());
if (password.empty()) {
password = std::move(tmp);
} else if (username.empty()) {
username = std::move(tmp);
}
b = ++i;
continue;
}
++i;
}
if (username.empty()) {
throw exceptions::authentication_exception("Authentication ID must not be null");
}
if (password.empty()) {
throw exceptions::authentication_exception("Password must not be null");
}
_credentials[USERNAME_KEY] = std::move(username);
_credentials[PASSWORD_KEY] = std::move(password);
_complete = true;
return {};
}
bool is_complete() const override {
return _complete;
}
future<authenticated_user> get_authenticated_user() const override {
return _self.authenticate(_credentials);
}
private:
credentials_map _credentials;
bool _complete = false;
};
return ::make_shared<plain_text_password_challenge>(*this);
}
}

View File

@@ -52,7 +52,7 @@ class migration_manager;
namespace auth {
extern const std::string_view password_authenticator_name;
const sstring& password_authenticator_name();
class password_authenticator : public authenticator {
cql3::query_processor& _qp;
@@ -61,7 +61,7 @@ class password_authenticator : public authenticator {
seastar::abort_source _as;
public:
static db::consistency_level consistency_for_user(std::string_view role_name);
static db::consistency_level consistency_for_user(stdx::string_view role_name);
password_authenticator(cql3::query_processor&, ::service::migration_manager&);
@@ -71,7 +71,7 @@ public:
virtual future<> stop() override;
virtual std::string_view qualified_java_name() const override;
virtual const sstring& qualified_java_name() const override;
virtual bool require_authentication() const override;
@@ -81,21 +81,19 @@ public:
virtual future<authenticated_user> authenticate(const credentials_map& credentials) const override;
virtual future<> create(std::string_view role_name, const authentication_options& options) const override;
virtual future<> create(stdx::string_view role_name, const authentication_options& options) const override;
virtual future<> alter(std::string_view role_name, const authentication_options& options) const override;
virtual future<> alter(stdx::string_view role_name, const authentication_options& options) const override;
virtual future<> drop(std::string_view role_name) const override;
virtual future<> drop(stdx::string_view role_name) const override;
virtual future<custom_options> query_custom_options(std::string_view role_name) const override;
virtual future<custom_options> query_custom_options(stdx::string_view role_name) const override;
virtual const resource_set& protected_resources() const override;
virtual ::shared_ptr<sasl_challenge> new_sasl_challenge() const override;
private:
future<> maybe_update_custom_options(std::string_view role_name, const authentication_options& options) const;
bool legacy_metadata_exists() const;
future<> migrate_legacy_metadata() const;

View File

@@ -24,9 +24,19 @@
#include "auth/authorizer.hh"
#include "auth/common.hh"
#include "auth/service.hh"
#include "db/config.hh"
namespace auth {
permissions_cache_config permissions_cache_config::from_db_config(const db::config& dc) {
permissions_cache_config c;
c.max_entries = dc.permissions_cache_max_entries();
c.validity_period = std::chrono::milliseconds(dc.permissions_validity_in_ms());
c.update_period = std::chrono::milliseconds(dc.permissions_update_interval_in_ms());
return c;
}
permissions_cache::permissions_cache(const permissions_cache_config& c, service& ser, logging::logger& log)
: _cache(c.max_entries, c.validity_period, c.update_period, log, [&ser, &log](const key_type& k) {
log.debug("Refreshing permissions for {}", k.first);

View File

@@ -22,7 +22,7 @@
#pragma once
#include <chrono>
#include <string_view>
#include <experimental/string_view>
#include <functional>
#include <iostream>
#include <optional>
@@ -37,6 +37,7 @@
#include "auth/resource.hh"
#include "auth/role_or_anonymous.hh"
#include "log.hh"
#include "stdx.hh"
#include "utils/hash.hh"
#include "utils/loading_cache.hh"
@@ -58,6 +59,8 @@ namespace auth {
class service;
struct permissions_cache_config final {
static permissions_cache_config from_db_config(const db::config&);
std::size_t max_entries;
std::chrono::milliseconds validity_period;
std::chrono::milliseconds update_period;

View File

@@ -61,7 +61,7 @@ std::ostream& operator<<(std::ostream& os, resource_kind kind) {
return os;
}
static const std::unordered_map<resource_kind, std::string_view> roots{
static const std::unordered_map<resource_kind, stdx::string_view> roots{
{resource_kind::data, "data"},
{resource_kind::role, "roles"}};
@@ -101,25 +101,24 @@ static permission_set applicable_permissions(const role_resource_view& rv) {
permission::DESCRIBE>();
}
resource::resource(resource_kind kind) : _kind(kind) {
_parts.emplace_back(roots.at(kind));
resource::resource(resource_kind kind) : _kind(kind), _parts{sstring(roots.at(kind))} {
}
resource::resource(resource_kind kind, utils::small_vector<sstring, 3> parts) : resource(kind) {
resource::resource(resource_kind kind, std::vector<sstring> parts) : resource(kind) {
_parts.reserve(parts.size() + 1);
_parts.insert(_parts.end(), std::make_move_iterator(parts.begin()), std::make_move_iterator(parts.end()));
}
resource::resource(data_resource_t, std::string_view keyspace) : resource(resource_kind::data) {
_parts.emplace_back(keyspace);
resource::resource(data_resource_t, stdx::string_view keyspace)
: resource(resource_kind::data, std::vector<sstring>{sstring(keyspace)}) {
}
resource::resource(data_resource_t, std::string_view keyspace, std::string_view table) : resource(resource_kind::data) {
_parts.emplace_back(keyspace);
_parts.emplace_back(table);
resource::resource(data_resource_t, stdx::string_view keyspace, stdx::string_view table)
: resource(resource_kind::data, std::vector<sstring>{sstring(keyspace), sstring(table)}) {
}
resource::resource(role_resource_t, std::string_view role) : resource(resource_kind::role) {
_parts.emplace_back(role);
resource::resource(role_resource_t, stdx::string_view role)
: resource(resource_kind::role, std::vector<sstring>{sstring(role)}) {
}
sstring resource::name() const {
@@ -174,7 +173,7 @@ data_resource_view::data_resource_view(const resource& r) : _resource(r) {
}
}
std::optional<std::string_view> data_resource_view::keyspace() const {
std::optional<stdx::string_view> data_resource_view::keyspace() const {
if (_resource._parts.size() == 1) {
return {};
}
@@ -182,7 +181,7 @@ std::optional<std::string_view> data_resource_view::keyspace() const {
return _resource._parts[1];
}
std::optional<std::string_view> data_resource_view::table() const {
std::optional<stdx::string_view> data_resource_view::table() const {
if (_resource._parts.size() <= 2) {
return {};
}
@@ -211,7 +210,7 @@ role_resource_view::role_resource_view(const resource& r) : _resource(r) {
}
}
std::optional<std::string_view> role_resource_view::role() const {
std::optional<stdx::string_view> role_resource_view::role() const {
if (_resource._parts.size() == 1) {
return {};
}
@@ -231,9 +230,9 @@ std::ostream& operator<<(std::ostream& os, const role_resource_view& v) {
return os;
}
resource parse_resource(std::string_view name) {
static const std::unordered_map<std::string_view, resource_kind> reverse_roots = [] {
std::unordered_map<std::string_view, resource_kind> result;
resource parse_resource(stdx::string_view name) {
static const std::unordered_map<stdx::string_view, resource_kind> reverse_roots = [] {
std::unordered_map<stdx::string_view, resource_kind> result;
for (const auto& pair : roots) {
result.emplace(pair.second, pair.first);
@@ -242,7 +241,7 @@ resource parse_resource(std::string_view name) {
return result;
}();
utils::small_vector<sstring, 3> parts;
std::vector<sstring> parts;
boost::split(parts, name, [](char ch) { return ch == '/'; });
if (parts.empty()) {

View File

@@ -41,7 +41,7 @@
#pragma once
#include <string_view>
#include <experimental/string_view>
#include <iostream>
#include <optional>
#include <stdexcept>
@@ -54,15 +54,15 @@
#include "auth/permission.hh"
#include "seastarx.hh"
#include "stdx.hh"
#include "utils/hash.hh"
#include "utils/small_vector.hh"
namespace auth {
class invalid_resource_name : public std::invalid_argument {
public:
explicit invalid_resource_name(std::string_view name)
: std::invalid_argument(format("The resource name '{}' is invalid.", name)) {
explicit invalid_resource_name(stdx::string_view name)
: std::invalid_argument(sprint("The resource name '%s' is invalid.", name)) {
}
};
@@ -98,16 +98,16 @@ struct role_resource_t final {};
class resource final {
resource_kind _kind;
utils::small_vector<sstring, 3> _parts;
std::vector<sstring> _parts;
public:
///
/// A root resource of a particular kind.
///
explicit resource(resource_kind);
resource(data_resource_t, std::string_view keyspace);
resource(data_resource_t, std::string_view keyspace, std::string_view table);
resource(role_resource_t, std::string_view role);
resource(data_resource_t, stdx::string_view keyspace);
resource(data_resource_t, stdx::string_view keyspace, stdx::string_view table);
resource(role_resource_t, stdx::string_view role);
resource_kind kind() const noexcept {
return _kind;
@@ -123,7 +123,7 @@ public:
permission_set applicable_permissions() const;
private:
resource(resource_kind, utils::small_vector<sstring, 3> parts);
resource(resource_kind, std::vector<sstring> parts);
friend class std::hash<resource>;
friend class data_resource_view;
@@ -131,7 +131,7 @@ private:
friend bool operator<(const resource&, const resource&);
friend bool operator==(const resource&, const resource&);
friend resource parse_resource(std::string_view);
friend resource parse_resource(stdx::string_view);
};
bool operator<(const resource&, const resource&);
@@ -150,7 +150,7 @@ class resource_kind_mismatch : public std::invalid_argument {
public:
explicit resource_kind_mismatch(resource_kind expected, resource_kind actual)
: std::invalid_argument(
format("This resource has kind '{}', but was expected to have kind '{}'.", actual, expected)) {
sprint("This resource has kind '%s', but was expected to have kind '%s'.", actual, expected)) {
}
};
@@ -166,9 +166,9 @@ public:
///
explicit data_resource_view(const resource& r);
std::optional<std::string_view> keyspace() const;
std::optional<stdx::string_view> keyspace() const;
std::optional<std::string_view> table() const;
std::optional<stdx::string_view> table() const;
};
std::ostream& operator<<(std::ostream&, const data_resource_view&);
@@ -187,7 +187,7 @@ public:
///
explicit role_resource_view(const resource&);
std::optional<std::string_view> role() const;
std::optional<stdx::string_view> role() const;
};
std::ostream& operator<<(std::ostream&, const role_resource_view&);
@@ -197,20 +197,20 @@ std::ostream& operator<<(std::ostream&, const role_resource_view&);
///
/// \throws \ref invalid_resource_name when the name is malformed.
///
resource parse_resource(std::string_view name);
resource parse_resource(stdx::string_view name);
const resource& root_data_resource();
inline resource make_data_resource(std::string_view keyspace) {
inline resource make_data_resource(stdx::string_view keyspace) {
return resource(data_resource_t{}, keyspace);
}
inline resource make_data_resource(std::string_view keyspace, std::string_view table) {
inline resource make_data_resource(stdx::string_view keyspace, stdx::string_view table) {
return resource(data_resource_t{}, keyspace, table);
}
const resource& root_role_resource();
inline resource make_role_resource(std::string_view role) {
inline resource make_role_resource(stdx::string_view role) {
return resource(role_resource_t{}, role);
}

View File

@@ -21,7 +21,7 @@
#pragma once
#include <string_view>
#include <experimental/string_view>
#include <memory>
#include <optional>
#include <stdexcept>
@@ -33,7 +33,7 @@
#include "auth/resource.hh"
#include "seastarx.hh"
#include "exceptions/exceptions.hh"
#include "stdx.hh"
namespace auth {
@@ -53,38 +53,38 @@ struct role_config_update final {
///
/// A logical argument error for a role-management operation.
///
class roles_argument_exception : public exceptions::invalid_request_exception {
class roles_argument_exception : public std::invalid_argument {
public:
using exceptions::invalid_request_exception::invalid_request_exception;
using std::invalid_argument::invalid_argument;
};
class role_already_exists : public roles_argument_exception {
public:
explicit role_already_exists(std::string_view role_name)
: roles_argument_exception(format("Role {} already exists.", role_name)) {
explicit role_already_exists(stdx::string_view role_name)
: roles_argument_exception(sprint("Role %s already exists.", role_name)) {
}
};
class nonexistant_role : public roles_argument_exception {
public:
explicit nonexistant_role(std::string_view role_name)
: roles_argument_exception(format("Role {} doesn't exist.", role_name)) {
explicit nonexistant_role(stdx::string_view role_name)
: roles_argument_exception(sprint("Role %s doesn't exist.", role_name)) {
}
};
class role_already_included : public roles_argument_exception {
public:
role_already_included(std::string_view grantee_name, std::string_view role_name)
role_already_included(stdx::string_view grantee_name, stdx::string_view role_name)
: roles_argument_exception(
format("{} already includes role {}.", grantee_name, role_name)) {
sprint("%s already includes role %s.", grantee_name, role_name)) {
}
};
class revoke_ungranted_role : public roles_argument_exception {
public:
revoke_ungranted_role(std::string_view revokee_name, std::string_view role_name)
revoke_ungranted_role(stdx::string_view revokee_name, stdx::string_view role_name)
: roles_argument_exception(
format("{} was not granted role {}, so it cannot be revoked.", revokee_name, role_name)) {
sprint("%s was not granted role %s, so it cannot be revoked.", revokee_name, role_name)) {
}
};
@@ -104,7 +104,7 @@ class role_manager {
public:
virtual ~role_manager() = default;
virtual std::string_view qualified_java_name() const noexcept = 0;
virtual stdx::string_view qualified_java_name() const noexcept = 0;
virtual const resource_set& protected_resources() const = 0;
@@ -115,17 +115,17 @@ public:
///
/// \returns an exceptional future with \ref role_already_exists for a role that has previously been created.
///
virtual future<> create(std::string_view role_name, const role_config&) const = 0;
virtual future<> create(stdx::string_view role_name, const role_config&) const = 0;
///
/// \returns an exceptional future with \ref nonexistant_role if the role does not exist.
///
virtual future<> drop(std::string_view role_name) const = 0;
virtual future<> drop(stdx::string_view role_name) const = 0;
///
/// \returns an exceptional future with \ref nonexistant_role if the role does not exist.
///
virtual future<> alter(std::string_view role_name, const role_config_update&) const = 0;
virtual future<> alter(stdx::string_view role_name, const role_config_update&) const = 0;
///
/// Grant `role_name` to `grantee_name`.
@@ -135,7 +135,7 @@ public:
/// \returns an exceptional future with \ref role_already_included if granting the role would be redundant, or
/// create a cycle.
///
virtual future<> grant(std::string_view grantee_name, std::string_view role_name) const = 0;
virtual future<> grant(stdx::string_view grantee_name, stdx::string_view role_name) const = 0;
///
/// Revoke `role_name` from `revokee_name`.
@@ -144,26 +144,26 @@ public:
///
/// \returns an exceptional future with \ref revoke_ungranted_role if the role was not granted.
///
virtual future<> revoke(std::string_view revokee_name, std::string_view role_name) const = 0;
virtual future<> revoke(stdx::string_view revokee_name, stdx::string_view role_name) const = 0;
///
/// \returns an exceptional future with \ref nonexistant_role if the role does not exist.
///
virtual future<role_set> query_granted(std::string_view grantee, recursive_role_query) const = 0;
virtual future<role_set> query_granted(stdx::string_view grantee, recursive_role_query) const = 0;
virtual future<role_set> query_all() const = 0;
virtual future<bool> exists(std::string_view role_name) const = 0;
virtual future<bool> exists(stdx::string_view role_name) const = 0;
///
/// \returns an exceptional future with \ref nonexistant_role if the role does not exist.
///
virtual future<bool> is_superuser(std::string_view role_name) const = 0;
virtual future<bool> is_superuser(stdx::string_view role_name) const = 0;
///
/// \returns an exceptional future with \ref nonexistant_role if the role does not exist.
///
virtual future<bool> can_login(std::string_view role_name) const = 0;
virtual future<bool> can_login(stdx::string_view role_name) const = 0;
};
}

View File

@@ -21,7 +21,7 @@
#pragma once
#include <string_view>
#include <experimental/string_view>
#include <functional>
#include <iosfwd>
#include <optional>
@@ -29,6 +29,7 @@
#include <seastar/core/sstring.hh>
#include "seastarx.hh"
#include "stdx.hh"
namespace auth {
@@ -37,7 +38,7 @@ public:
std::optional<sstring> name{};
role_or_anonymous() = default;
role_or_anonymous(std::string_view name) : name(name) {
role_or_anonymous(stdx::string_view name) : name(name) {
}
};

View File

@@ -36,23 +36,25 @@ namespace meta {
namespace roles_table {
std::string_view creation_query() {
stdx::string_view creation_query() {
static const sstring instance = sprint(
"CREATE TABLE %s ("
" %s text PRIMARY KEY,"
" can_login boolean,"
" is_superuser boolean,"
" member_of set<text>,"
" salted_hash text,"
" options frozen<map<text, text>>,"
" salted_hash text"
")",
qualified_name,
qualified_name(),
role_col_name);
return instance;
}
constexpr std::string_view qualified_name("system_auth.roles");
stdx::string_view qualified_name() noexcept {
static const sstring instance = AUTH_KS + "." + sstring(name);
return instance;
}
}
@@ -61,21 +63,23 @@ constexpr std::string_view qualified_name("system_auth.roles");
future<bool> default_role_row_satisfies(
cql3::query_processor& qp,
std::function<bool(const cql3::untyped_result_set_row&)> p) {
static const sstring query = format("SELECT * FROM {} WHERE {} = ?",
meta::roles_table::qualified_name,
static const sstring query = sprint(
"SELECT * FROM %s WHERE %s = ?",
meta::roles_table::qualified_name(),
meta::roles_table::role_col_name);
return do_with(std::move(p), [&qp](const auto& p) {
return qp.execute_internal(
return qp.process(
query,
db::consistency_level::ONE,
infinite_timeout_config,
{meta::DEFAULT_SUPERUSER_NAME},
true).then([&qp, &p](::shared_ptr<cql3::untyped_result_set> results) {
if (results->empty()) {
return qp.execute_internal(
return qp.process(
query,
db::consistency_level::QUORUM,
internal_distributed_query_state(),
internal_distributed_timeout_config(),
{meta::DEFAULT_SUPERUSER_NAME},
true).then([&p](::shared_ptr<cql3::untyped_result_set> results) {
if (results->empty()) {
@@ -94,13 +98,13 @@ future<bool> default_role_row_satisfies(
future<bool> any_nondefault_role_row_satisfies(
cql3::query_processor& qp,
std::function<bool(const cql3::untyped_result_set_row&)> p) {
static const sstring query = format("SELECT * FROM {}", meta::roles_table::qualified_name);
static const sstring query = sprint("SELECT * FROM %s", meta::roles_table::qualified_name());
return do_with(std::move(p), [&qp](const auto& p) {
return qp.execute_internal(
return qp.process(
query,
db::consistency_level::QUORUM,
internal_distributed_query_state()).then([&p](::shared_ptr<cql3::untyped_result_set> results) {
internal_distributed_timeout_config()).then([&p](::shared_ptr<cql3::untyped_result_set> results) {
if (results->empty()) {
return false;
}

View File

@@ -21,12 +21,13 @@
#pragma once
#include <string_view>
#include <experimental/string_view>
#include <functional>
#include <seastar/core/future.hh>
#include "seastarx.hh"
#include "stdx.hh"
namespace cql3 {
class query_processor;
@@ -39,13 +40,13 @@ namespace meta {
namespace roles_table {
std::string_view creation_query();
stdx::string_view creation_query();
constexpr std::string_view name{"roles", 5};
constexpr stdx::string_view name{"roles", 5};
extern const std::string_view qualified_name;
stdx::string_view qualified_name() noexcept;
constexpr std::string_view role_col_name{"role", 4};
constexpr stdx::string_view role_col_name{"role", 4};
}

View File

@@ -1,102 +0,0 @@
/*
* Licensed to the Apache Software Foundation (ASF) under one
* or more contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. The ASF licenses this file
* to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
/*
* Copyright (C) 2019 ScyllaDB
*
* Modified by ScyllaDB
*/
/*
* This file is part of Scylla.
*
* Scylla is free software: you can redistribute it and/or modify
* it under the terms of the GNU Affero General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* Scylla is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with Scylla. If not, see <http://www.gnu.org/licenses/>.
*/
#include "auth/sasl_challenge.hh"
#include "exceptions/exceptions.hh"
namespace auth {
/**
* SASL PLAIN mechanism specifies that credentials are encoded in a
* sequence of UTF-8 bytes, delimited by 0 (US-ASCII NUL).
* The form is : {code}authzId<NUL>authnId<NUL>password<NUL>{code}
* authzId is optional, and in fact we don't care about it here as we'll
* set the authzId to match the authnId (that is, there is no concept of
* a user being authorized to act on behalf of another).
*
* @param bytes encoded credentials string sent by the client
* @return map containing the username/password pairs in the form an IAuthenticator
* would expect
* @throws javax.security.sasl.SaslException
*/
bytes plain_sasl_challenge::evaluate_response(bytes_view client_response) {
sstring username, password;
auto b = client_response.crbegin();
auto e = client_response.crend();
auto i = b;
while (i != e) {
if (*i == 0) {
sstring tmp(i.base(), b.base());
if (password.empty()) {
password = std::move(tmp);
} else if (username.empty()) {
username = std::move(tmp);
}
b = ++i;
continue;
}
++i;
}
if (username.empty()) {
throw exceptions::authentication_exception("Authentication ID must not be null");
}
if (password.empty()) {
throw exceptions::authentication_exception("Password must not be null");
}
_username = std::move(username);
_password = std::move(password);
return {};
}
bool plain_sasl_challenge::is_complete() const {
return _username && _password;
}
future<authenticated_user> plain_sasl_challenge::get_authenticated_user() const {
return _when_complete(*_username, *_password);
}
}

Some files were not shown because too many files have changed in this diff Show More