Compare commits

..

149 Commits

Author SHA1 Message Date
Yaron Kaikov
6c0825e2a6 release: prepare for 4.6.11 2022-11-28 15:45:26 +02:00
Nadav Har'El
db3dd3bdf6 Merge 'cql3: don't ignore other restrictions when a multi column restriction is present during filtering' from Jan Ciołek
When filtering with multi column restriction present all other restrictions were ignored.
So a query like:
`SELECT * FROM WHERE pk = 0 AND (ck1, ck2) < (0, 0) AND regular_col = 0 ALLOW FILTERING;`
would ignore the restriction `regular_col = 0`.

This was caused by a bug in the filtering code:
2779a171fc/cql3/selection/selection.cc (L433-L449)

When multi column restrictions were detected, the code checked if they are satisfied and returned immediately.
This is fixed by returning only when these restrictions are not satisfied. When they are satisfied the other restrictions are checked as well to ensure all of them are satisfied.

This code was introduced back in 2019, when fixing #3574.
Perhaps back then it was impossible to mix multi column and regular columns and this approach was correct.

Fixes: #6200
Fixes: #12014

Closes #12031

* github.com:scylladb/scylladb:
  cql-pytest: add a reproducer for #12014, verify that filtering multi column and regular restrictions works
  boost/restrictions-test: uncomment part of the test that passes now
  cql-pytest: enable test for filtering combined multi column and regular column restrictions
  cql3: don't ignore other restrictions when a multi column restriction is present during filtering

(cherry picked from commit 2d2034ea28)

Closes #12086
2022-11-27 00:15:04 +02:00
Pavel Emelyanov
4ad24180f5 Merge '[branch-4.6] multishard_mutation_query: don't unpop partition header of spent partition ' from Botond Dénes
When stopping the read, the multishard reader will dismantle the
compaction state, pushing back (unpopping) the currently processed
partition's header to its originating reader. This ensures that if the
reader stops in the middle of a partition, on the next page the
partition-header is re-emitted as the compactor (and everything
downstream from it) expects.
It can happen however that there is nothing more for the current
partition in the reader and the next fragment is another partition.
Since we only push back the partition header (without a partition-end)
this can result in two partitions being emitted without being separated
by a partition end.
We could just add the missing partition-end when needed but it is
pointless, if the partition has no more data, just drop the header, we
won't need it on the next page.

The missing partition-end can generate an "IDL frame truncated" message
as it ends up causing the query result writer to create a corrupt
partition entry.

Fixes: https://github.com/scylladb/scylladb/issues/9482

Closes #11914

* github.com:scylladb/scylladb:
  test/cql-pytest: add regression test for "IDL frame truncated" error
  mutation_compactor: detach_state(): make it no-op if partition was exhausted
  treewide: fix headers
2022-11-16 11:52:51 +03:00
Anna Mikhlin
755c7eeb6a release: prepare for 4.6.10 2022-11-14 10:30:20 +02:00
Eliran Sinvani
8914ca8c58 cql: Fix crash upon use of the word empty for service level name
Wrong access to an uninitialized token instead of the actual
generated string caused the parser to crash, this wasn't
detected by the ANTLR3 compiler because all the temporary
variables defined in the ANTLR3 statements are global in the
generated code. This essentialy caused a null dereference.

Tests: 1. The fixed issue scenario from github.
       2. Unit tests in release mode.

Fixes #11774

Signed-off-by: Eliran Sinvani <eliransin@scylladb.com>
Message-Id: <20190612133151.20609-1-eliransin@scylladb.com>

Closes #11777

(cherry picked from commit ab7429b77d)
2022-11-10 20:43:44 +02:00
Botond Dénes
e82e4bbed3 test/cql-pytest: add regression test for "IDL frame truncated" error
(cherry picked from commit 11af489e84)
2022-11-07 16:51:14 +02:00
Botond Dénes
f9c457778e mutation_compactor: detach_state(): make it no-op if partition was exhausted
detach_state() allows the user to resume a compaction process later,
without having to keep the compactor object alive. This happens by
generating and returning the mutation fragments the user has to re-feed
to a newly constructed compactor to bring it into the exact same state
the current compactor was at the point of stopping the compaction.
This state includes the partition-header (partition-start and static-row
if any) and the currently active range tombstone.
Detaching the state is pointless however when the compaction was stopped
such that the currently compacted partition was completely exhausted.
Allowing the state to be detached in this case seems benign but it
caused a subtle bug in the main user of this feature: the partition
range scan algorithm, where the fragments included in the detached state
were pushed back into the reader which produced them. If the partition
happened to be exhausted -- meaning the next fragment in the reader was
a partition-start or EOS -- this resulted in the partition being
re-emitted later without a partition-end, resulting in corrupt
query-result being generated, in turn resulting in an obscure "IDL frame
truncated" error.

This patch solves this seemingly benign but sinister bug by making the
return value of `detach_state()` an std::optional and returning a
disengaged optional when the partition was exhausted.

(cherry picked from commit 70b4158ce0)
2022-11-07 16:51:14 +02:00
Botond Dénes
8315a7b164 treewide: fix headers
To fix CI.
2022-11-07 16:51:14 +02:00
Nadav Har'El
291ca8db60 cql3: fix cql3::util::maybe_quote() for keywords
cql3::util::maybe_quote() is a utility function formatting an identifier
name (table name, column name, etc.) that needs to be embedded in a CQL
statement - and might require quoting if it contains non-alphanumeric
characters, uppercase characters, or a CQL keyword.

maybe_quote() made an effort to only quote the identifier name if neccessary,
e.g., a lowercase name usually does not need quoting. But lowercase names
that are CQL keywords - e.g., to or where - cannot be used as identifiers
without quoting. This can cause problems for code that wants to generate
CQL statements, such as the materialized-view problem in issue #9450 - where
a user had a column called "to" and wanted to create a materialized view
for it.

So in this patch we fix maybe_quote() to recognize invalid identifiers by
using the CQL parser, and quote them. This will quote reserved keywords,
but not so-called unreserved keywords, which *are* allowed as identifiers
and don't need quoting. This addition slows down maybe_quote(), but
maybe_quote() is anyway only used in heavy operations which need to
generate CQL.

This patch also adds two tests that reproduce the bug and verify its
fix:

1. Add to the low-level maybe_quote() test (a C++ unit test) also tests
   that maybe_quote() quotes reserved keywords like "to", but doesn't
   quote unreserved keywords like "int".

2. Add a test reproducing issue #9450 - creating a materialized view
   whose key column is a keyword. This new test passes on Cassandra,
   failed on Scylla before this patch, and passes after this patch.

It is worth noting that maybe_quote() now has a "forward compatiblity"
problem: If we save CQL statements generated by maybe_quote(), and a
future version introduces a new reserved keyword, the parser of the
future version may not be able to parse the saved CQL statement that
was generated with the old mayb_quote() and didn't quote what is now
a keyword. This problem can be solved in two ways:

1. Try hard not to introduced new reserved keywords. Instead, introduce
   unreserved keywords. We've been doing this even before recognizing
   this maybe_quote() future-compatibility problem.

2. In the next patch we will introduce quote() - which unconditionally
   quotes identifier names, even if lowercase. These quoted names will
   be uglier for lowercase names - but will be safe from future
   introduction of new keywords. So we can consider switching some or
   all uses of maybe_quote() to quote().

Fixes #9450

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20220118161217.231811-1-nyh@scylladb.com>
(cherry picked from commit 5d2f694a90)
2022-11-07 10:38:10 +02:00
Jadw1
4da5fbaa24 CQL3: fromJson accepts string as bool
The problem was incompatibility with cassandra, which accepts bool
as a string in `fromJson()` UDF. The difference between Cassandra and
Scylla now is Scylla accepts whitespaces around word in string,
Cassandra don't. Both are case insensitive.

Fixes: #7915
(cherry picked from commit 1902dbc9ff)
2022-11-07 10:38:10 +02:00
Takuya ASADA
fc16664d81 locator::ec2_snitch: Retry HTTP request to EC2 instance metadata service
EC2 instance metadata service can be busy, ret's retry to connect with
interval, just like we do in scylla-machine-image.

Fixes #10250

Signed-off-by: Takuya ASADA <syuu@scylladb.com>

Closes #11688

(cherry picked from commit 6b246dc119)
(cherry picked from commit e2809674d2)
2022-11-06 15:43:58 +02:00
Botond Dénes
80bea5341e Merge 'Alternator, MV: fix bug in some view updates which set the view key to its existing value' from Nadav Har'El
As described in issue #11801, we saw in Alternator when a GSI has both partition and sort keys which were non-key attributes in the base, cases where updating the GSI-sort-key attribute to the same value it already had caused the entire GSI row to be deleted.

In this series fix this bug (it was a bug in our materialized views implementation) and add a reproducing test (plus a few more tests for similar situations which worked before the patch, and continue to work after it).

Fixes #11801

Closes #11808

* github.com:scylladb/scylladb:
  test/alternator: add test for issue 11801
  MV: fix handling of view update which reassign the same key value
  materialized views: inline used-once and confusing function, replace_entry()

(cherry picked from commit e981bd4f21)
2022-11-01 13:31:51 +02:00
Botond Dénes
6ecc772b56 mutation_partition: deletable_row::apply(shadowable_tombstone): remove redundant maybe_shadow()
Shadowing is already checked by the underlying row_tombstone::apply().
This redundant check was introduced by a previous fix to #9483
(6a76e12768). The rest of that patch is
good.

Refs: #9483
Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <20211115091513.181233-1-bdenes@scylladb.com>
(cherry picked from commit b136746040)
2022-10-16 11:53:04 +03:00
Benny Halevy
0b2e951954 range_tombstone_list: insert_from: correct rev.update range_tombstone in not overlapping case
2nd std::move(start) looks like a typo
in fe2fa3f20d.

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20220404124741.1775076-1-bhalevy@scylladb.com>
(cherry picked from commit 2d80057617)
2022-10-14 12:29:56 +02:00
Pavel Emelyanov
f2a738497f compaction_manager: Swallow ENOSPCs in ::stop()
When being stopped compaction manager may step on ENOSPC. This is not a
reason to fail stopping process with abort, better to warn this fact in
logs and proceed as if nothing happened

refs: #11245

Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
2022-10-13 16:02:33 +03:00
Pavel Emelyanov
badf7c816f exceptions: Mark storage_io_error::code() with noexcept
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
2022-10-13 16:02:32 +03:00
Pavel Emelyanov
bfb86f2c78 compaction_manager: Shuffle really_do_stop()
Make it the future-returning method and setup the _stop_future in its
only caller. Makes next patch much simpler

Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
2022-10-13 16:02:31 +03:00
Beni Peled
18e7a46038 release: prepare for 4.6.9 2022-10-09 08:54:33 +03:00
Nadav Har'El
cbcfa31e51 cql: validate bloom_filter_fp_chance up-front
Scylla's Bloom filter implementation has a minimal false-positive rate
that it can support (6.71e-5). When setting bloom_filter_fp_chance any
lower than that, the compute_bloom_spec() function, which writes the bloom
filter, throws an exception. However, this is too late - it only happens
while flushing the memtable to disk, and a failure at that point causes
Scylla to crash.

Instead, we should refuse the table creation with the unsupported
bloom_filter_fp_chance. This is also what Cassandra did six years ago -
see CASSANDRA-11920.

This patch also includes a regression test, which crashes Scylla before
this patch but passes after the patch (and also passes on Cassandra).

Fixes #11524.

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

Closes #11576

(cherry picked from commit 4c93a694b7)
2022-10-04 16:23:25 +03:00
Nadav Har'El
5ee69ff3a9 alternator: return ProvisionedThroughput in DescribeTable
DescribeTable is currently hard-coded to return PAY_PER_REQUEST billing
mode. Nevertheless, even in PAY_PER_REQUEST mode, the DescribeTable
operation must return a ProvisionedThroughput structure, listing both
ReadCapacityUnits and WriteCapacityUnits as 0. This requirement is not
stated in some DynamoDB documentation but is explictly mentioned in
https://docs.aws.amazon.com/amazondynamodb/latest/APIReference/API_ProvisionedThroughput.html
Also in empirically, DynamoDB returns ProvisionedThroughput with zeros
even in PAY_PER_REQUEST mode. We even had an xfailing test to confirm this.

The ProvisionedThroughput structure being missing was a problem for
applications like DynamoDB connectors for Spark, if they implicitly
assume that ProvisionedThroughput is returned by DescribeTable, and
fail (as described in issue #11222) if it's outright missing.

So this patch adds the missing ProvisionedThroughput structure, and
the xfailing test starts to pass.

Note that this patch doesn't change the fact that attempting to set
a table to PROVISIONED billing mode is ignored: DescribeTable continues
to always return PAY_PER_REQUEST as the billing mode and zero as the
provisioned capacities.

Fixes #11222

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

Closes #11298

(cherry picked from commit 941c719a23)
2022-10-03 14:29:22 +03:00
Tomasz Grabiec
949103d22a test: lib: random_mutation_generator: Don't generate mutations with marker uncompacted with shadowable tombstone
The generator was first setting the marker then applied tombstones.

The marker was set like this:

  row.marker() = random_row_marker();

Later, when shadowable tombstones were applied, they were compacted
with the marker as expected.

However, the key for the row was chosen randomly in each iteration and
there are multiple keys set, so there was a possibility of a key clash
with an earlier row. This could override the marker without applying
any tombstones, which is conditional on random choice.

This could generate rows with markers uncompacted with shadowable tombstones.

This broken row_cache_test::test_concurrent_reads_and_eviction on
comparison between expected and read mutations. The latter was
compacted because it went through an extra merge path, which compacts
the row.

Fix by making sure there are no key clashes.

Closes #11663

(cherry picked from commit 5268f0f837)
2022-10-03 09:00:28 +03:00
Botond Dénes
549cb60f4c sstables: crawling mx-reader: make on_out_of_clustering_range() no-op
Said method currently emits a partition-end. This method is only called
when the last fragment in the stream is a range tombstone change with a
position after all clustered rows. The problem is that
consume_partition_end() is also called unconditionally, resulting in two
partition-end fragments being emitted. The fix is simple: make this
method a no-op, there is nothing to do there.

Also add two tests: one targeted to this bug and another one testing the
crawling reader with random mutations generated for random schema.

Fixes: #11421

Closes #11422

(cherry picked from commit be9d1c4df4)
2022-09-30 17:56:58 +03:00
Botond Dénes
37633c5576 test/lib/random_schema: add a simpler overload for fixed partition count
Some tests want to generate a fixed amount of random partitions, make
their life easier.

(cherry picked from commit 98f3d516a2)

Ref #11421 (prerequisite)
2022-09-30 17:56:10 +03:00
Michael Livshin
abd9f43fa7 batchlog_manager: warn when a batch fails to replay
Only for reasons other than "no such KS", i.e. when the failure is
presumed transient and the batch in question is not deleted from
batchlog and will be retried in the future.

(Would info be more appropriate here than warning?)

Signed-off-by: Michael Livshin <michael.livshin@scylladb.com>

Closes #10556

Fixes #10636

(cherry picked from commit 00ed4ac74c)
2022-09-29 12:13:21 +03:00
Raphael S. Carvalho
d41d4db5c0 compaction: Make cleanup withstand better disk pressure scenario
It's not uncommong for cleanup to be issued against an entire keyspace,
which may be composed of tons of tables. To increase chances of success
if low on space, cleanup will now start from smaller tables first, such
that bigger tables will have more space available, once they're reached,
to satisfy their space requirement.

parallel_for_each() is dropped and wasn't needed given that manager
performs per-shard serialization of cleanup jobs.

Refs #9504.

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Message-Id: <20211130133712.64517-1-raphaelsc@scylladb.com>
(cherry picked from commit 0d5ac845e1)
2022-09-29 10:15:29 +03:00
Michał Radwański
c500043a78 flat_mutation_reader: allow destructing readers which are not closed and didn't initiate any IO.
In functions such as upgrade_to_v2 (excerpt below), if the constructor
of transforming_reader throws, r needs to be destroyed, however it
hasn't been closed. However, if a reader didn't start any operations, it
is safe to destruct such a reader. This issue can potentially manifest
itself in many more readers and might be hard to track down. This commit
adds a bool indicating whether a close is anticipated, thus avoiding
errors in the destructor.

Code excerpt:
flat_mutation_reader_v2 upgrade_to_v2(flat_mutation_reader r) {
    class transforming_reader : public flat_mutation_reader_v2::impl {
        // ...
    };
    return make_flat_mutation_reader_v2<transforming_reader>(std::move(r));
}

Fixes #9065.

(cherry picked from commit 9ada63a9cb)
2022-09-29 09:40:07 +03:00
Pavel Emelyanov
af4752a526 messaging_service: Fix gossiper verb group
When configuring tcp-nodelay unconditionally, messaging service thinks
gossiper uses group index 1, though it had changed some time ago and now
those verbs belong to group 0.

fixes: #11465

Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
(cherry picked from commit 2c74062962)
2022-09-19 10:32:49 +03:00
Anna Mikhlin
0aa9a8c266 release: prepare for 4.6.8 2022-09-19 09:30:09 +03:00
Michał Chojnowski
85fd6ab377 sstables: add a flag for disabling long-term index caching
Long-term index caching in the global cache, as introduced in 4.6, is a major
pessimization for workloads where accesses to the index are (spacially) sparse.
We want to have a way to disable it for the affected workloads.

There is already infrastructure in place for disabling it for BYPASS CACHE
queries. One way of solving the issue is hijacking that infrastructure.

This patch adds a global flag (and a corresponding CLI option) which controls
index caching. Setting the flag to `false` causes all index reads to behave
like they would in BYPASS CACHE queries.

Consequences of this choice:

- The per-SSTable partition_index_cache is unused. Every index_reader has
  its own, and they die together. Independent reads can no longer reuse the
  work of other reads which hit the same index pages. This is not crucial,
  since partition accesses have no (natural) spatial locality. Note that
  the original reason for partition_index_cache -- the ability to share
  reads for the lower and upper bound of the query -- is unaffected.
- The per-SSTable cached_file is unused. Every index_reader has its own
  (uncached) input stream from the index file, and every
  bsearch_clustered_cursor has its own cached_file, which dies together with
  the cursor. Note that the cursor still can perform its binary search with
  caching. However, it won't be able to reuse the file pages read by
  index_reader. In particular, if the promoted index is small, and fits inside
  the same file page as its index_entry, that page will be re-read.
  It can also happen that index_reader will read the same index file page
  multiple times. When the summary is so dense that multiple index pages fit in
  one index file page, advancing the upper bound, which reads the next index
  page, will read the same index file page. Since summary:disk ratio is 1:2000,
  this is expected to happen for partitions with size greater than 2000
  partition keys.

Fixes #11202

(cherry picked from commit cdb3e71045)
2022-09-18 13:30:28 +03:00
Beni Peled
7c79c513d1 release: prepare for 4.6.7 2022-09-07 11:17:55 +03:00
Karol Baryła
9a8e73f0c3 transport/server.cc: Return correct size of decompressed lz4 buffer
An incorrect size is returned from the function, which could lead to
crashes or undefined behavior. Fix by erroring out in these cases.

Fixes #11476

(cherry picked from commit 1c2eef384d)
2022-09-07 10:58:54 +03:00
Benny Halevy
fac0443200 snapshot-ctl: run_snapshot_modify_operation: reject views and secondary index using the schema
Detecting a secondary index by checking for a dot
in the table name is wrong as tables generated by Alternator
may contain a dot in their name.

Instead detect bot hmaterialized view and secondary indexes
using the schema()->is_view() method.

Fixes #10526

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
(cherry picked from commit aa127a2dbb)
2022-09-06 17:56:30 +03:00
Piotr Sarna
6bcfef2cfa cql3: fix misleading error message for service level timeouts
The error message incorrectly stated that the timeout value cannot
be longer than 24h, but it can - the actual restriction is that the
value cannot be expressed in units like days or months, which was done
in order to significantly simplify the parsing routines (and the fact
that timeouts counted in days are not expected to be common).

Fixes #10286

Closes #10294

(cherry picked from commit 85e95a8cc3)
2022-09-01 20:34:22 +03:00
Juliusz Stasiewicz
d2c67a2429 cdc/check_and_repair_cdc_streams: ignore LEFT endpoints
When `check_and_repair_cdc_streams` encountered a node with status LEFT, Scylla
would throw. This behavior is fixed so that LEFT nodes are simply ignored.

Fixes #9771

Closes #9778

(cherry picked from commit 351f142791)
2022-09-01 15:44:35 +03:00
Avi Kivity
d6c2f228e7 Merge 'row_cache: Fix missing row if upper bound of population range is evicted and has adjacent dummy' from Tomasz Grabiec
Scenario:

cache = [
    row(pos=2, continuous=false),
    row(pos=after(2), dummy=true)
]

Scanning read starts, starts populating [-inf, before(2)] from sstables.

row(pos=2) is evicted.

cache = [
    row(pos=after(2), dummy=true)
]

Scanning read finishes reading from sstables.

Refreshes cache cursor via
partition_snapshot_row_cursor::maybe_refresh(), which calls
partition_snapshot_row_cursor::advance_to() because iterators are
invalidated. This advances the cursor to
after(2). no_clustering_row_between(2, after(2)) returns true, so
advance_to() returns true, and maybe_refresh() returns true. This is
interpreted by the cache reader as "the cursor has not moved forward",
so it marks the range as complete, without emitting the row with
pos=2. Also, it marks row(pos=after(2)) as continuous, so later reads
will also miss the row.

The bug is in advance_to(), which is using
no_clustering_row_between(a, b) to determine its result, which by
definition excludes the starting key.

Discovered by row_cache_test.cc::test_concurrent_reads_and_eviction
with reduced key range in the random_mutation_generator (1024 -> 16).

Fixes #11239

Closes #11240

* github.com:scylladb/scylladb:
  test: mvcc: Fix illegal use of maybe_refresh()
  tests: row_cache_test: Add test_eviction_of_upper_bound_of_population_range()
  tests: row_cache_test: Introduce one_shot mode to throttle
  row_cache: Fix missing row if upper bound of population range is evicted and has adjacent dummy
2022-08-11 19:19:30 +02:00
Yaron Kaikov
a1b1df2074 release: prepare for 4.6.6 2022-08-07 16:24:51 +03:00
Avi Kivity
14e13ecbd4 Merge 'Backport: Fix map subscript crashes when map or subscript is null' from Nadav Har'El
This is a backport of https://github.com/scylladb/scylla/pull/10420 to branch 5.0.
Branch 5.0 had somewhat different code in this expression area, so the backport was not automatically, but nevertheless was fairly straightforward - just copy the exact same checking code to its right place, and keep the exact same tests to see we indeed fixed the bug.

Refs #10535.

The original cover letter from https://github.com/scylladb/scylla/pull/10420:

In the filtering expression "WHERE m[?] = 2", our implementation was buggy when either the map, or the subscript, was NULL (and also when the latter was an UNSET_VALUE). Our code ended up dereferencing null objects, yielding bizarre errors when we were lucky, or crashes when we were less lucky - see examples of both in issues https://github.com/scylladb/scylla/issues/10361, https://github.com/scylladb/scylla/issues/10399, https://github.com/scylladb/scylla/pull/10401. The existing test test_null.py::test_map_subscript_null reproduced all these bugs sporadically.

In this series we improve the test to reproduce the separate bugs separately, and also reproduce additional problems (like the UNSET_VALUE). We then define both m[NULL] and NULL[2] to result in NULL instead of the existing undefined (and buggy, and crashing) behavior. This new definition is consistent with our usual SQL-inspired tradition that NULL "wins" in expressions - e.g., NULL < 2 is also defined as resulting in NULL.

However, this decision differs from Cassandra, where m[NULL] is considered an error but NULL[2] is allowed. We believe that making m[NULL] be a NULL instead of an error is more consistent, and moreover - necessary if we ever want to support more complicate expressions like m[a], where the column a can be NULL for some rows and non-NULL for others, and it doesn't make sense to return an "invalid query" error in the middle of the scan.

Fixes https://github.com/scylladb/scylla/issues/10361
Fixes https://github.com/scylladb/scylla/issues/10399
Fixes https://github.com/scylladb/scylla/pull/10401

Closes #11142

* github.com:scylladb/scylla:
  test/cql-pytest: reproducer for CONTAINS NULL bug
  expressions: don't dereference invalid map subscript in filter
  expressions: fix invalid dereference in map subscript evaluation
  test/cql-pytest: improve tests for map subscripts and nulls

(cherry picked from commit 23a34d7e42)
2022-07-31 15:44:00 +03:00
Benny Halevy
b8740bde6e multishard_mutation_query: do_query: stop ctx if lookup_readers fails
lookup_readers might fail after populating some readers
and those better be closed before returning the exception.

Fixes #10351

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>

Closes #10425

(cherry picked from commit 055141fc2e)
2022-07-25 14:52:58 +03:00
Benny Halevy
1b23f8d038 sstables: time_series_sstable_set: insert: make exception safe
Need to erase the shared sstable from _sstables
if insertion to _sstables_reversed fails.

Fixes #10787

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
(cherry picked from commit cd68b04fbf)
2022-07-25 14:22:08 +03:00
Tomasz Grabiec
05a228e4c5 memtable: Fix missing range tombstones during reads under ceratin rare conditions
There is a bug introduced in e74c3c8 (4.6.0) which makes memtable
reader skip one a range tombstone for a certain pattern of deletions
and under certain sequence of events.

_rt_stream contains the result of deoverlapping range tombstones which
had the same position, which were sipped from all the versions. The
result of deoverlapping may produce a range tombstone which starts
later, at the same position as a more recent tombstone which has not
been sipped from the partition version yet. If we consume the old
range tombstone from _rt_stream and then refresh the iterators, the
refresh will skip over the newer tombstone.

The fix is to drop the logic which drains _rt_stream so that
_rt_stream is always merged with partition versions.

For the problem to trigger, there have to be multiple MVCC versions
(at least 2) which contain deletions of the following form:

[a, c] @ t0
[a, b) @ t1, [b, d] @ t2

c > b

The proper sequence for such versions is (assuming d > c):

[a, b) @ t1,
[b, d] @ t2

Due to the bug, the reader will produce:

[a, b) @ t1,
[b, c] @ t0

The reader also needs to be preempted right before processing [b, d] @
t2 and iterators need to get invalidated so that
lsa_partition_reader::do_refresh_state() is called and it skips over
[b, d] @ t2. Otherwise, the reader will emit [b, d] @ t2 later. If it
does emit the proper range tombstone, it's possible that it will violate
fragment order in the stream if _rt_stream accumulated remainders
(possible with 3 MVCC versions).

The problem goes away once MVCC versions merge.

Fixes #10913
Fixes #10830

Closes #10914

(cherry picked from commit a6aef60b93)

[avi: backport prerequisite position_range_to_clustering_range() too]
2022-07-19 19:27:15 +03:00
Yaron Kaikov
2ec293ab0e release: prepare for 4.6.5 2022-07-19 16:02:46 +03:00
Pavel Emelyanov
b60f14601e azure_snitch: Do nothing on non-io-cpu
All snitch drivers are supposed to snitch info on some shard and
replicate the dc/rack info across others. All, but azure really do so.
The azure one gets dc/rack on all shards, which's excessive but not
terrible, but when all shards start to replicate their data to all the
others, this may lead to use-after-frees.

fixes: #10494

Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
(cherry picked from commit c6d0bc87d0)
2022-07-17 14:22:29 +03:00
Raphael S. Carvalho
284dd21ef7 compaction_manager: Fix race when selecting sstables for rewrite operations
Rewrite operations are scrub, cleanup and upgrade.

Race can happen because 'selection of sstables' and 'mark sstables as
compacting' are decoupled. So any deferring point in between can lead
to a parallel compaction picking the same files. After commit 2cf0c4bbf,
files are marked as compacting before rewrite starts, but it didn't
take into account the commit c84217ad which moved retrieval of
candidates to a deferring thread, before rewrite_sstables() is even
called.

Scrub isn't affected by this because it uses a coarse grained approach
where whole operation is run with compaction disabled, which isn't good
because regular compaction cannot run until its completion.

From now on, selection of files and marking them as compacting will
be serialized by running them with compaction disabled.

Now cleanup will also retrieve sstables with compaction disabled,
meaning it will no longer leave uncleaned files behind, which is
important to avoid data resurrection if node regains ownership of
data in uncleaned files.

Fixes #8168.
Refs #8155.

[backport notes:
- minor conflict around run_with_compaction_disabled()
- bumped into our old friend
  https://gcc.gnu.org/bugzilla/show_bug.cgi?id=95111,
so I had to use std::ref() on local copy of lambda
- with the yielding part of candidate retrieval now happening in
rewrite_sstables(), task registration is moved to after run_with_
compaction_disabled() call, so the latter won't incorrectly try
to stop the task that called it, which triggers an assert in
debug mode.
]

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Message-Id: <20211129133107.53011-1-raphaelsc@scylladb.com>
(cherry picked from commit 80a1ebf0f3)

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>

Closes #10963
2022-07-13 18:45:36 +03:00
Pavel Emelyanov
8b52f1d6e7 view: Fix trace-state pointer use after move
It's moved into .mutate_locally() but it captured and used in its
continuation. It works well just because moved-from pointer looks like
nullptr and all the tracing code checks for it to be non-such.

tests: https://jenkins.scylladb.com/job/releng/job/Scylla-CI/1266/
       (CI job failed on post-actions thus it's red)

Fixes #11015

Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Message-Id: <20220711134152.30346-1-xemul@scylladb.com>
(cherry picked from commit 5526738794)
2022-07-12 14:21:11 +03:00
Piotr Sarna
157951f756 view: exclude using static columns in the view filter
The code which applied view filtering (i.e. a condition placed
on a view column, e.g. "WHERE v = 42") erroneously used a wildcard
selection, which also assumes that static columns are needed,
if the base table contains any such columns.
The filtering code currently assumes that no such columns are fetched,
so the selection is amended to only ask for regular columns
(primary key columns are sent anyway, because they are enabled
via slice options, so no need to ask for them explicitly).

Fixes #10851

Closes #10855

(cherry picked from commit bc3a635c42)
2022-07-11 17:07:22 +03:00
Juliusz Stasiewicz
4f643ed4a5 cdc: check_and_repair_cdc_streams: regenerate if too many streams are present
If the number of streams exceeds the number of token ranges
it indicates that some spurious streams from decommissioned
nodes are present.

In such a situation - simply regenerate.

Fixes #9772

Closes #9780

(cherry picked from commit ea46439858)
2022-07-07 18:53:14 +02:00
Avi Kivity
b598629b7f messaging: do isolate default tenants
In 10dd08c9 ("messaging_service: supply and interpret rpc isolation_cookies",
4.2), we added a mechanism to perform rpc calls in remote scheduling groups
based on the connection identity (rather than the verb), so that
connection processing itself can run in the correct group (not just
verb processing), and so that one verb can run in different groups according
to need.

In 16d8cdadc ("messaging_service: introduce the tenant concept", 4.2), we
changed the way isolation cookies are sent:

 scheduling_group
 messaging_service::scheduling_group_for_verb(messaging_verb verb) const {
     return _scheduling_info_for_connection_index[get_rpc_client_idx(verb)].sched_group;
@@ -665,11 +694,14 @@ shared_ptr<messaging_service::rpc_protocol_client_wrapper> messaging_service::ge
     if (must_compress) {
         opts.compressor_factory = &compressor_factory;
     }
     opts.tcp_nodelay = must_tcp_nodelay;
     opts.reuseaddr = true;
-    opts.isolation_cookie = _scheduling_info_for_connection_index[idx].isolation_cookie;
+    // We send cookies only for non-default statement tenant clients.
+    if (idx > 3) {
+        opts.isolation_cookie = _scheduling_info_for_connection_index[idx].isolation_cookie;
+    }

This effectively disables the mechanism for the default tenant. As a
result some verbs will be executed in whatever group the messaging
service listener was started in. This used to be the main group,
but in 554ab03 ("main: Run init_server and join_cluster inside
maintenance scheduling group", 4.5), this was change to the maintenance
group. As a result normal read/writes now compete with maintenance
operations, raising their latency significantly.

Fix by sending the isolation cookie for all connections. With this,
a 2-node cassandra-stress load has 99th percentile increase by just
3ms during repair, compared to 10ms+ before.

Fixes #9505.

Closes #10673

(cherry picked from commit c83393e819)
2022-07-05 13:42:10 +03:00
Nadav Har'El
43f82047b9 Merge 'types: fix is_string for reversed types' from Piotr Sarna
Checking if the type is string is subtly broken for reversed types,
and these types will not be recognized as strings, even though they are.
As a result, if somebody creates a column with DESC order and then
tries to use operator LIKE on it, it will fail because the type
would not be recognized as a string.

Fixes #10183

Closes #10181

* github.com:scylladb/scylla:
  test: add a case for LIKE operator on a descending order column
  types: fix is_string for reversed types

(cherry picked from commit 733672fc54)
2022-07-03 17:59:56 +03:00
Benny Halevy
ec3c07de6e compaction_manager: perform_offstrategy: run_offstrategy_compaction in maintenance scheduling group
It was assumed that offstrategy compaction is always triggered by streaming/repair
where it would inherit the caller's scheduling group.

However, offstrategy is triggered by a timer via table::_off_strategy_trigger so I don't see
how the expiration of this timer will inherit anything from streaming/repair.

Also, since d309a86, offstrategy compaction
may be triggered by the api where it will run in the default scheduling group.

The bottom line is that the compaction manager needs to explicitly perform offstrategy compaction
in the maintenance scheduling group similar to `perform_sstable_scrub_validate_mode`.

Fixes #10151

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20220302084821.2239706-1-bhalevy@scylladb.com>
(cherry picked from commit 0764e511bb)
2022-07-03 14:30:54 +03:00
Takuya ASADA
82572e8cfe scylla_coredump_setup: support new format of Storage field
Storage field of "coredumpctl info" changed at systemd-v248, it added
"(present)" on the end of line when coredump file available.

Fixes #10669

Closes #10714

(cherry picked from commit ad2344a864)
2022-07-03 13:55:25 +03:00
Nadav Har'El
2b9ed79c6f alternator: forbid empty AttributesToGet
In DynamoDB one can retrieve only a subset of the attributes using the
AttributesToGet or ProjectionExpression paramters to read requests.
Neither allows an empty list of attributes - if you don't want any
attributes, you should use Select=COUNT instead.

Currently we correctly refuse an empty ProjectionExpression - and have
a test for it:
test_projection_expression.py::test_projection_expression_toplevel_syntax

However, Alternator is missing the same empty-forbidding logic for
AttributesToGet. An empty AttributesToGet is currently allowed, and
basically says "retrieve everything", which is sort of unexpected.

So this patch adds the missing logic, and the missing test (actually
two tests for the same thing - one using GetItem and the other Query).

Fixes #10332

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20220405113700.9768-1-nyh@scylladb.com>
(cherry picked from commit 9c1ebdceea)
2022-07-03 13:36:02 +03:00
Avi Kivity
ab0b6fd372 Update seastar submodule (json crash in describe_ring)
* seastar 7a430a0830...8b2c13b346 (1):
  > Merge 'stream_range_as_array: always close output stream' from Benny Halevy

Fixes #10592.
2022-06-08 16:49:53 +03:00
Nadav Har'El
12f1718ef4 alternator: allow DescribeTimeToLive even without TTL enabled
We still consider the TTL support in Alternator to be experimental, so we
don't want to allow a user to enable TTL on a table without turning on a
"--experimental-features" flag. However, there is no reason not to allow
the DescribeTimeToLive call when this experimental flag is off - this call
would simply reply with the truth - that the TTL feature is disabled for
the table!

This is important for client code (such as the Terraform module
described in issue #10660) which uses DescribeTimeToLive for
information, even when it never intends to actually enable TTL.

The patch is trivial - we simply remove the flag check in
DescribeTimeToLive, the code works just as before.

After this patch, the following test now works on Scylla without
experimental flags turned on:

    test/alternator/run test_ttl.py::test_describe_ttl_without_ttl

Refs #10660

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
(cherry picked from commit 8ecf1e306f)
2022-05-30 20:40:34 +03:00
Tomasz Grabiec
322dfe8403 sstable: partition_index_cache: Fix abort on bad_alloc during page loading
When entry loading fails and there is another request blocked on the
same page, attempt to erase the failed entry will abort because that
would violate entry_ptr guarantees, which is supposed to keep the
entry alive.

The fix in 92727ac36c was incomplete. It
only helped for the case of a single loader. This patch makes a more
general approach by relaxing the assert.

The assert manifested like this:

scylla: ./sstables/partition_index_cache.hh:71: sstables::partition_index_cache::entry::~entry(): Assertion `!is_referenced()' failed.

Fixes #10617

Closes #10653

(cherry picked from commit f87274f66a)
2022-05-30 13:00:46 +03:00
Beni Peled
11f008e8fd release: prepare for 4.6.4 2022-05-16 15:20:35 +03:00
Benny Halevy
fd7314a362 table: clear: serialize with ongoing flush
Get all flush permits to serialize with any
ongoing flushes and preventing further flushes
during table::clear, in particular calling
discard_completed_segments for every table and
clearing the memtables in clear_and_add.

Fixes #10423

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
(cherry picked from commit aae532a96b)
2022-05-15 13:43:43 +03:00
Raphael S. Carvalho
d27468f078 compaction: LCS: don't write to disengaged optional on compaction completion
Dtest triggers the problem by:
1) creating table with LCS
2) disabling regular compaction
3) writing a few sstables
4) running maintenance compaction, e.g. cleanup

Once the maintenance compaction completes, disengaged optional _last_compacted_keys
triggers an exception in notify_completion().

_last_compacted_keys is used by regular for its round-robin file picking
policy. It stores the last compacted key for each level. Meaning it's
irrelevant for any other compaction type.

Regular compaction is responsible for initializing it when it runs for
the first time to pick files. But with it disabled, notify_completion()
will find it uninitialized, therefore resulting in bad_optional_access.

To fix this, the procedure is skipped if _last_compacted_keys is
disengaged. Regular compaction, once re-enabled, will be able to
fill _last_compacted_keys by looking at metadata of the files.

compaction_test.py::TestCompaction::test_disable_autocompaction_doesnt_
block_user_initiated_compactions[CLEANUP-LeveledCompactionStrategy]
now passes.

Fixes #10378.

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>

Closes #10508

(cherry picked from commit 8e99d3912e)
2022-05-15 13:20:30 +03:00
Juliusz Stasiewicz
74ef1ee961 CQL: Replace assert by exception on invalid auth opcode
One user observed this assertion fail, but it's an extremely rare event.
The root cause - interlacing of processing STARTUP and OPTIONS messages -
is still there, but now it's harmless enough to leave it as is.

Fixes #10487

Closes #10503

(cherry picked from commit 603dd72f9e)
2022-05-10 14:03:03 +02:00
Benny Halevy
07549d159c compaction: time_window_compaction_strategy: reset estimated_remaining_tasks when running out of candidates
_estimated_remaining_tasks gets updated via get_next_non_expired_sstables ->
get_compaction_candidates, but otherwise if we return earlier from
get_sstables_for_compaction, it does not get updated and may go out of sync.

Refs #10418
(to be closed when the fix reaches branch-4.6)

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>

Closes #10419

(cherry picked from commit 01f41630a5)
2022-05-09 09:36:22 +03:00
Eliran Sinvani
189bbcd82d prepared_statements: Invalidate batch statement too
It seams that batch prepared statements always return false for
depends_on, this in turn renders the removal criteria from the
prepared statements cache to always be false which result by the
queries not being evicted.
Here we change the function to return the true state meaning,
they will return true if one of the sub queries is dependant
upon the keyspace and/ or column family.

Fixes #10129

Signed-off-by: Eliran Sinvani <eliransin@scylladb.com>
(cherry picked from commit 4eb0398457)
2022-05-08 12:33:00 +03:00
Eliran Sinvani
70e6921125 cql3 statements: Change dependency test API to express better it's
purpose

Cql statements used to have two API functions, depends_on_keyspace and
depends_on_column_family. The former, took as a parameter only a table
name, which makes no sense. There could be multiple tables with the same
name each in a different keyspace and it doesn't make sense to
generalize the test - i.e to ask "Does a statement depend on any table
named XXX?"
In this change we unify the two calls to one - depends on that takes a
keyspace name and optionally also a table name, that way every logical
dependency tests that makes sense is supported by a single API call.

(cherry picked from commit bf50dbd35b)

Ref #10129
2022-05-08 12:32:41 +03:00
Calle Wilund
e314158708 cdc: Ensure columns removed from log table are registered as dropped
If we are redefining the log table, we need to ensure any dropped
columns are registered in "dropped_columns" table, otherwise clients will not
be able to read data older than now.
Includes unit test.

Should probably be backported to all CDC enabled versions.

Fixes #10473
Closes #10474

(cherry picked from commit 78350a7e1b)
2022-05-05 11:34:56 +02:00
Tomasz Grabiec
46586532c9 loading_cache: Make invalidation take immediate effect
There are two issues with current implementation of remove/remove_if:

  1) If it happens concurrently with get_ptr(), the latter may still
  populate the cache using value obtained from before remove() was
  called. remove() is used to invalidate caches, e.g. the prepared
  statements cache, and the expected semantic is that values
  calculated from before remove() should not be present in the cache
  after invalidation.

  2) As long as there is any active pointer to the cached value
  (obtained by get_ptr()), the old value from before remove() will be
  still accessible and returned by get_ptr(). This can make remove()
  have no effect indefinitely if there is persistent use of the cache.

One of the user-perceived effects of this bug is that some prepared
statements may not get invalidated after a schema change and still use
the old schema (until next invalidation). If the schema change was
modifying UDT, this can cause statement execution failures. CQL
coordinator will try to interpret bound values using old set of
fields. If the driver uses the new schema, the coordinaotr will fail
to process the value with the following exception:

  User Defined Type value contained too many fields (expected 5, got 6)

The patch fixes the problem by making remove()/remove_if() erase old
entries from _loading_values immediately.

The predicate-based remove_if() variant has to also invalidate values
which are concurrently loading to be safe. The predicate cannot be
avaluated on values which are not ready. This may invalidate some
values unnecessarily, but I think it's fine.

Fixes #10117

Message-Id: <20220309135902.261734-1-tgrabiec@scylladb.com>
(cherry picked from commit 8fa704972f)
2022-05-04 15:38:11 +03:00
Avi Kivity
0114244363 Merge 'replica/database: drop_column_family(): properly cleanup stale querier cache entries' from Botond Dénes
Said method has to evict all querier cache entries, belonging to the to-be-dropped table. This is already the case, but there was a window where new entries could sneak in, causing a stale reference to the table to be de-referenced later when they are evicted due to TTL. This window is now closed, the entries are evicted after the method has waited for all ongoing operations on said table to stop.

Fixes: #10450

Closes #10451

* github.com:scylladb/scylla:
  replica/database: drop_column_family(): drop querier cache entries after waiting for ops
  replica/database: finish coroutinizing drop_column_family()
  replica/database: make remove(const column_family&) private

(cherry picked from commit 7f1e368e92)
2022-05-01 17:11:52 +03:00
Avi Kivity
f154c8b719 Update tools/java submodule (bad IPv6 addresses in nodetool)
* tools/java 05ec511bbb...46744a92ff (1):
  > CASSANDRA-17581 fix NodeProbe: Malformed IPv6 address at index

Fixes #10442
2022-04-28 11:35:09 +03:00
Beni Peled
8bf149fdd6 release: prepare for 4.6.3 2022-04-14 14:16:52 +03:00
Tomasz Grabiec
0265d56173 utils/chunked_managed_vector: Fix sigsegv during reserve()
Fixes the case of make_room() invoked with last_chunk_capacity_deficit
but _size not in the last reserved chunk.

Found during code review, no user impact.

Fixes #10364.

Message-Id: <20220411224741.644113-1-tgrabiec@scylladb.com>
(cherry picked from commit 0c365818c3)
2022-04-13 10:29:30 +03:00
Tomasz Grabiec
e50452ba43 utils/chunked_vector: Fix sigsegv during reserve()
Fixes the case of make_room() invoked with last_chunk_capacity_deficit
but _size not in the last reserved chunk.

Found during code review, no known user impact.

Fixes #10363.

Message-Id: <20220411222605.641614-1-tgrabiec@scylladb.com>
(cherry picked from commit 01eeb33c6e)

[avi: make max_chunk_capacity() public for backport]
2022-04-13 10:29:03 +03:00
Avi Kivity
a205f644cb transport: return correct error codes when downgrading v4 {WRITE,READ}_FAILURE to {WRITE,READ}_TIMEOUT
Protocol v4 added WRITE_FAILURE and READ_FAILURE. When running under v3
we downgrade these exceptions to WRITE_TIMEOUT and READ_TIMEOUT (since
the client won't understand the v4 errors), but we still send the new
error codes. This causes the client to become confused.

Fix by updating the error codes.

A better fix is to move the error code from the constructor parameter
list and hard-code it in the constructor, but that is left for a follow-up
after this minimal fix.

Fixes #5610.

Closes #10362

(cherry picked from commit 987e6533d2)
2022-04-13 09:49:02 +03:00
Tomasz Grabiec
f136b5b950 utils/chunked_managed_vector: Fix corruption in case there is more than one chunk
If reserve() allocates more than one chunk, push_back() should not
work with the last chunk. This can result in items being pushed to the
wrong chunk, breaking internal invariants.

Also, pop_back() should not work with the last chunk. This breaks when
there is more than one chunk.

Currently, the container is only used in the sstable partition index
cache.

Manifests by crashes in sstable reader which touch sstables which have
partition index pages with more than 1638 partition entries.

Introduced in 78e5b9fd85 (4.6.0)

Fixes #10290

Message-Id: <20220407174023.527059-1-tgrabiec@scylladb.com>
(cherry picked from commit 41fe01ecff)
2022-04-08 10:53:52 +03:00
Takuya ASADA
69a1325884 docker: enable --log-to-stdout which mistakenly disabled
Since our Docker image moved to Ubuntu, we mistakenly copy
dist/docker/etc/sysconfig/scylla-server to /etc/sysconfig, which is not
used in Ubuntu (it should be /etc/default).
So /etc/default/scylla-server is just default configuration of
scylla-server .deb package, --log-to-stdout is 0, same as normal installation.

We don't want keep the duplicated configuration file anyway,
so let's drop dist/docker/etc/sysconfig/scylla-server and configure
/etc/default/scylla-server in build_docker.sh.

Fixes #10270

Closes #10280

(cherry picked from commit bdefea7c82)
2022-04-07 12:13:35 +03:00
Avi Kivity
ab153c9b94 Update seastar submodule (logger deadlock with large messages)
* seastar 34e58f9995...94a462d94b (2):
  > log: Fix silencer to be shard-local and logger-global
  > log: Silence logger when logging

Fixes #10336.
2022-04-05 19:43:49 +03:00
Beni Peled
eb372d7f03 release: prepare for 4.6.2 2022-04-05 16:59:53 +03:00
Takuya ASADA
e232711e7e docker: run scylla as root
Previous versions of Docker image runs scylla as root, but cb19048
accidently modified it to scylla user.
To keep compatibility we need to revert this to root.

Fixes #10261

Closes #10325

(cherry picked from commit f95a531407)
2022-04-05 12:46:12 +03:00
Takuya ASADA
0a440b6d4a docker: revert scylla-server.conf service name change
We changed supervisor service name at cb19048, but this breaks
compatibility with scylla-operator.
To fix the issue we need to revert the service name to previous one.

Fixes #10269

Closes #10323

(cherry picked from commit 41edc045d9)
2022-04-05 12:42:36 +03:00
Piotr Sarna
00bb1e8145 cql3: fix qualifying restrictions with IN for indexing
When a query contains IN restriction on its partition key,
it's currently not eligible for indexing. It was however
erroneously qualified as such, which lead to fetching incorrect
results. This commit fixes the issue by not allowing such queries
to undergo indexing, and comes with a regression test.

Fixes #10300

Closes #10302

(cherry picked from commit c0fd53a9d7)
2022-04-03 11:21:43 +03:00
Avi Kivity
e30dbee2db Update seastar submodule (pidof command not installed)
* seastar 50e1549b2c...34e58f9995 (1):
  > seastar-cpu-map.sh: switch from pidof to pgrep
Fixes #10238.
2022-03-29 12:40:17 +03:00
Beni Peled
2309d6b51e release: prepare for 4.6.1 2022-03-28 10:57:31 +03:00
Benny Halevy
b77ca07709 atomic_cell: compare_atomic_cell_for_merge: compare ttl if expiry is equal
Following up on a57c087c89,
compare_atomic_cell_for_merge should compare the ttl value in the
reverse order since, when comparing two cells that are identical
in all attributes but their ttl, we want to keep the cell with the
smaller ttl value rather than the larger ttl, since it was written
at a later (wall-clock) time, and so would remain longer after it
expires, until purged after gc_grace seconds.

Fixes #10173

Test: mutation_test.test_cell_ordering, unit(dev)

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20220302154328.2400717-1-bhalevy@scylladb.com>
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20220306091913.106508-1-bhalevy@scylladb.com>
(cherry picked from commit a085ef74ff)
2022-03-24 18:08:07 +02:00
Benny Halevy
bb0a38f889 atomic_cell: compare_atomic_cell_for_merge: compare ttl if expiry is equal
Unlike atomic_cell_or_collection::equals, compare_atomic_cell_for_merge
currently returns std::strong_ordering::equal if two cells are equal in
every way except their ttl:s.

The problem with that is that the cells' hashes are different and this
will cause repair to keep trying to repair discrepancies caused by the
ttl being different.

This may be triggered by e.g. the spark migrator that computes the ttl
based on the expiry time by subtracting the expiry time from the current
time to produce a respective ttl.

If the cell is migrated multiple times at different times, it will generate
cells that the same expiry (by design) but have different ttl values.

Fixes #10156

Test: mutation_test.test_cell_ordering, unit(dev)

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20220302154328.2400717-1-bhalevy@scylladb.com>
(cherry picked from commit a57c087c89)
2022-03-24 18:08:07 +02:00
Benny Halevy
c48fd03463 atomic_cell: compare_atomic_cell_for_merge: fixup indentation
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20220302113833.2308533-2-bhalevy@scylladb.com>
(cherry picked from commit d43da5d6dc)

Ref #10156
2022-03-24 18:07:54 +02:00
Benny Halevy
eb78e6d4b8 atomic_cell: compare_atomic_cell_for_merge: simplify expiry/deltion_time comparison
No need to check first the the cells' expiry is different
or that deletion_time is different before comparing them
with `<=>`.

If they are the same the function returns std::strong_ordering::equal
anyhow and that is the same as `<=>` comparing identical values.

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20220302113833.2308533-1-bhalevy@scylladb.com>
(cherry picked from commit be865a29b8)

Ref #10156
2022-03-24 18:07:32 +02:00
Avi Kivity
4b1b0a55c0 replica, atomic_cell: move atomic_cell merge code from replica module to atomic_cell.cc
compare_atomic_cell_for_merge() was placed in database.cc, before
atomic_cell.cc existed. Move it to its correct place.

Closes #9889

(cherry picked from commit 6c53717a39)
2022-03-24 18:07:11 +02:00
Benny Halevy
172a8628d5 main: shutdown: do not abort on certain system errors
Currently any unhandled error during deferred shutdown
is rethrown in a noexcept context (in ~deferred_action),
generating a core dump.

The core dump is not helpful if the cause of the
error is "environmental", i.e. in the system, rather
than in scylla itself.

This change detects several such errors and calls
_Exit(255) to exit the process early, without leaving
a coredump behind.  Otherwise, call abort() explicitly,
rather than letting terminate() be called implicitly
by the destructor exception handling code.

Fixes #9573

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20220227101054.1294368-1-bhalevy@scylladb.com>
(cherry picked from commit 132c9d5933)
2022-03-24 14:49:24 +02:00
Nadav Har'El
5688b125e6 Seastar: backport Seastar fix for missing scring escape in JSON output
Backported Seastar fix:
  > Merge 'json/formatter: Escape strings' from Juliusz Stasiewicz

Fixes #9061

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
2022-03-23 21:27:13 +02:00
Piotr Sarna
6da4acb41e expression: fix get_value for mismatched column definitions
As observed in #10026, after schema changes it somehow happened
that a column defition that does not match any of the base table
columns was passed to expression verification code.
The function that looks up the index of a column happens to return
-1 when it doesn't find anything, so using this returned index
without checking if it's nonnegative results in accessing invalid
vector data, and a segfault or silent memory corruption.
Therefore, an explicit check is added to see if the column was actually
found. This serves two purposes:
 - avoiding segfaults/memory corruption
 - making it easier to investigate the root cause of #10026

Closes #10039

(cherry picked from commit 7b364fec9849e9a342af1c240e3a7185bf5401ef)
2022-03-21 10:46:34 +01:00
Botond Dénes
f09cc9a01d Merge 'service: storage_service: announce new CDC generation immediately with RBNO' from Kamil Braun
When a new CDC generation is created (during bootstrap or otherwise), it
is assigned a timestamp. The timestamp must be propagated as soon as
possible, so all live nodes can learn about the generation before their
clocks reach the generation's timestamp. The propagation mechanism for
generation timestamps is gossip.

When bootstrap RBNO was enabled this was not the case: the generation
timestamp was inserted into gossiper state too late, after the repair
phase finished. Fix this.

Also remove an obsolete comment.

Fixes https://github.com/scylladb/scylla/issues/10149.

Closes #10154

* github.com:scylladb/scylla:
  service: storage_service: announce new CDC generation immediately with RBNO
  service: storage_service: fix indentation

(cherry picked from commit f1b2ff1722)
2022-03-16 12:27:24 +01:00
Raphael S. Carvalho
cd2e33ede4 compaction_manager: Abort reshape for tables waiting for a chance to run
Tables waiting for a chance to run reshape wouldn't trigger stop
exception, as the exception was only being triggered for ongoing
compactions. Given that stop reshape API must abort all ongoing
tasks and all pending ones, let's change run_custom_job() to
trigger the exception if it found that the pending task was
asked to stop.

Tests:
dtest: compaction_additional_test.py::TestCompactionAdditional::test_stop_reshape_with_multiple_keyspaces
unit: dev

Fixes #9836.

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Message-Id: <20211223002157.215571-1-raphaelsc@scylladb.com>
(cherry picked from commit 07fba4ab5d)
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Message-Id: <20220311183053.46625-1-raphaelsc@scylladb.com>
2022-03-15 16:58:47 +02:00
Benny Halevy
32d0698d78 compaction_manager: rewrite_sstables: do not acquire table write lock
Since regular compaction may run in parallel no lock
is required per-table.

We still acquire a read lock in this patch, for backporting
purposes, in case the branch doesn't contain
6737c88045.
But it can be removed entirely in master in a follow-up patch.

This should solve some of the slowness in cleanup compaction (and
likely in upgrade sstables seen in #10060, and
possibly #10166.

Fixes #10175

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>

Closes #10177

(cherry picked from commit 11ea2ffc3c)
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20220314151416.2496374-1-bhalevy@scylladb.com>
2022-03-14 18:15:49 +02:00
Piotr Jastrzebski
93cf43ae4b cdc: Handle compact storage correctly in preimage
Base tables that use compact storage may have a special artificial
column that has an empty type.

c010cefc4d fixed the main CDC path to
handle such columns correctly and to not include them in the CDC Log
schema.

This patch makes sure that generation of preimage ignores such empty
column as well.

Fixes #9876
Closes #9910

Signed-off-by: Piotr Jastrzebski <piotr@scylladb.com>
(cherry picked from commit 09d4438a0d)
2022-03-10 14:25:02 +02:00
Nadav Har'El
2f2d22a864 cql: INSERT JSON should refuse empty-string partition key
Add the missing partition-key validation in INSERT JSON statements.

Scylla, following the lead of Cassandra, forbids an empty-string partition
key (please note that this is not the same as a null partition key, and
that null clustering keys *are* allowed).

Trying to INSERT, UPDATE or DELETE a partition with an empty string as
the partition key fails with a "Key may not be empty". However, we had a
loophole - you could insert such empty-string partition keys using an
"INSERT ... JSON" statement.

The problem was that the partition key validation was done in one place -
`modification_statement::build_partition_keys()`. The INSERT, UPDATE and
DELETE statements all inherited this same method and got the correct
validation. But the INSERT JSON statement - insert_prepared_json_statement
overrode the build_partition_keys() method and this override forgot to call
the validation function. So in this patch we add the missing validation.

Note that the validation function checks for more than just empty strings -
there is also a length limit for partition keys.

This patch also adds a cql-pytest reproducer for this bug. Before this
patch, the test passed on Cassandra but failed on Scylla.

Reported by @FortTell
Fixes #9853.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20220116085216.21774-1-nyh@scylladb.com>
(cherry picked from commit 8fd5041092)
2022-03-02 22:00:15 +02:00
Avi Kivity
5f92f54f06 Merge 'utils: cached_file: Fix alloc-dealloc mismatch during eviction' from Tomasz Grabiec
cached_page::on_evicted() is invoked in the LSA allocator context, set in the
reclaimer callback installed by the cache_tracker. However,
cached_pages are allocated in the standard allocator context (note:
page content is allocated inside LSA via lsa_buffer). The LSA region
will happily deallocate these, thinking that they these are large
objects which were delegated to the standard allocator. But the
_non_lsa_memory_in_use metric will underflow. When it underflows
enough, shard_segment_pool.total_memory() will become 0 and memory
reclamation will stop doing anything, leading to apparent OOM.

The fix is to switch to the standard allocator context inside
cached_page::on_evicted(). evict_range() was also given the same
treatment as a precaution, it currently is only invoked in the
standard allocator context.

The series also adds two safety checks to LSA to catch such problems earlier.

Fixes #10056

\cc @slivne @bhalevy

Closes #10130

* github.com:scylladb/scylla:
  lsa: Abort when trying to free a standard allocator object not allocated through the region
  lsa: Abort when _non_lsa_memory_in_use goes negative
  tests: utils: cached_file: Validate occupancy after eviction
  test: sstable_partition_index_cache_test: Fix alloc-dealloc mismatch
  utils: cached_file: Fix alloc-dealloc mismatch during eviction

(cherry picked from commit ff2cd72766)
2022-02-26 11:28:53 +02:00
Benny Halevy
395f2459b4 cql3: result_set: remove std::ref from comperator&
Applying std::ref on `RowComparator& cmp` hits the
following compilation error on Fedora 34 with
libstdc++-devel-11.2.1-9.fc34.x86_64

```
FAILED: build/dev/cql3/statements/select_statement.o
clang++ -MD -MT build/dev/cql3/statements/select_statement.o -MF build/dev/cql3/statements/select_statement.o.d -I/home/bhalevy/dev/scylla/seastar/include -I/home/bhalevy/dev/scylla/build/dev/seastar/gen/include -std=gnu++20 -U_FORTIFY_SOURCE -DSEASTAR_SSTRING -Werror=unused-result -fstack-clash-protection -DSEASTAR_API_LEVEL=6 -DSEASTAR_ENABLE_ALLOC_FAILURE_INJECTION -DSEASTAR_SCHEDULING_GROUPS_COUNT=16 -DSEASTAR_TYPE_ERASE_MORE -DFMT_LOCALE -DFMT_SHARED -I/usr/include/p11-kit-1  -DDEVEL -DSEASTAR_ENABLE_ALLOC_FAILURE_INJECTION -DSCYLLA_ENABLE_ERROR_INJECTION -O2 -DSCYLLA_ENABLE_WASMTIME -iquote. -iquote build/dev/gen --std=gnu++20  -ffile-prefix-map=/home/bhalevy/dev/scylla=.  -march=westmere -DBOOST_TEST_DYN_LINK   -Iabseil -fvisibility=hidden  -Wall -Werror -Wno-mismatched-tags -Wno-tautological-compare -Wno-parentheses-equality -Wno-c++11-narrowing -Wno-sometimes-uninitialized -Wno-return-stack-address -Wno-missing-braces -Wno-unused-lambda-capture -Wno-overflow -Wno-noexcept-type -Wno-error=cpp -Wno-ignored-attributes -Wno-overloaded-virtual -Wno-unused-command-line-argument -Wno-defaulted-function-deleted -Wno-redeclared-class-member -Wno-unsupported-friend -Wno-unused-variable -Wno-delete-non-abstract-non-virtual-dtor -Wno-braced-scalar-init -Wno-implicit-int-float-conversion -Wno-delete-abstract-non-virtual-dtor -Wno-uninitialized-const-reference -Wno-psabi -Wno-narrowing -Wno-array-bounds -Wno-nonnull -Wno-error=deprecated-declarations -DXXH_PRIVATE_API -DSEASTAR_TESTING_MAIN -DHAVE_LZ4_COMPRESS_DEFAULT  -c -o build/dev/cql3/statements/select_statement.o cql3/statements/select_statement.cc
In file included from cql3/statements/select_statement.cc:14:
In file included from ./cql3/statements/select_statement.hh:16:
In file included from ./cql3/statements/raw/select_statement.hh:16:
In file included from ./cql3/statements/raw/cf_statement.hh:16:
In file included from ./cql3/cf_name.hh:16:
In file included from ./cql3/keyspace_element_name.hh:16:
In file included from /home/bhalevy/dev/scylla/seastar/include/seastar/core/sstring.hh:25:
In file included from /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/algorithm:74:
In file included from /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/pstl/glue_algorithm_defs.h:13:
In file included from /usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/functional:58:
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/refwrap.h:319:40: error: exception specification of 'function<__gnu_cxx::__ops::_Iter_comp_iter<std::reference_wrapper<const std::function<bool (const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &, const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &)>>>, void>' uses itself
                = decltype(reference_wrapper::_S_fun(std::declval<_Up>()))>
                                                     ^
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/refwrap.h:319:40: note: in instantiation of exception specification for 'function<__gnu_cxx::__ops::_Iter_comp_iter<std::reference_wrapper<const std::function<bool (const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &, const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &)>>>, void>' requested here
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/refwrap.h:321:2: note: in instantiation of default argument for 'reference_wrapper<__gnu_cxx::__ops::_Iter_comp_iter<std::reference_wrapper<const std::function<bool (const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &, const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &)>>>, void>' required here
        reference_wrapper(_Up&& __uref)
        ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/type_traits:1017:57: note: while substituting deduced template arguments into function template 'reference_wrapper' [with _Up = __gnu_cxx::__ops::_Iter_comp_iter<std::reference_wrapper<const std::function<bool (const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &, const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &)>>>, $1 = (no value), $2 = (no value)]
      = __bool_constant<__is_nothrow_constructible(_Tp, _Args...)>;
                                                        ^
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/type_traits:1023:14: note: in instantiation of template type alias '__is_nothrow_constructible_impl' requested here
    : public __is_nothrow_constructible_impl<_Tp, _Args...>::type
             ^
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/type_traits:153:14: note: in instantiation of template class 'std::is_nothrow_constructible<__gnu_cxx::__ops::_Iter_comp_iter<std::reference_wrapper<const std::function<bool (const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &, const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &)>>>, __gnu_cxx::__ops::_Iter_comp_iter<std::reference_wrapper<const std::function<bool (const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &, const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &)>>>>' requested here
    : public conditional<_B1::value, _B2, _B1>::type
             ^
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/std_function.h:298:11: note: (skipping 8 contexts in backtrace; use -ftemplate-backtrace-limit=0 to see all)
          return __and_<typename _Base::_Local_storage,
                 ^
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algo.h:1933:13: note: in instantiation of function template specialization 'std::__partial_sort<utils::chunked_vector<std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>>, 131072>::iterator_type<std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>>>, __gnu_cxx::__ops::_Iter_comp_iter<std::reference_wrapper<const std::function<bool (const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &, const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &)>>>>' requested here
              std::__partial_sort(__first, __last, __last, __comp);
                   ^
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algo.h:1954:9: note: in instantiation of function template specialization 'std::__introsort_loop<utils::chunked_vector<std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>>, 131072>::iterator_type<std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>>>, long, __gnu_cxx::__ops::_Iter_comp_iter<std::reference_wrapper<const std::function<bool (const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &, const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &)>>>>' requested here
          std::__introsort_loop(__first, __last,
               ^
/usr/lib/gcc/x86_64-redhat-linux/11/../../../../include/c++/11/bits/stl_algo.h:4875:12: note: in instantiation of function template specialization 'std::__sort<utils::chunked_vector<std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>>, 131072>::iterator_type<std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>>>, __gnu_cxx::__ops::_Iter_comp_iter<std::reference_wrapper<const std::function<bool (const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &, const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &)>>>>' requested here
      std::__sort(__first, __last, __gnu_cxx::__ops::__iter_comp_iter(__comp));
           ^
./cql3/result_set.hh:168:14: note: in instantiation of function template specialization 'std::sort<utils::chunked_vector<std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>>, 131072>::iterator_type<std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>>>, std::reference_wrapper<const std::function<bool (const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &, const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &)>>>' requested here
        std::sort(_rows.begin(), _rows.end(), std::ref(cmp));
             ^
cql3/statements/select_statement.cc:773:21: note: in instantiation of function template specialization 'cql3::result_set::sort<std::function<bool (const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &, const std::vector<std::optional<seastar::basic_sstring<signed char, unsigned int, 31, false>>> &)>>' requested here
                rs->sort(_ordering_comparator);
                    ^
1 error generated.
ninja: build stopped: subcommand failed.
```

Fixes #10079.

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20220215071955.316895-3-bhalevy@scylladb.com>
(cherry picked from commit 3e20fee070)

[avi: backport for developer quality-of-life rather than as a bug fix]
2022-02-16 10:08:24 +02:00
Raphael S. Carvalho
019d50bb5c Revert "sstables/compaction_manager: rewrite_sstables(): resolve maintenance group FIXME"
This reverts commit 4c05e5f966.

Moving cleanup to maintenance group made its operation time up to
10x slower than previous release. It's a blocker to 4.6 release,
so let's revert it until we figure this all out.

Probably this happens because maintenance group is fixed at a
relatively small constant, and cleanup may be incrementally
generating backlog for regular compaction, where the former is
fighting for resources against the latter.

Fixes #10060.

Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Message-Id: <20220213165147.56204-1-raphaelsc@scylladb.com>

Ref: a9427f150a
2022-02-14 12:10:38 +02:00
Avi Kivity
bbe775b926 utils: logalloc: correct and adjust timing unit in stall report
The stall report uses the millisecond unit, but actually reports
nanoseconds.

Switch to microseconds (milliseconds are a bit too coarse) and
use the safer "duration / 1us" style rather than "duration::count()"
that leads to unit confusion.

Fixes #9733.

Closes #9734

(cherry picked from commit f907205b92)
2022-02-12 15:56:42 +02:00
Yaron Kaikov
469c94ea17 release: prepare for 4.6.0 2022-02-08 16:45:50 +02:00
Nadav Har'El
4c780d0265 alternator: allow REMOVE of non-existent nested attribute
DynamoDB allows an UpdateItem operation "REMOVE x.y" when a map x
exists in the item, but x.y doesn't - the removal silently does
nothing. Alternator incorrectly generated an error in this case,
and unfortunately we didn't have a test for this case.

So in this patch we add the missing test (which fails on Alternator
before this patch - and passes on DynamoDB) and then fix the behavior.
After this patch, "REMOVE x.y" will remain an error if "x" doesn't
exist (saying "document paths not valid for this item"), but if "x"
exists and is a map, but "x.y" doesn't, the removal will silently
do nothing and will not be an error.

Fixes #10043.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20220207133652.181994-1-nyh@scylladb.com>
(cherry picked from commit 9982a28007)
2022-02-08 11:48:18 +02:00
Michael Livshin
0181de1f2c shard_reader: check that _reader is valid before dereferencing
After fc729a804, `shard_reader::close()` is not interrupted with an
exception any more if read-ahead fails, so `_reader` may in fact be
null.

Fixes #9923

Signed-off-by: Michael Livshin <michael.livshin@scylladb.com>
Message-Id: <20220117120405.152927-1-michael.livshin@scylladb.com>
(cherry picked from commit d7a993043d)
2022-02-07 10:10:58 +02:00
Benny Halevy
7597a79ef9 shard_reader: Continue after read_ahead error
If read ahead failed, just issue a log warning
and proceed to close the reader.

Currently co_await will throw and the evictable reader
won't be closed.

This is seen occasionally in testing, e.g.
https://jenkins.scylladb.com/view/master/job/scylla-master/job/dtest-debug/1010/artifact/logs-all.debug.2/1640918573898_lwt_banking_load_test.py%3A%3ATestLWTBankingLoad%3A%3Atest_bank_with_nemesis/node2.log
```
ERROR 2021-12-31 02:40:56,160 [shard 0] mutation_reader - shard_reader::close(): failed to stop reader on shard 1: seastar::named_semaphore_timed_out (Semaphore timed out: _system_read_concurrency_sem)
```

Fixes #9865.

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20220102124636.2791544-1-bhalevy@scylladb.com>
(cherry picked from commit fc729a804b)
2022-02-07 10:09:05 +02:00
Nadav Har'El
8f5148e921 docker: don't repeat "--alternator-address" option twice
If the Docker startup script is passed both "--alternator-port" and
"--alternator-https-port", a combination which is supposed to be
allowed, it passes to Scylla the "--alternator-address" option twice.
This isn't necessary, and worse - not allowed.

So this patch fixes the scyllasetup.py script to only pass this
parameter once.

Fixes #10016.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20220202165814.1700047-1-nyh@scylladb.com>
(cherry picked from commit cb6630040d)
2022-02-03 18:39:47 +02:00
Yaron Kaikov
5694ec189f release: prepare for 4.6.rc5 2022-02-03 16:19:46 +02:00
Calle Wilund
34d470967a commitlog: Fix double clearing of _segment_allocating shared_future.
Fixes #10020

Previous fix 445e1d3 tried to close one double invocation,  but added
another, since it failed to ensure all potential nullings of the opt
shared_future happened before a new allocator could reset it.

This simplifies the code by making clearing the shared_future a
pre-requisite for resolving its contents (as read by waiters).

Also removes any need for try-catch etc.

Closes #10024

(cherry picked from commit 1e66043412)
2022-02-03 07:43:18 +02:00
Calle Wilund
61db571a44 commitlog: Ensure we never have more than one new_segment call at a time
Refs #9896

Found by @eliransin. Call to new_segment was wrapped in with_timeout.
This means that if primary caller timed out, we would leave new_segment
calls running, but potentially issue new ones for next caller.

This could lead to reserve segment queue being read simultanously. And
it is not what we want.

Change to always use the shared_future wait, all callers, and clear it
only on result (exception or segment)

Closes #10001

(cherry picked from commit 445e1d3e41)
2022-02-01 09:10:27 +02:00
Tomasz Grabiec
5b5a300a9e util: cached_file: Fix corruption after memory reclamation was triggered from population
If memory reclamation is triggered inside _cache.emplace(), the _cache
btree can get corrupted. Reclaimers erase from it, and emplace()
assumes that the tree is not modified during its execution. It first
locates the target node and then does memory allocation.

Fix by running emplace() under allocating section, which disables
memory reclamation.

The bug manifests with assert failures, e.g:

./utils/bptree.hh:1699: void bplus::node<unsigned long, cached_file::cached_page, cached_file::page_idx_less_comparator, 12, bplus::key_search::linear, bplus::with_debug::no>::refill(Less) [Key = unsigned long, T = cached_file::cached_page, Less = cached_file::page_idx_less_comparator, NodeSize = 12, Search = bplus::key_search::linear, Debug = bplus::with_debug::no]: Assertion `p._kids[i].n == this' failed.

Fixes #9915

Message-Id: <20220130175639.15258-1-tgrabiec@scylladb.com>
(cherry picked from commit b734615f51)
2022-01-31 01:24:47 +02:00
Avi Kivity
148a65d0d6 Update seastar submodule (gratuitous exceptions on allocation failure)
* seastar a189cdc45d...a375681303 (1):
  > core: memory: Avoid current_backtrace() on alloc failure when logging suppressed

Fixes #9982.
2022-01-30 20:02:24 +02:00
Avi Kivity
e3ad14d55f Point seastar submodule at scylla-seastar.git
This allows us to backport fixes to seastar selectively.
2022-01-30 20:01:12 +02:00
Calle Wilund
2b506c2d4a commitlog: Ensure we don't run continuation (task switch) with queues modified
Fixes #9955

In #9348 we handled the problem of failing to delete segment files on disk, and
the need to recompute disk footprint to keep data flow consistent across intermittent
failures. However, because _reserve_segments and _recycled_segments are queues, we
have to empty them to inspect the contents. One would think it is ok for these
queues to be empty for a while, whilst we do some recaclulating, including
disk listing -> continuation switching. But then one (i.e. I) misses the fact
that these queues use the pop_eventually mechanism, which does _not_ handle
a scenario where we push something into an empty queue, thus triggering the
future that resumes a waiting task, but then pop the element immediately, before
the waiting task is run. In fact, _iff_ one does this, not only will things break,
they will in fact start creating undefined behaviour, because the underlying
std::queue<T, circular_buffer> will _not_ do any bounds checks on the pop/push
operations -> we will pop an empty queue, immediately making it non-empty, but
using undefined memory (with luck null/zeroes).

Strictly speakging, seastar::queue::pop_eventually should be fixed to handle
the scenario, but nontheless we can fix the usage here as well, by simply copy
objects and do the calculation "in background" while we potentially start
popping queue again.

Closes #9966

(cherry picked from commit 43f51e9639)
2022-01-27 10:24:03 +02:00
Avi Kivity
50aad1c668 Merge 'scylla_raid_setup: use mdmonitor only when RAID level > 0' from Takuya ASADA
We found that monitor mode of mdadm does not work on RAID0, and it is
not a bug, expected behavior according to RHEL developer.
Therefore, we should stop enabling mdmonitor when RAID0 is specified.

Fixes #9540

----

This reverts 0d8f932 and introduce correct fix.

Closes #9970

* github.com:scylladb/scylla:
  scylla_raid_setup: use mdmonitor only when RAID level > 0
  Revert "scylla_raid_setup: workaround for mdmonitor.service issue on CentOS8"

(cherry picked from commit df22396a34)
2022-01-27 10:21:25 +02:00
Yaron Kaikov
7bf3f37cd1 release: prepare for 4.6.rc4 2022-01-23 10:44:09 +02:00
Botond Dénes
0f7f8585f2 reader_permit: release_base_resources(): also update _resources
If the permit was admitted, _base_resources was already accounted in
_resource and therefore has to be deducted from it, otherwise the permit
will think it leaked some resources on destruction.

Test:
dtest(repair_additional_test.py.test_repair_one_missing_row_diff_shard_count)

Refs: #9751
Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <20220119132550.532073-1-bdenes@scylladb.com>
(cherry picked from commit a65b38a9f7)
2022-01-20 18:39:25 +02:00
Pavel Emelyanov
2c65c4a569 Merge 'db: range_tombstone_list: Deoverlap empty range tombstones' from Tomasz Grabiec
Appending an empty range adjacent to an existing range tombstone would
not deoverlap (by dropping the empty range tombstone) resulting in
different (non canoncial) result depending on the order of appending.

Suppose that range tombstone [a, b] covers range tombstone [x, x), and [a, x) and [x, b) are range tombstones which correspond to [a, b] split around position x.

Appending [a, x) then [x, b) then [x, x) would give [a, b)
Appending [a, x) then [x, x) then [x, b) would give [a, x), [x, x), [x, b)

The fix is to drop empty range tombstones in range_tombstone_list so that the result is canonical.

Fixes #9661

Closes #9764

* github.com:scylladb/scylla:
  range_tombstone_list: Deoverlap adjacent empty ranges
  range_tombstone_list: Convert to work in terms of position_in_partition

(cherry picked from commit b2a62d2b59)
2022-01-20 12:35:21 +02:00
Avi Kivity
f85cd289bc Merge "repair: make sure there is one permit per repair with count res" from Botond
"
Repair obtains a permit for each repair-meta instance it creates. This
permit is supposed to track all resources consumed by that repair as
well as ensure concurrency limit is respected. However when the
non-local reader path is used (shard config of master != shard config of
follower), a second permit will be obtained -- for the shard reader of
the multishard reader. This creates a situation where the repair-meta's
permit can block the shard permit, creating a deadlock situation.
This patch solves this by dropping the count resource on the
repair-meta's permit when a non-local reader path is executed -- that is
a multishard reader is created.

Fixes: #9751
"

* 'repair-double-permit-block/v4' of https://github.com/denesb/scylla:
  repair: make sure there is one permit per repair with count res
  reader_permit: add release_base_resource()

(cherry picked from commit 52b7778ae6)
2022-01-17 16:02:55 +02:00
Beni Peled
5e661af9a4 release: prepare for 4.6.rc3 2022-01-17 13:11:54 +02:00
Calle Wilund
5629b67d25 messaging_service: Make dc/rack encryption check for connection more strict
Fixes #9653

When doing an outgoing connection, in a internode_encryption=dc/rack situation
we should not use endpoint/local broadcast solely to determine if we can
downgrade a connection.

If gossip/message_service determines that we will connect to a different
address than the "official" endpoint address, we should use this to determine
association of target node, and similarly, if we bind outgoing connection
to interface != bc we need to use this to decide local one.

Note: This will effectively _disable_ internode_encryption=dc/rack on ec2 etc
until such time that gossip can give accurate info on dc/rack for "internal"
ip addresses of nodes.

(cherry picked from commit 4778770814)
2022-01-16 19:10:57 +02:00
Takuya ASADA
ad632cf7fc dist: fix scylla-housekeeping uuid file chmod call
Should use chmod() on a file, not fchmod()

Fixes #9683

Closes #9802

(cherry picked from commit 7064ae3d90)
2022-01-10 16:57:34 +02:00
Botond Dénes
ca24bebcf2 sstables/partition_index_cache: destroy entry ptr on error
The error-handling code removes the cache entry but this leads to an
assertion because the entry is still referenced by the entry pointer
instance which is returned on the normal path. To avoid this clear the
pointer on the error path and make sure there are no additional
references kept to it.

Fixes #9887

Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <20220105140859.586234-2-bdenes@scylladb.com>
(cherry picked from commit 92727ac36c)
2022-01-07 21:21:44 +01:00
Calle Wilund
7dc5abb6f8 commitlog: Don't allow error_handler to swallow exception
Fixes #9798

If an exception in allocate_segment_ex is (sub)type of std::system_error,
commit_error_handler might _not_ cause throw (doh), in which case the error
handling code would forget the current exception and return an unusable
segment.

Now only used as an exception pointer replacer.

Closes #9870

(cherry picked from commit 3c02cab2f7)
2022-01-06 14:10:18 +02:00
Yaron Kaikov
e8a1cfb6f8 release: prepare for 4.6.rc2 2022-01-02 09:15:47 +02:00
Tomasz Grabiec
fc312b3021 lsa: Fix segment leak on memory reclamation during alloc_buf
alloc_buf() calls new_buf_active() when there is no active segment to
allocate a new active segment. new_buf_active() allocates memory
(e.g. a new segment) so may cause memory reclamation, which may cause
segment compaction, which may call alloc_buf() and re-enter
new_buf_active(). The first call to new_buf_active() would then
override _buf_active and cause the segment allocated during segment
compaction to be leaked.

This then causes abort when objects from the leaked segment are freed
because the segment is expected to be present in _closed_segments, but
isn't. boost::intrusive::list::erase() will fail on assertion that the
object being erased is linked.

Introduced in b5ca0eb2a2.

Fixes #9821
Fixes #9192
Fixes #9825
Fixes #9544
Fixes #9508
Refs #9573

Message-Id: <20211229201443.119812-1-tgrabiec@scylladb.com>
(cherry picked from commit 7038dc7003)
2021-12-30 18:56:28 +02:00
Nadav Har'El
7b82aaf939 alternator: fix error on UpdateTable for non-existent table
When the UpdateTable operation is called for a non-existent table, the
appropriate error is ResourceNotFoundException, but before this patch
we ran into an exception, which resulted in an ugly "internal server
error".

In this patch we use the existing get_table() function which most other
operations use, and which does all the appropriate verifications and
generates the appropriate Alternator api_error instead of letting
internal Scylla exceptions escape to the user.

This patch also includes a test for UpdateTable on a non-existent table,
which used to fail before this patch and pass afterwards. We also add a
test for DeleteTable in the same scenario, and see it didn't have this
bug. As usual, both tests pass on DynamoDB, which confirms we generate
the right error codes.

Fixes #9747.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20211206181605.1182431-1-nyh@scylladb.com>
(cherry picked from commit 31eeb44d28)
2021-12-29 22:59:25 +02:00
Nadav Har'El
894a4abfae commitlog: fix missing wait for semaphore units
Commit dcc73c5d4e introduced a semaphore
for excluding concurrent recalculations - _reserve_recalculation_guard.

Unfortunately, the two places in the code which tried to take this
guard just called get_units() - which returns a future<units>, not
units - and never waited for this future to become available.

So this patch adds the missing "co_await" needed to wait for the
units to become available.

Fixes #9770.

Signed-off-by: Nadav Har'El <nyh@scylladb.com>
Message-Id: <20211214122612.1462436-1-nyh@scylladb.com>
(cherry picked from commit b8786b96f4)
2021-12-29 13:18:59 +02:00
Takuya ASADA
4dcf023470 scylla_raid_setup: workaround for mdmonitor.service issue on CentOS8
On CentOS8, mdmonitor.service does not works correctly when using
mdadm-4.1-15.el8.x86_64 and later versions.
Until we find a solution, let's pinning the package version to older one
which does not cause the issue (4.1-14.el8.x86_64).

Fixes #9540

Closes #9782

(cherry picked from commit 0d8f932f0b)
2021-12-28 11:38:04 +02:00
Benny Halevy
283788828e compaction: scrub_validate_mode_validate_reader: throw compaction_stopped_exception if stop is requested
Currently when scrub/validate is stopped (e.g. via the api),
scrub_validate_mode_validate_reader co_return:s without
closing the reader passed to it - causing a crash due
to internal error check, see #9766.

Throwing a compaction_stopped_exception rather than co_return:ing
an exception will be handled as any other exeption, including closing
the reader.

Fixes #9766

Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
Message-Id: <20211213125528.2422745-1-bhalevy@scylladb.com>
(cherry picked from commit c89876c975)
2021-12-15 15:03:59 +02:00
Pavel Emelyanov
730a147ba6 row-cache: Handle exception (un)safety of rows_entry insertion
The B-tree's insert_before() is throwing operation, its caller
must account for that. When the rows_entry's collection was
switched on B-tree all the risky places were fixed by ee9e1045,
but few places went under the radar.

In the cache_flat_mutation_reader there's a place where a C-pointer
is inserted into the tree, thus potentially leaking the entry.

In the partition_snapshot_row_cursor there are two places that not
only leak the entry, but also leave it in the LRU list. The latter
it quite nasty, because those entry can be evicted, eviction code
tries to get rows_entry iterator from "this", but the hook happens
to be unattached (because insertion threw) and fails the assert.

fixes: #9728

Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
(cherry picked from commit ee103636ac)
2021-12-14 15:53:42 +02:00
Pavel Emelyanov
9897e83029 partition_snapshot_row_cursor: Shuffle ensure_result creation
Both places get the C-pointer on the freshly allocated rows_entry,
insert it where needed and return back the dereferenced pointer.

The C-pointer is going to become smart-pointer that would go out
of scope before return. This change prepares for that by constructing
the ensure_result from the iterator, that's returned from insertion
of the entry.

Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
(cherry picked from commit 9fd8db318d)

Ref #9728
2021-12-14 15:52:37 +02:00
Asias He
1a9b64e6f6 storage_service: Wait for seastar::get_units in node_ops
The seastar::get_units returns a future, we have to wait for it.

Fixes #9767

Closes #9768

(cherry picked from commit 9859c76de1)
2021-12-12 18:42:20 +02:00
Takuya ASADA
49fe9e2c8e dist: allow running scylla-housekeeping with strict umask setting
To avoid failing scylla-housekeeping in strict umask environment,
we need to chmod a+r on repository file and housekeeping.uuid.

Fixes #9683

Closes #9739

(cherry picked from commit ea20f89c56)
2021-12-12 14:25:57 +02:00
Takuya ASADA
d0580c41ee dist: add support im4gn/is4gen instance on AWS
Add support next-generation, storage-optimized ARM64 instance types.

Fixes #9711

Closes #9730

(cherry picked from commit 097a6ee245)
2021-12-08 14:29:44 +02:00
Beni Peled
542394c82f release: prepare for 4.6.rc1 2021-12-08 11:08:45 +02:00
Avi Kivity
018ad3f6f4 test: refine test suite names exposed via xunit format
The test suite names seen by Jenkins are suboptimal: there is
no distinction between modes, and the ".cc" suffix of file names
is interpreted as a class name, which is converted to a tree node
that must be clicked to expand. Massage the names to remove
unnecessary information and add the mode.

Closes #9696

(cherry picked from commit ef3edcf848)

Fixes #9738.
2021-12-05 19:58:22 +02:00
Avi Kivity
9b8b7efb54 tests: consolidate boost xunit result files
The recent parallelization of boost unit tests caused an increase
in xml result files. This is challenging to Jenkins, since it
appears to use rpc-over-ssh to read the result files, and as a result
it takes more than an hour to read all result files when the Jenkins
main node is not on the same continent as the agent.

To fix this, merge the result files in test.py and leave one result
file per mode. Later we can leave one result file overall (integrating
the mode into the testsuite name), but that can wait.

Tested on a local Jenkins instance (just reading the result files,
not the entire build).

Closes #9668

(cherry picked from commit b23af15432)

Fixes #9738
2021-12-05 19:57:39 +02:00
Botond Dénes
1c3e63975f Merge 'Backport of #9348 (xceptions in commitlog::segment_manager::delete_segments could cause footprint counters to loose track)' from Calle Wilund
Backport of series to 4.6
Upstream merge commit: e2c27ee743.
Refs #9348

Closes #9702

* github.com:scylladb/scylla:
  commitlog: Recalculate footprint on delete_segment exceptions
  commitlog_test: Add test for exception in alloc w. deleted underlying file
  commitlog: Ensure failed-to-create-segment is re-deleted
  commitlog::allocate_segment_ex: Don't re-throw out of function
2021-12-02 09:22:19 +02:00
Calle Wilund
11bb03e46d commitlog: Recalculate footprint on delete_segment exceptions
Fixes #9348

If we get exceptions in delete_segments, we can, and probably will, loose
track of footprint counters. We need to recompute the used disk footprint,
otherwise we will flush too often, and even block indefinately on new_seg
iff using hard limits.
2021-11-29 14:56:48 +00:00
Calle Wilund
810e410c5d commitlog_test: Add test for exception in alloc w. deleted underlying file
Tests that we can handle exception-in-alloc cleanup if the file actually
does not exist. This however uncovers another weakness (addressed in next
patch) - that we can loose track of disk footprint here, and w. hard limits
end up waiting for disk space that never comes. Thus test does not use hard
limit.
2021-11-29 14:56:43 +00:00
Calle Wilund
97f6da0c3e commitlog: Ensure failed-to-create-segment is re-deleted
Fixes #9343

If we fail in allocate_segment_ex, we should push the file opened/created
to the delete set to ensure we reclaim the disk space. We should also
ensure that if we did not recycle a file in delete_segments, we still
wake up any recycle waiters iff we made a file delete instead.

Included a small unit test.
2021-11-29 14:51:39 +00:00
Calle Wilund
c229fe9694 commitlog::allocate_segment_ex: Don't re-throw out of function
Fixes #9342

commitlog_error_handler rethrows. But we want to not. And run post-handler
cleanup (co_await)
2021-11-29 14:51:39 +00:00
Tomasz Grabiec
ee1ca8ae4d lsa: Add sanity checks around lsa_buffer operations
We've been observing hard to explain crashes recently around
lsa_buffer destruction, where the containing segment is absent in
_segment_descs which causes log_heap::adjust_up to abort. Add more
checks to catch certain impossible senarios which can lead to this
sooner.

Refs #9192.
Message-Id: <20211116122346.814437-1-tgrabiec@scylladb.com>

(cherry picked from commit bf6898a5a0)
2021-11-24 15:17:37 +01:00
Tomasz Grabiec
6bfd322e3b lsa: Mark compact_segment_locked() as noexcept
We cannot recover from a failure in this method. The implementation
makes sure it never happens. Invariants will be broken if this
throws. Detect violations early by marking as noexcept.

We could make it exception safe and try to leave the data structures
in a consistent state but the reclaimer cannot make progress if this throws, so
it's pointless.

Refs #9192
Message-Id: <20211116122019.813418-1-tgrabiec@scylladb.com>

(cherry picked from commit 4d627affc3)
2021-11-24 15:17:35 +01:00
Tomasz Grabiec
afc18d5070 cql: Fix missing data in indexed queries with base table short reads
Indexed queries are using paging over the materialized view
table. Results of the view read are then used to issue reads of the
base table. If base table reads are short reads, the page is returned
to the user and paging state is adjusted accordingly so that when
paging is resumed it will query the view starting from the row
corresponding to the next row in the base which was not yet
returned. However, paging state's "remaining" count was not reset, so
if the view read was exhausted the reading will stop even though the
base table read was short.

Fix by restoring the "remaining" count when adjusting the paging state
on short read.

Tests:

  - index_with_paging_test
  - secondary_index_test

Fixes #9198
Message-Id: <20210818131840.1160267-1-tgrabiec@scylladb.com>

(cherry picked from commit 1e4da2dcce)
2021-11-23 11:22:00 +02:00
Tomasz Grabiec
2ec22c2404 sstables: partition_index_cache: Avoid abort due to benign bad_alloc inside allocating section
shared_promise::get_shared_future() is marked noexcept, but can
allocate memory. It is invoked by sstable partition index cache inside
an allocating section, which means that allocations can throw
bad_alloc even though there is memory to reclaim, so under normal
conditions.

Fix by allocating the shared_promise in a stable memory, in the
standard allocator via lw_shared_ptr<>, so that it can be accessed outside
allocating section.

Fixes #9666

Tests:

  - build/dev/test/boost/sstable_partition_index_cache_test

Message-Id: <20211122165100.1606854-1-tgrabiec@scylladb.com>
(cherry picked from commit 1d84bc6c3b)
2021-11-23 11:21:27 +02:00
Avi Kivity
19da778271 Merge "Run gossiper message handlers in a gate" from Pavel E
"
When gossiper processes its messages in the background some of
the continuations may pop up after the gossiper is shutdown.
This, in turn, may result in unwanted code to be executed when
it doesn't expect.

In particular, storage_service notification hooks may try to
update system keyspace (with "fresh" peer info/state/tokens/etc).
This update doesn't work after drain because drain shuts down
commitlog. The intention was that gossiper did _not_ notify
anyone after drain, because it's shut down during drain too.
But since there are background continuations left, it's not
working as expected.

refs: #9567
tests: unit(dev), dtest.concurrent_schema_changes.snapshot(dev)
"

* 'br-gossiper-background-messages-2' of https://github.com/xemul/scylla:
  gossiper: Guard background processing with gate
  gossiper: Helper for background messaging processing

(cherry picked from commit 9e2b6176a2)
2021-11-19 07:25:26 +02:00
Avi Kivity
cbd4c13ba6 Merge 'Revert "scylla_util.py: return bool value on systemd_unit.is_active()"' from Takuya ASADA
On scylla_unit.py, we provide `systemd_unit.is_active()` to return `systemctl is-active` output.
When we introduced systemd_unit class, we just returned `systemctl is-active` output as string, but we changed the return value to bool after that (2545d7fd43).
This was because `if unit.is_active():` always becomes True even it returns "failed" or "inactive", to avoid such scripting bug.
However, probably this was mistake.
Because systemd unit state is not 2 state, like "start" / "stop", there are many state.

And we already using multiple unit state ("activating", "failed", "inactive", "active") in our Cloud image login prompt:
https://github.com/scylladb/scylla-machine-image/blob/next/common/scylla_login#L135
After we merged 2545d7fd43, the login prompt is broken, because it does not return string as script expected (https://github.com/scylladb/scylla-machine-image/issues/241).

I think we should revert 2545d7fd43, it should return exactly same value as `systemctl is-active` says.

Fixes #9627
Fixes scylladb/scylla-machine-image#241

Closes #9628

* github.com:scylladb/scylla:
  scylla_ntp_setup: use string in systemd_unit.is_active()
  Revert "scylla_util.py: return bool value on systemd_unit.is_active()"

(cherry picked from commit c17101604f)
2021-11-18 11:44:11 +02:00
Pavel Emelyanov
338871802d generic_server: Keep server alive during conn background processing
There's at least one tiny race in generic_server code. The trailing
.handle_exception after the conn->process() captures this, but since the
whole continuation chain happens in the background, that this can be
released thus causing the whole lambda to execute on freed generic_server
instance. This, in turn, is not nice because captured this is used to get
a _logger from.

The fix is based on the observation that all connections pin the server
in memory until all of them (connections) are destructed. Said that, to
keep the server alive in the aforementioned lambda it's enough to make
sure the conn variable (it's lw_shared_ptr on the connection) is alive in
it. Not to generate a bunch of tiny continuations with identical set of
captures -- tail the single .then_wrapped() one and do whatever is needed
to wrap up the connection processing in it.

tests: unit(dev)
fixes: #9316

Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
Message-Id: <20211115105818.11348-1-xemul@scylladb.com>
(cherry picked from commit ba16318457)
2021-11-17 10:21:11 +02:00
Yaron Kaikov
8b5b1b8af6 dist/docker/debian/build_docker.sh: debian version fix for rc releases
When building a docker we relay on `VERSION` value from
`SCYLLA-VERSION-GEN` . For `rc` releases only there is a different
between the configured version (X.X.rcX) and the actualy debian package
we generate (X.X~rcX)

Using a similar solution as i did in dcb10374a5

Fixes: #9616

Closes #9617

(cherry picked from commit 060a91431d)
2021-11-12 20:07:19 +02:00
Takuya ASADA
ea89eff95d dist/docker: fix bashrc filename for Ubuntu
For Debian variants, correct filename is /etc/bash.bashrc.

Fixes #9588

Closes #9589

(cherry picked from commit 201a97e4a4)
2021-11-10 14:25:27 +02:00
Michał Radwański
96421e7779 memtable: fix gcc function argument evaluation order induced use after move
clang evaluates function arguments from left to right, while gcc does so
in reverse. Therefore, this code can be correct on clang and incorrect
on gcc:
```
f(x.sth(), std::move(x))
```

This patch fixes one such instance of this bug, in memtable.cc.

Fixes #9605.

Closes #9606

(cherry picked from commit eff392073c)
2021-11-10 08:58:09 +02:00
Botond Dénes
142336ca53 mutation_writer/feed_writer: don't drop readers with small amount of content
Due to an error in transforming the above routine, readers who have <= a
buffer worth of content are dropped without consuming them.
This is due to the outer consume loop being conditioned on
`is_end_of_stream()`, which will be set for readers that eagerly
pre-fill their buffer and also have no more data then what is in their
buffer.
Change the condition to also check for `is_buffer_empty()` and only drop
the reader if both of these are true.

Fixes: #9594

Tests: unit(mutation_writer_test --repeat=200, dev)
Signed-off-by: Botond Dénes <bdenes@scylladb.com>
Message-Id: <20211108092923.104504-1-bdenes@scylladb.com>
(cherry picked from commit 4b6c0fe592)
2021-11-09 14:13:21 +02:00
Calle Wilund
492f12248c commitlog: Add explicit track var for "wasted space" to avoid double counting
Refs #9331

In segment::close() we add space to managers "wasted" counter. In destructor,
if we can cleanly delete/recycle the file we remove it. However, if we never
went through close (shutdown - ok, exception in batch_cycle - not ok), we can
end up subtracting numbers that were never added in the first place.
Just keep track of the bytes added in a var.

Observed behaviour in above issue is timeouts in batch_cycle, where we
declare the segment closed early (because we cannot add anything more safely
- chunks could get partial/misplaced). Exception will propagate to caller(s),
but the segment will not go through actual close() call -> destructor should
not assume such.

Closes #9598

(cherry picked from commit 3929b7da1f)
2021-11-09 14:07:04 +02:00
Yaron Kaikov
7eb7a0e5fe release: prepare for 4.6.rc0 2021-11-08 09:18:26 +02:00
3289 changed files with 110002 additions and 241796 deletions

1
.gitattributes vendored
View File

@@ -1,3 +1,2 @@
*.cc diff=cpp
*.hh diff=cpp
*.svg binary

42
.github/CODEOWNERS vendored
View File

@@ -2,17 +2,17 @@
auth/* @elcallio @vladzcloudius
# CACHE
row_cache* @tgrabiec
*mutation* @tgrabiec
test/boost/mvcc* @tgrabiec
row_cache* @tgrabiec @haaawk
*mutation* @tgrabiec @haaawk
test/boost/mvcc* @tgrabiec @haaawk
# CDC
cdc/* @kbr- @elcallio @piodul @jul-stas
test/cql/cdc_* @kbr- @elcallio @piodul @jul-stas
test/boost/cdc_* @kbr- @elcallio @piodul @jul-stas
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 @eliransin
db/commitlog/* @elcallio
db/batch* @elcallio
# COORDINATOR
@@ -25,15 +25,11 @@ compaction/* @raphaelsc @nyh
transport/*
# CQL QUERY LANGUAGE
cql3/* @tgrabiec @cvybhu @nyh
cql3/* @tgrabiec @psarna @cvybhu
# COUNTERS
counters* @jul-stas
tests/counter_test* @jul-stas
# DOCS
docs/* @annastuchlik @tzach
docs/alternator @annastuchlik @tzach @nyh @havaker @nuivall
counters* @haaawk @jul-stas
tests/counter_test* @haaawk @jul-stas
# GOSSIP
gms/* @tgrabiec @asias
@@ -45,9 +41,9 @@ dist/docker/*
utils/logalloc* @tgrabiec
# MATERIALIZED VIEWS
db/view/* @nyh @cvybhu @piodul
cql3/statements/*view* @nyh @cvybhu @piodul
test/boost/view_* @nyh @cvybhu @piodul
db/view/* @nyh @psarna
cql3/statements/*view* @nyh @psarna
test/boost/view_* @nyh @psarna
# PACKAGING
dist/* @syuu1228
@@ -62,9 +58,9 @@ service/migration* @tgrabiec @nyh
schema* @tgrabiec @nyh
# SECONDARY INDEXES
index/* @nyh @cvybhu @piodul
cql3/statements/*index* @nyh @cvybhu @piodul
test/boost/*index* @nyh @cvybhu @piodul
db/index/* @nyh @psarna
cql3/statements/*index* @nyh @psarna
test/boost/*index* @nyh @psarna
# SSTABLES
sstables/* @tgrabiec @raphaelsc @nyh
@@ -74,11 +70,11 @@ streaming/* @tgrabiec @asias
service/storage_service.* @tgrabiec @asias
# ALTERNATOR
alternator/* @nyh @havaker @nuivall
test/alternator/* @nyh @havaker @nuivall
alternator/* @nyh @psarna
test/alternator/* @nyh @psarna
# HINTED HANDOFF
db/hints/* @piodul @vladzcloudius @eliransin
db/hints/* @haaawk @piodul @vladzcloudius
# REDIS
redis/* @nyh @syuu1228

View File

@@ -1,17 +0,0 @@
name: "Docs / Amplify enhanced"
on: issue_comment
jobs:
build:
runs-on: ubuntu-latest
if: ${{ github.event.issue.pull_request }}
steps:
- name: Checkout
uses: actions/checkout@v3
with:
fetch-depth: 0
- name: Amplify enhanced
env:
TOKEN: ${{ secrets.GITHUB_TOKEN }}
uses: scylladb/sphinx-scylladb-theme/.github/actions/amplify-enhanced@master

View File

@@ -1,40 +0,0 @@
name: "Docs / Publish"
# For more information,
# see https://sphinx-theme.scylladb.com/stable/deployment/production.html#available-workflows
env:
FLAG: ${{ github.repository == 'scylladb/scylla-enterprise' && 'enterprise' || 'opensource' }}
on:
push:
branches:
- 'master'
- 'enterprise'
paths:
- "docs/**"
workflow_dispatch:
jobs:
release:
runs-on: ubuntu-latest
steps:
- name: Checkout
uses: actions/checkout@v3
with:
persist-credentials: false
fetch-depth: 0
- name: Set up Python
uses: actions/setup-python@v3
with:
python-version: 3.7
- name: Set up env
run: make -C docs FLAG="${{ env.FLAG }}" setupenv
- name: Build docs
run: make -C docs FLAG="${{ env.FLAG }}" multiversion
- name: Build redirects
run: make -C docs FLAG="${{ env.FLAG }}" redirects
- name: Deploy docs to GitHub Pages
run: ./docs/_utils/deploy.sh
if: (github.ref_name == 'master' && env.FLAG == 'opensource') || (github.ref_name == 'enterprise' && env.FLAG == 'enterprise')
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

29
.github/workflows/docs-pages@v2.yaml vendored Normal file
View File

@@ -0,0 +1,29 @@
name: "Docs / Publish"
on:
push:
branches:
- master
paths:
- "docs/**"
workflow_dispatch:
jobs:
release:
runs-on: ubuntu-latest
steps:
- name: Checkout
uses: actions/checkout@v2
with:
persist-credentials: false
fetch-depth: 0
- name: Set up Python
uses: actions/setup-python@v1
with:
python-version: 3.7
- name: Build docs
run: make -C docs multiversion
- name: Deploy
run: ./docs/_utils/deploy.sh
env:
GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

View File

@@ -1,32 +0,0 @@
name: "Docs / Build PR"
# For more information,
# see https://sphinx-theme.scylladb.com/stable/deployment/production.html#available-workflows
env:
FLAG: ${{ github.repository == 'scylladb/scylla-enterprise' && 'enterprise' || 'opensource' }}
on:
pull_request:
branches:
- master
- enterprise
paths:
- "docs/**"
jobs:
build:
runs-on: ubuntu-latest
steps:
- name: Checkout
uses: actions/checkout@v3
with:
persist-credentials: false
fetch-depth: 0
- name: Set up Python
uses: actions/setup-python@v3
with:
python-version: 3.7
- name: Set up env
run: make -C docs FLAG="${{ env.FLAG }}" setupenv
- name: Build docs
run: make -C docs FLAG="${{ env.FLAG }}" test

25
.github/workflows/docs-pr@v1.yaml vendored Normal file
View File

@@ -0,0 +1,25 @@
name: "Docs / Build PR"
on:
pull_request:
branches:
- master
paths:
- "docs/**"
jobs:
build:
name: Build
runs-on: ubuntu-latest
steps:
- name: Checkout
uses: actions/checkout@v2
with:
persist-credentials: false
fetch-depth: 0
- name: Set up Python
uses: actions/setup-python@v1
with:
python-version: 3.7
- name: Build docs
run: make -C docs test

3
.gitignore vendored
View File

@@ -22,7 +22,6 @@ resources
.pytest_cache
/expressions.tokens
tags
!db/tags/
testlog
test/*/*.reject
.vscode
@@ -30,5 +29,3 @@ docs/_build
docs/poetry.lock
compile_commands.json
.ccls-cache/
.mypy_cache
.envrc

9
.gitmodules vendored
View File

@@ -6,6 +6,12 @@
path = swagger-ui
url = ../scylla-swagger-ui
ignore = dirty
[submodule "libdeflate"]
path = libdeflate
url = ../libdeflate
[submodule "abseil"]
path = abseil
url = ../abseil-cpp
[submodule "scylla-jmx"]
path = tools/jmx
url = ../scylla-jmx
@@ -15,6 +21,3 @@
[submodule "scylla-python3"]
path = tools/python3
url = ../scylla-python3
[submodule "tools/cqlsh"]
path = tools/cqlsh
url = ../scylla-cqlsh

View File

@@ -1,3 +0,0 @@
Avi Kivity <avi@scylladb.com> Avi Kivity' via ScyllaDB development <scylladb-dev@googlegroups.com>
Raphael S. Carvalho <raphaelsc@scylladb.com> Raphael S. Carvalho' via ScyllaDB development <scylladb-dev@googlegroups.com>
Pavel Emelyanov <xemul@scylladb.com> Pavel Emelyanov' via ScyllaDB development <scylladb-dev@googlegroups.com>

View File

@@ -2,214 +2,794 @@ cmake_minimum_required(VERSION 3.18)
project(scylla)
include(CTest)
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")
endif()
list(APPEND CMAKE_MODULE_PATH
${CMAKE_CURRENT_SOURCE_DIR}/cmake
${CMAKE_CURRENT_SOURCE_DIR}/seastar/cmake)
if(CMAKE_BUILD_TYPE)
string(TOLOWER "${CMAKE_BUILD_TYPE}" BUILD_TYPE)
else()
set(BUILD_TYPE "release")
endif()
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)
else()
set(${arch} "" PARENT_SCOPE)
endif()
endfunction()
default_target_arch(target_arch)
if(target_arch)
set(target_arch_flag "-march=${target_arch}")
endif()
set(cxx_coro_flag)
if (CMAKE_CXX_COMPILER_ID MATCHES GNU)
set(cxx_coro_flag -fcoroutines)
endif()
set(CMAKE_BUILD_TYPE "${CMAKE_BUILD_TYPE}" CACHE
STRING "Choose the type of build." FORCE)
# Set the possible values of build type for cmake-gui
set_property(CACHE CMAKE_BUILD_TYPE PROPERTY STRINGS
"Debug" "Release" "Dev" "Sanitize")
string(TOUPPER "${CMAKE_BUILD_TYPE}" build_mode)
include(mode.${build_mode})
include(mode.common)
add_compile_definitions(
${Seastar_DEFINITIONS_${build_mode}}
FMT_DEPRECATED_OSTREAM)
include(limit_jobs)
# Configure Seastar compile options to align with Scylla
set(CMAKE_CXX_STANDARD "20" CACHE INTERNAL "")
set(CMAKE_CXX_EXTENSIONS ON CACHE INTERNAL "")
set(CMAKE_CXX_VISIBILITY_PRESET hidden)
set(Seastar_CXX_FLAGS ${cxx_coro_flag} ${target_arch_flag} CACHE INTERNAL "" FORCE)
set(Seastar_CXX_DIALECT gnu++20 CACHE INTERNAL "" FORCE)
set(Seastar_TESTING ON CACHE BOOL "" FORCE)
set(Seastar_API_LEVEL 6 CACHE STRING "" FORCE)
add_subdirectory(seastar)
add_subdirectory(abseil)
# Exclude absl::strerror from the default "all" target since it's not
# used in Scylla build and, moreover, makes use of deprecated glibc APIs,
# such as sys_nerr, which are not exposed from "stdio.h" since glibc 2.32,
# which happens to be the case for recent Fedora distribution versions.
#
# Need to use the internal "absl_strerror" target name instead of namespaced
# variant because `set_target_properties` does not understand the latter form,
# unfortunately.
set_target_properties(absl_strerror PROPERTIES EXCLUDE_FROM_ALL TRUE)
# System libraries dependencies
find_package(Boost REQUIRED
COMPONENTS filesystem program_options system thread regex unit_test_framework)
find_package(Boost COMPONENTS filesystem program_options system thread regex REQUIRED)
find_package(Lua REQUIRED)
find_package(ZLIB REQUIRED)
find_package(ICU COMPONENTS uc i18n REQUIRED)
find_package(absl COMPONENTS hash raw_hash_set REQUIRED)
find_package(libdeflate REQUIRED)
find_package(libxcrypt REQUIRED)
find_package(Snappy REQUIRED)
find_package(RapidJSON REQUIRED)
find_package(Thrift REQUIRED)
find_package(xxHash REQUIRED)
find_package(ICU COMPONENTS uc REQUIRED)
set(scylla_gen_build_dir "${CMAKE_BINARY_DIR}/gen")
file(MAKE_DIRECTORY "${scylla_gen_build_dir}")
set(scylla_build_dir "${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()
add_library(scylla-main STATIC)
target_sources(scylla-main
PRIVATE
# 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} ${outputs} PARENT_SCOPE)
endfunction()
scylla_generate_thrift(
TARGET scylla_thrift_gen_cassandra
VAR scylla_thrift_gen_cassandra_files
IN_FILE "${CMAKE_SOURCE_DIR}/interface/cassandra.thrift"
OUT_DIR ${scylla_gen_build_dir}
SERVICE Cassandra)
# Parse antlr3 grammar files and generate C++ sources
function(scylla_generate_antlr3)
set(one_value_args TARGET VAR IN_FILE OUT_DIR)
cmake_parse_arguments(args "" "${one_value_args}" "" ${ARGN})
get_filename_component(in_file_pure_name ${args_IN_FILE} NAME)
get_filename_component(stem ${in_file_pure_name} NAME_WE)
set(outputs
"${args_OUT_DIR}/${stem}Lexer.hpp"
"${args_OUT_DIR}/${stem}Lexer.cpp"
"${args_OUT_DIR}/${stem}Parser.hpp"
"${args_OUT_DIR}/${stem}Parser.cpp")
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}/${stem}Parser.cpp"
VERBATIM)
add_custom_target(${args_TARGET}
DEPENDS ${outputs})
set(${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_${grammar_file_name}_files
IN_FILE "${CMAKE_SOURCE_DIR}/${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 "${CMAKE_SOURCE_DIR}/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_${fname}_files
IN_FILE "${CMAKE_SOURCE_DIR}/${f}"
OUT_DIR "${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/hinted_handoff.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/raft.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_${idl_target}_files
IN_FILE "${CMAKE_SOURCE_DIR}/${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/conditions.cc
alternator/controller.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
client_data.cc
caching_options.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
compaction/compaction.cc
compaction/compaction_manager.cc
compaction/compaction_strategy.cc
compaction/leveled_compaction_strategy.cc
compaction/size_tiered_compaction_strategy.cc
compaction/time_window_compaction_strategy.cc
compress.cc
connection_notifier.cc
converting_mutation_partition_applier.cc
counters.cc
direct_failure_detector/failure_detector.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/expr/term_expr.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/prepare_context.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_service_level_statement.cc
cql3/statements/alter_table_statement.cc
cql3/statements/alter_type_statement.cc
cql3/statements/alter_view_statement.cc
cql3/statements/attach_service_level_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_aggregate_statement.cc
cql3/statements/create_function_statement.cc
cql3/statements/create_index_statement.cc
cql3/statements/create_keyspace_statement.cc
cql3/statements/create_service_level_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/detach_service_level_statement.cc
cql3/statements/drop_aggregate_statement.cc
cql3/statements/drop_function_statement.cc
cql3/statements/drop_index_statement.cc
cql3/statements/drop_keyspace_statement.cc
cql3/statements/drop_service_level_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_service_level_attachments_statement.cc
cql3/statements/list_service_level_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/service_level_statement.cc
cql3/statements/sl_prop_defs.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/util.cc
cql3/ut_name.cc
cql3/values.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/host_filter.cc
db/hints/manager.cc
db/hints/resource_manager.cc
db/hints/sync_point.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
db/virtual_table.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
generic_server.cc
debug.cc
gms/application_state.cc
gms/endpoint_state.cc
gms/failure_detector.cc
gms/feature_service.cc
gms/gossip_digest_ack2.cc
gms/gossip_digest_ack.cc
gms/gossip_digest_syn.cc
gms/gossiper.cc
gms/inet_address.cc
gms/versioned_value.cc
gms/version_generator.cc
hashers.cc
index/secondary_index.cc
index/secondary_index_manager.cc
init.cc
keys.cc
lister.cc
locator/abstract_replication_strategy.cc
locator/azure_snitch.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
lang/lua.cc
main.cc
memtable.cc
message/messaging_service.cc
multishard_mutation_query.cc
mutation.cc
mutation_fragment.cc
mutation_partition.cc
mutation_partition_serializer.cc
mutation_partition_view.cc
mutation_query.cc
mutation_reader.cc
mutation_writer/feed_writers.cc
mutation_writer/multishard_writer.cc
mutation_writer/partition_based_splitting_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.cc
query_ranges_to_vnodes.cc
query-result-set.cc
tombstone_gc_options.cc
tombstone_gc.cc
raft/fsm.cc
raft/log.cc
raft/raft.cc
raft/server.cc
raft/tracker.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
release.cc
repair/repair.cc
repair/row_level.cc
row_cache.cc
schema.cc
schema_mutations.cc
schema_registry.cc
serializer.cc
sstables_loader.cc
service/client_state.cc
service/migration_manager.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/qos/qos_common.cc
service/qos/service_level_controller.cc
service/qos/standard_service_level_distributed_data_accessor.cc
service/raft/raft_gossip_failure_detector.cc
service/raft/raft_group_registry.cc
service/raft/raft_rpc.cc
service/raft/raft_sys_table_storage.cc
service/raft/schema_raft_state_machine.cc
service/storage_proxy.cc
service/storage_service.cc
sstables/compress.cc
sstables/integrity_checked_file_impl.cc
sstables/kl/reader.cc
sstables/metadata_collector.cc
sstables/m_format_read_helpers.cc
sstables/mx/reader.cc
sstables/mx/writer.cc
sstables/prepended_input_stream.cc
sstables/random_access_reader.cc
sstables/sstable_directory.cc
sstables/sstable_mutation_reader.cc
sstables/sstables.cc
sstables/sstable_set.cc
sstables/sstables_manager.cc
sstables/sstable_version.cc
sstables/writer.cc
streaming/consumer.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
tasks/task_manager.cc
thrift/controller.cc
thrift/handler.cc
thrift/server.cc
thrift/thrift_validation.cc
timeout_config.cc
tools/scylla-sstable-index.cc
tools/scylla-types.cc
tracing/traced_file.cc
tracing/trace_keyspace_helper.cc
tracing/trace_state.cc
tracing/tracing_backend_registry.cc
tracing/tracing.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/arch/powerpc/crc32-vpmsum/crc32_wrapper.cc
utils/array-search.cc
utils/ascii.cc
utils/base64.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/gz/gen_crc_combine_table.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
utils/UUID_gen.cc
validation.cc
vint-serialization.cc
zstd.cc)
target_link_libraries(scylla-main
PRIVATE
db
absl::hash
absl::raw_hash_set
Seastar::seastar
Snappy::snappy
systemd
ZLIB::ZLIB)
add_subdirectory(api)
add_subdirectory(alternator)
add_subdirectory(db)
add_subdirectory(auth)
add_subdirectory(cdc)
add_subdirectory(compaction)
add_subdirectory(cql3)
add_subdirectory(data_dictionary)
add_subdirectory(dht)
add_subdirectory(gms)
add_subdirectory(idl)
add_subdirectory(index)
add_subdirectory(interface)
add_subdirectory(lang)
add_subdirectory(locator)
add_subdirectory(mutation)
add_subdirectory(mutation_writer)
add_subdirectory(readers)
add_subdirectory(redis)
add_subdirectory(replica)
add_subdirectory(raft)
add_subdirectory(repair)
add_subdirectory(rust)
add_subdirectory(schema)
add_subdirectory(service)
add_subdirectory(sstables)
add_subdirectory(streaming)
add_subdirectory(test)
add_subdirectory(thrift)
add_subdirectory(tools)
add_subdirectory(tracing)
add_subdirectory(transport)
add_subdirectory(types)
add_subdirectory(utils)
include(add_version_library)
add_version_library(scylla_version
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}")
add_executable(scylla
main.cc)
target_link_libraries(scylla PRIVATE
scylla-main
api
auth
alternator
db
cdc
compaction
cql3
data_dictionary
dht
gms
idl
index
lang
locator
mutation
mutation_writer
raft
readers
redis
repair
replica
schema
scylla_version
service
sstables
streaming
test-perf
thrift
tools
tracing
transport
types
utils)
target_link_libraries(Boost::regex
INTERFACE
ICU::i18n
ICU::uc)
${scylla_sources}
${scylla_gen_sources})
target_link_libraries(scylla PRIVATE
seastar
Boost::program_options)
# Force SHA1 build-id generation
set(default_linker_flags "-Wl,--build-id=sha1")
include(CheckLinkerFlag)
set(Scylla_USE_LINKER
""
CACHE
STRING
"Use specified linker instead of the default one")
if(Scylla_USE_LINKER)
set(linkers "${Scylla_USE_LINKER}")
else()
set(linkers "lld" "gold")
endif()
foreach(linker ${linkers})
set(linker_flag "-fuse-ld=${linker}")
check_linker_flag(CXX ${linker_flag} "CXX_LINKER_HAVE_${linker}")
if(CXX_LINKER_HAVE_${linker})
string(APPEND default_linker_flags " ${linker_flag}")
break()
elseif(Scylla_USE_LINKER)
message(FATAL_ERROR "${Scylla_USE_LINKER} is not supported.")
endif()
endforeach()
set(CMAKE_EXE_LINKER_FLAGS "${default_linker_flags}" CACHE INTERNAL "")
# 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)
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
target_compile_options(scylla PRIVATE
-std=gnu++20
${cxx_coro_flag}
${target_arch_flag})
# Hacks needed to expose internal APIs for xxhash dependencies
target_compile_definitions(scylla PRIVATE XXH_PRIVATE_API HAVE_LZ4_COMPRESS_DEFAULT)
target_include_directories(scylla PRIVATE
"${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
${cxx_coro_flag}
${target_arch_flag})
add_dependencies(scylla crc_combine_table)
# Generate an additional source file at build time that is needed for Scylla compilation
add_custom_command(OUTPUT "${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 ${CMAKE_SOURCE_DIR}/SCYLLA-VERSION-GEN --output-dir "${CMAKE_BINARY_DIR}/gen" RESULT_VARIABLE scylla_version_gen_res)
if(scylla_version_gen_res)
message(SEND_ERROR "Version file generation failed. Return code: ${scylla_version_gen_res}")
endif()
file(READ "${CMAKE_BINARY_DIR}/gen/SCYLLA-VERSION-FILE" scylla_version)
string(STRIP "${scylla_version}" scylla_version)
file(READ "${CMAKE_BINARY_DIR}/gen/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 "${CMAKE_SOURCE_DIR}/libdeflate"
BUILD_DIR=../build/${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)

View File

@@ -2,7 +2,7 @@
## Asking questions or requesting help
Use the [ScyllaDB Community Forum](https://forum.scylladb.com) or the [Slack workspace](http://slack.scylladb.com) for general questions and help.
Use the [Scylla Users mailing list](https://groups.google.com/g/scylladb-users) or the [Slack workspace](http://slack.scylladb.com) for general questions and help.
Join the [Scylla Developers mailing list](https://groups.google.com/g/scylladb-dev) for deeper technical discussions and to discuss your ideas for contributions.
@@ -18,5 +18,3 @@ If you need help formatting or sending patches, [check out these instructions](h
The Scylla C++ source code uses the [Seastar coding style](https://github.com/scylladb/seastar/blob/master/coding-style.md) so please adhere to that in your patches. Note that Scylla code is written with `using namespace seastar`, so should not explicitly add the `seastar::` prefix to Seastar symbols. You will usually not need to add `using namespace seastar` to new source files, because most Scylla header files have `#include "seastarx.hh"`, which does this.
Header files in Scylla must be self-contained, i.e., each can be included without having to include specific other headers first. To verify that your change did not break this property, run `ninja dev-headers`. If you added or removed header files, you must `touch configure.py` first - this will cause `configure.py` to be automatically re-run to generate a fresh list of header files.
For more criteria on what reviewers consider good code, see the [review checklist](https://github.com/scylladb/scylla/blob/master/docs/dev/review-checklist.md).

View File

@@ -195,7 +195,7 @@ $ # Edit configuration options as appropriate
$ SCYLLA_HOME=$HOME/scylla build/release/scylla
```
The `scylla.yaml` file in the repository by default writes all database data to `/var/lib/scylla`, which likely requires root access. Change the `data_file_directories`, `commitlog_directory` and `schema_commitlog_directory` fields as appropriate.
The `scylla.yaml` file in the repository by default writes all database data to `/var/lib/scylla`, which likely requires root access. Change the `data_file_directories` and `commitlog_directory` fields as appropriate.
Scylla has a number of requirements for the file-system and operating system to operate ideally and at peak performance. However, during development, these requirements can be relaxed with the `--developer-mode` flag.
@@ -383,40 +383,6 @@ Open the link printed at the end. Be horrified. Go and write more tests.
For more details see `./scripts/coverage.py --help`.
### Resolving stack backtraces
Scylla may print stack backtraces to the log for several reasons.
For example:
- When aborting (e.g. due to assertion failure, internal error, or segfault)
- When detecting seastar reactor stalls (where a seastar task runs for a long time without yielding the cpu to other tasks on that shard)
The backtraces contain code pointers so they are not very helpful without resolving into code locations.
To resolve the backtraces, one needs the scylla relocatable package that contains the scylla binary (with debug information),
as well as the dynamic libraries it is linked against.
Builds from our automated build system are uploaded to the cloud
and can be searched on http://backtrace.scylladb.com/
Make sure you have the scylla server exact `build-id` to locate
its respective relocatable package, required for decoding backtraces it prints.
The build-id is printed to the system log when scylla starts.
It can also be found by executing `scylla --build-id`, or
by using the `file` utility, for example:
```
$ scylla --build-id
4cba12e6eb290a406bfa4930918db23941fd4be3
$ file scylla
scylla: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically linked, interpreter /////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////////lib64/ld-linux-x86-64.so.2, for GNU/Linux 3.2.0, BuildID[sha1]=4cba12e6eb290a406bfa4930918db23941fd4be3, with debug_info, not stripped, too many notes (256)
```
To find the build-id of a coredump, use the `eu-unstrip` utility as follows:
```
$ eu-unstrip -n --core <coredump> | awk '/scylla$/ { s=$2; sub(/@.*$/, "", s); print s; exit(0); }'
4cba12e6eb290a406bfa4930918db23941fd4be3
```
### Core dump debugging
See [debugging.md](docs/dev/debugging.md).
See [debugging.md](debugging.md).

View File

@@ -30,9 +30,9 @@ requirements - you just need to meet the frozen toolchain's prerequisites
Building Scylla with the frozen toolchain `dbuild` is as easy as:
```bash
$ git submodule update --init --force --recursive
$ ./tools/toolchain/dbuild ./configure.py
$ ./tools/toolchain/dbuild ninja build/release/scylla
$ git submodule update --init --force --recursive
$ ./tools/toolchain/dbuild ./configure.py
$ ./tools/toolchain/dbuild ninja build/release/scylla
```
For further information, please see:
@@ -42,7 +42,7 @@ For further information, please see:
* [Docker image build documentation] for information on how to build Docker images.
[developer documentation]: HACKING.md
[build documentation]: docs/dev/building.md
[build documentation]: docs/guides/building.md
[docker image build documentation]: dist/docker/debian/README.md
## Running Scylla
@@ -60,12 +60,12 @@ Please note that you need to run Scylla with `dbuild` if you built it with the f
For more run options, run:
```bash
$ ./tools/toolchain/dbuild ./build/release/scylla --help
$ ./tools/toolchain/dbuild ./build/release/scylla --help
```
## Testing
See [test.py manual](docs/dev/testing.md).
See [test.py manual](docs/guides/testing.md).
## Scylla APIs and compatibility
By default, Scylla is compatible with Apache Cassandra and its APIs - CQL and
@@ -78,7 +78,7 @@ and the current compatibility of this feature as well as Scylla-specific extensi
## Documentation
Documentation can be found [here](docs/dev/README.md).
Documentation can be found [here](https://scylla.docs.scylladb.com).
Seastar documentation can be found [here](http://docs.seastar.io/master/index.html).
User documentation can be found [here](https://docs.scylladb.com/).
@@ -100,10 +100,10 @@ If you are a developer working on Scylla, please read the [developer guidelines]
## Contact
* The [community forum] and [Slack channel] are for users to discuss configuration, management, and operations of the ScyllaDB open source.
* 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.
[Community forum]: https://forum.scylladb.com/
[Users mailing list]: https://groups.google.com/forum/#!forum/scylladb-users
[Slack channel]: http://slack.scylladb.com/

View File

@@ -1,12 +1,11 @@
#!/bin/sh
USAGE=$(cat <<-END
Usage: $(basename "$0") [-h|--help] [-o|--output-dir PATH] [--date-stamp DATE] -- generate Scylla version and build information files.
Usage: $(basename "$0") [-h|--help] [-o|--output-dir PATH] -- generate Scylla version and build information files.
Options:
-h|--help show this help message.
-o|--output-dir PATH specify destination path at which the version files are to be created.
-d|--date-stamp DATE manually set date for release parameter
By default, the script will attempt to parse 'version' file
in the current directory, which should contain a string of
@@ -32,9 +31,7 @@ using '-o PATH' option.
END
)
DATE=""
while [ $# -gt 0 ]; do
while [[ $# -gt 0 ]]; do
opt="$1"
case $opt in
-h|--help)
@@ -46,11 +43,6 @@ while [ $# -gt 0 ]; do
shift
shift
;;
--date-stamp)
DATE="$2"
shift
shift
;;
*)
echo "Unexpected argument found: $1"
echo
@@ -66,33 +58,24 @@ if [ -z "$OUTPUT_DIR" ]; then
OUTPUT_DIR="$SCRIPT_DIR/build"
fi
if [ -z "$DATE" ]; then
DATE=$(date --utc +%Y%m%d)
fi
# Default scylla product/version tags
PRODUCT=scylla
VERSION=5.3.0-rc1
VERSION=4.6.11
if test -f version
then
SCYLLA_VERSION=$(cat version | awk -F'-' '{print $1}')
SCYLLA_RELEASE=$(cat version | awk -F'-' '{print $2}')
else
DATE=$(date +%Y%m%d)
GIT_COMMIT=$(git -C "$SCRIPT_DIR" log --pretty=format:'%h' -n 1)
SCYLLA_VERSION=$VERSION
if [ -z "$SCYLLA_RELEASE" ]; then
DATE=$(date --utc +%Y%m%d)
GIT_COMMIT=$(git -C "$SCRIPT_DIR" log --pretty=format:'%h' -n 1 --abbrev=12)
# For custom package builds, replace "0" with "counter.your_name",
# where counter starts at 1 and increments for successive versions.
# This ensures that the package manager will select your custom
# package over the standard release.
SCYLLA_BUILD=0
SCYLLA_RELEASE=$SCYLLA_BUILD.$DATE.$GIT_COMMIT
elif [ -f "$OUTPUT_DIR/SCYLLA-RELEASE-FILE" ]; then
echo "setting SCYLLA_RELEASE only makes sense in clean builds" 1>&2
exit 1
fi
# For custom package builds, replace "0" with "counter.your_name",
# where counter starts at 1 and increments for successive versions.
# This ensures that the package manager will select your custom
# package over the standard release.
SCYLLA_BUILD=0
SCYLLA_RELEASE=$SCYLLA_BUILD.$DATE.$GIT_COMMIT
fi
if [ -f "$OUTPUT_DIR/SCYLLA-RELEASE-FILE" ]; then

1
abseil Submodule

Submodule abseil added at f70eadadd7

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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"

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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

View File

@@ -1,30 +0,0 @@
include(generate_cql_grammar)
generate_cql_grammar(
GRAMMAR expressions.g
SOURCES cql_grammar_srcs)
add_library(alternator STATIC)
target_sources(alternator
PRIVATE
controller.cc
server.cc
executor.cc
stats.cc
serialization.cc
expressions.cc
conditions.cc
auth.cc
streams.cc
ttl.cc
${cql_grammar_srcs})
target_include_directories(alternator
PUBLIC
${CMAKE_SOURCE_DIR}
${CMAKE_BINARY_DIR}
PRIVATE
${RAPIDJSON_INCLUDE_DIRS})
target_link_libraries(alternator
cql3
idl
Seastar::seastar
xxHash::xxhash)

View File

@@ -3,13 +3,28 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "hashers.hh"
#include "bytes.hh"
#include "alternator/auth.hh"
#include <fmt/format.h>
@@ -19,6 +34,7 @@
#include "service/storage_proxy.hh"
#include "alternator/executor.hh"
#include "cql3/selection/selection.hh"
#include "database.hh"
#include "query-result-set.hh"
#include "cql3/result_set.hh"
#include <seastar/core/coroutine.hh>
@@ -27,35 +43,127 @@ 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 apply_sha256(const std::vector<temporary_buffer<char>>& msg) {
sha256_hasher hasher;
for (const temporary_buffer<char>& buf : msg) {
hasher.update(buf.get(), buf.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,
const std::vector<temporary_buffer<char>>& 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(service::storage_proxy& proxy, std::string username) {
schema_ptr schema = proxy.data_dictionary().find_schema("system_auth", "roles");
schema_ptr schema = proxy.get_db().local().find_schema("system_auth", "roles");
partition_key pk = partition_key::from_single_value(*schema, utf8_type->decompose(username));
dht::partition_range_vector partition_ranges{dht::partition_range(dht::decorate_key(*schema, pk))};
std::vector<query::clustering_range> bounds{query::clustering_range::make_open_ended_both_sides()};
const column_definition* salted_hash_col = schema->get_column_definition(bytes("salted_hash"));
if (!salted_hash_col) {
co_await coroutine::return_exception(api_error::unrecognized_client(format("Credentials cannot be fetched for: {}", username)));
co_return coroutine::make_exception(api_error::unrecognized_client(format("Credentials cannot be fetched for: {}", username)));
}
auto selection = cql3::selection::selection::for_columns(schema, {salted_hash_col});
auto partition_slice = query::partition_slice(std::move(bounds), {}, query::column_id_vector{salted_hash_col->id}, selection->get_query_options());
auto command = ::make_lw_shared<query::read_command>(schema->id(), schema->version(), partition_slice,
proxy.get_max_result_size(partition_slice), query::tombstone_limit(proxy.get_tombstone_limit()));
auto command = ::make_lw_shared<query::read_command>(schema->id(), schema->version(), partition_slice, proxy.get_max_result_size(partition_slice));
auto cl = auth::password_authenticator::consistency_for_user(username);
service::client_state client_state{service::client_state::internal_tag()};
service::storage_proxy::coordinator_query_result qr = co_await proxy.query(schema, std::move(command), std::move(partition_ranges), cl,
service::storage_proxy::coordinator_query_options(executor::default_timeout(), empty_service_permit(), client_state));
cql3::selection::result_set_builder builder(*selection, gc_clock::now());
cql3::selection::result_set_builder builder(*selection, gc_clock::now(), cql_serialization_format::latest());
query::result_view::consume(*qr.query_result, partition_slice, cql3::selection::result_set_builder::visitor(builder, *schema, *selection));
auto result_set = builder.build();
if (result_set->empty()) {
co_await coroutine::return_exception(api_error::unrecognized_client(format("User not found: {}", username)));
co_return coroutine::make_exception(api_error::unrecognized_client(format("User not found: {}", username)));
}
const managed_bytes_opt& salted_hash = result_set->rows().front().front(); // We only asked for 1 row and 1 column
const bytes_opt& salted_hash = result_set->rows().front().front(); // We only asked for 1 row and 1 column
if (!salted_hash) {
co_await coroutine::return_exception(api_error::unrecognized_client(format("No password found for user: {}", username)));
co_return coroutine::make_exception(api_error::unrecognized_client(format("No password found for user: {}", username)));
}
co_return value_cast<sstring>(utf8_type->deserialize(*salted_hash));
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -20,7 +33,13 @@ class storage_proxy;
namespace alternator {
using key_cache = utils::loading_cache<std::string, std::string, 1>;
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,
const std::vector<temporary_buffer<char>>& body_content, std::string_view region, std::string_view service, std::string_view query_string);
future<std::string> get_key_from_roles(service::storage_proxy& proxy, std::string username);

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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>
@@ -232,14 +245,7 @@ bool check_BEGINS_WITH(const rjson::value* v1, const rjson::value& v2,
if (it2->name == "S") {
return rjson::to_string_view(it1->value).starts_with(rjson::to_string_view(it2->value));
} else /* it2->name == "B" */ {
try {
return base64_begins_with(rjson::to_string_view(it1->value), rjson::to_string_view(it2->value));
} catch(std::invalid_argument&) {
// determine if any of the malformed values is from query and raise an exception if so
unwrap_bytes(it1->value, v1_from_query);
unwrap_bytes(it2->value, v2_from_query);
return false;
}
return base64_begins_with(rjson::to_string_view(it1->value), rjson::to_string_view(it2->value));
}
}
@@ -248,7 +254,7 @@ static bool is_set_of(const rjson::value& type1, const rjson::value& type2) {
}
// Check if two JSON-encoded values match with the CONTAINS relation
bool check_CONTAINS(const rjson::value* v1, const rjson::value& v2, bool v1_from_query, bool v2_from_query) {
bool check_CONTAINS(const rjson::value* v1, const rjson::value& v2) {
if (!v1) {
return false;
}
@@ -257,12 +263,7 @@ bool check_CONTAINS(const rjson::value* v1, const rjson::value& v2, bool v1_from
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") {
auto d_kv1 = unwrap_bytes(kv1.value, v1_from_query);
auto d_kv2 = unwrap_bytes(kv2.value, v2_from_query);
if (!d_kv1 || !d_kv2) {
return false;
}
return d_kv1->find(*d_kv2) != bytes::npos;
return rjson::base64_decode(kv1.value).find(rjson::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) {
@@ -285,11 +286,11 @@ bool check_CONTAINS(const rjson::value* v1, const rjson::value& v2, bool v1_from
}
// 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, bool v1_from_query, bool v2_from_query) {
static bool check_NOT_CONTAINS(const rjson::value* v1, const rjson::value& v2) {
if (!v1) {
return false;
}
return !check_CONTAINS(v1, v2, v1_from_query, v2_from_query);
return !check_CONTAINS(v1, v2);
}
// Check if a JSON-encoded value equals any element of an array, which must have at least one element.
@@ -386,12 +387,7 @@ bool check_compare(const rjson::value* v1, const rjson::value& v2, const Compara
std::string_view(kv2.value.GetString(), kv2.value.GetStringLength()));
}
if (kv1.name == "B") {
auto d_kv1 = unwrap_bytes(kv1.value, v1_from_query);
auto d_kv2 = unwrap_bytes(kv2.value, v2_from_query);
if(!d_kv1 || !d_kv2) {
return false;
}
return cmp(*d_kv1, *d_kv2);
return cmp(rjson::base64_decode(kv1.value), rjson::base64_decode(kv2.value));
}
// cannot reach here, as check_comparable_type() verifies the type is one
// of the above options.
@@ -481,13 +477,7 @@ static bool check_BETWEEN(const rjson::value* v, const rjson::value& lb, const r
bounds_from_query);
}
if (kv_v.name == "B") {
auto d_kv_v = unwrap_bytes(kv_v.value, v_from_query);
auto d_kv_lb = unwrap_bytes(kv_lb.value, lb_from_query);
auto d_kv_ub = unwrap_bytes(kv_ub.value, ub_from_query);
if(!d_kv_v || !d_kv_lb || !d_kv_ub) {
return false;
}
return check_BETWEEN(*d_kv_v, *d_kv_lb, *d_kv_ub, bounds_from_query);
return check_BETWEEN(rjson::base64_decode(kv_v.value), rjson::base64_decode(kv_lb.value), rjson::base64_decode(kv_ub.value), bounds_from_query);
}
if (v_from_query) {
throw api_error::validation(
@@ -580,7 +570,7 @@ static bool verify_expected_one(const rjson::value& condition, const rjson::valu
format("CONTAINS operator requires a single AttributeValue of type String, Number, or Binary, "
"got {} instead", argtype));
}
return check_CONTAINS(got, arg, false, true);
return check_CONTAINS(got, arg);
}
case comparison_operator_type::NOT_CONTAINS:
{
@@ -594,7 +584,7 @@ static bool verify_expected_one(const rjson::value& condition, const rjson::valu
format("CONTAINS operator requires a single AttributeValue of type String, Number, or Binary, "
"got {} instead", argtype));
}
return check_NOT_CONTAINS(got, arg, false, true);
return check_NOT_CONTAINS(got, arg);
}
}
throw std::logic_error(format("Internal error: corrupted operator enum: {}", int(op)));

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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/>.
*/
/*
@@ -38,7 +51,7 @@ 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 v1_from_query, bool v2_from_query);
bool check_CONTAINS(const rjson::value* v1, const rjson::value& v2);
bool check_BEGINS_WITH(const rjson::value* v1, const rjson::value& v2, bool v1_from_query, bool v2_from_query);
bool verify_condition_expression(

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 <seastar/net/dns.hh>
@@ -14,8 +27,6 @@
#include "db/config.hh"
#include "cdc/generation_service.hh"
#include "service/memory_limiter.hh"
#include "auth/service.hh"
#include "service/qos/service_level_controller.hh"
using namespace seastar;
@@ -30,8 +41,6 @@ controller::controller(
sharded<db::system_distributed_keyspace>& sys_dist_ks,
sharded<cdc::generation_service>& cdc_gen_svc,
sharded<service::memory_limiter>& memory_limiter,
sharded<auth::service>& auth_service,
sharded<qos::service_level_controller>& sl_controller,
const db::config& config)
: _gossiper(gossiper)
, _proxy(proxy)
@@ -39,32 +48,12 @@ controller::controller(
, _sys_dist_ks(sys_dist_ks)
, _cdc_gen_svc(cdc_gen_svc)
, _memory_limiter(memory_limiter)
, _auth_service(auth_service)
, _sl_controller(sl_controller)
, _config(config)
{
}
sstring controller::name() const {
return "alternator";
}
sstring controller::protocol() const {
return "dynamodb";
}
sstring controller::protocol_version() const {
return version;
}
std::vector<socket_address> controller::listen_addresses() const {
return _listen_addresses;
}
future<> controller::start_server() {
future<> controller::start() {
return seastar::async([this] {
_listen_addresses.clear();
auto preferred = _config.listen_interface_prefer_ipv6() ? std::make_optional(net::inet_address::family::INET6) : std::nullopt;
auto family = _config.enable_ipv6_dns_lookup() || preferred ? std::nullopt : std::make_optional(net::inet_address::family::INET);
@@ -76,32 +65,27 @@ future<> controller::start_server() {
_ssg = create_smp_service_group(c).get0();
rmw_operation::set_default_write_isolation(_config.alternator_write_isolation());
executor::set_default_timeout(std::chrono::milliseconds(_config.alternator_timeout_in_ms()));
net::inet_address addr = utils::resolve(_config.alternator_address, family).get0();
net::inet_address addr;
try {
addr = net::dns::get_host_by_name(_config.alternator_address(), family).get0().addr_list.front();
} catch (...) {
std::throw_with_nested(std::runtime_error(fmt::format("Unable to resolve alternator_address {}", _config.alternator_address())));
}
auto get_cdc_metadata = [] (cdc::generation_service& svc) { return std::ref(svc.get_cdc_metadata()); };
auto get_timeout_in_ms = [] (const db::config& cfg) -> utils::updateable_value<uint32_t> {
return cfg.alternator_timeout_in_ms;
};
_executor.start(std::ref(_gossiper), std::ref(_proxy), std::ref(_mm), std::ref(_sys_dist_ks),
sharded_parameter(get_cdc_metadata, std::ref(_cdc_gen_svc)), _ssg.value(),
sharded_parameter(get_timeout_in_ms, std::ref(_config))).get();
_server.start(std::ref(_executor), std::ref(_proxy), std::ref(_gossiper), std::ref(_auth_service), std::ref(_sl_controller)).get();
// Note: from this point on, if start_server() throws for any reason,
// it must first call stop_server() to stop the executor and server
// services we just started - or Scylla will cause an assertion
// failure when the controller object is destroyed in the exception
// unwinding.
_executor.start(std::ref(_gossiper), std::ref(_proxy), std::ref(_mm), std::ref(_sys_dist_ks), sharded_parameter(get_cdc_metadata, std::ref(_cdc_gen_svc)), _ssg.value()).get();
_server.start(std::ref(_executor), std::ref(_proxy), std::ref(_gossiper)).get();
std::optional<uint16_t> alternator_port;
if (_config.alternator_port()) {
alternator_port = _config.alternator_port();
_listen_addresses.push_back({addr, *alternator_port});
}
std::optional<uint16_t> alternator_https_port;
std::optional<tls::credentials_builder> creds;
if (_config.alternator_https_port()) {
alternator_https_port = _config.alternator_https_port();
_listen_addresses.push_back({addr, *alternator_https_port});
creds.emplace();
auto opts = _config.alternator_encryption_options();
if (opts.empty()) {
@@ -118,13 +102,7 @@ future<> controller::start_server() {
}
opts.erase("require_client_auth");
opts.erase("truststore");
try {
utils::configure_tls_creds_builder(creds.value(), std::move(opts)).get();
} catch(...) {
logger.error("Failed to set up Alternator TLS credentials: {}", std::current_exception());
stop_server().get();
std::throw_with_nested(std::runtime_error("Failed to set up Alternator TLS credentials"));
}
utils::configure_tls_creds_builder(creds.value(), std::move(opts)).get();
}
bool alternator_enforce_authorization = _config.alternator_enforce_authorization();
_server.invoke_on_all(
@@ -132,10 +110,6 @@ future<> controller::start_server() {
return server.init(addr, alternator_port, alternator_https_port, creds, alternator_enforce_authorization,
&_memory_limiter.local().get_semaphore(),
_config.max_concurrent_requests_per_shard);
}).handle_exception([this, addr, alternator_port, alternator_https_port] (std::exception_ptr ep) {
logger.error("Failed to set up Alternator HTTP server on {} port {}, TLS port {}: {}",
addr, alternator_port ? std::to_string(*alternator_port) : "OFF", alternator_https_port ? std::to_string(*alternator_https_port) : "OFF", ep);
return stop_server().then([ep = std::move(ep)] { return make_exception_future<>(ep); });
}).then([addr, alternator_port, alternator_https_port] {
logger.info("Alternator server listening on {}, HTTP port {}, HTTPS port {}",
addr, alternator_port ? std::to_string(*alternator_port) : "OFF", alternator_https_port ? std::to_string(*alternator_https_port) : "OFF");
@@ -143,20 +117,12 @@ future<> controller::start_server() {
});
}
future<> controller::stop_server() {
future<> controller::stop() {
return seastar::async([this] {
if (!_ssg) {
return;
}
_server.stop().get();
_executor.stop().get();
_listen_addresses.clear();
destroy_smp_service_group(_ssg.value()).get();
});
}
future<> controller::request_stop_server() {
return stop_server();
}
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -11,8 +24,6 @@
#include <seastar/core/sharded.hh>
#include <seastar/core/smp.hh>
#include "protocol_server.hh"
namespace service {
class storage_proxy;
class migration_manager;
@@ -34,38 +45,22 @@ class gossiper;
}
namespace auth {
class service;
}
namespace qos {
class service_level_controller;
}
namespace alternator {
// This is the official DynamoDB API version.
// It represents the last major reorganization of that API, and all the features
// that were added since did NOT increment this version string.
constexpr const char* version = "2012-08-10";
using namespace seastar;
class executor;
class server;
class controller : public protocol_server {
class controller {
sharded<gms::gossiper>& _gossiper;
sharded<service::storage_proxy>& _proxy;
sharded<service::migration_manager>& _mm;
sharded<db::system_distributed_keyspace>& _sys_dist_ks;
sharded<cdc::generation_service>& _cdc_gen_svc;
sharded<service::memory_limiter>& _memory_limiter;
sharded<auth::service>& _auth_service;
sharded<qos::service_level_controller>& _sl_controller;
const db::config& _config;
std::vector<socket_address> _listen_addresses;
sharded<executor> _executor;
sharded<server> _server;
std::optional<smp_service_group> _ssg;
@@ -78,17 +73,10 @@ public:
sharded<db::system_distributed_keyspace>& sys_dist_ks,
sharded<cdc::generation_service>& cdc_gen_svc,
sharded<service::memory_limiter>& memory_limiter,
sharded<auth::service>& auth_service,
sharded<qos::service_level_controller>& sl_controller,
const db::config& config);
virtual sstring name() const override;
virtual sstring protocol() const override;
virtual sstring protocol_version() const override;
virtual std::vector<socket_address> listen_addresses() const override;
virtual future<> start_server() override;
virtual future<> stop_server() override;
virtual future<> request_stop_server() override;
future<> start();
future<> stop();
};
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -23,7 +36,7 @@ namespace alternator {
// api_error into a JSON object, and that is returned to the user.
class api_error final : public std::exception {
public:
using status_type = http::reply::status_type;
using status_type = httpd::reply::status_type;
status_type _http_code;
std::string _type;
std::string _msg;
@@ -70,14 +83,8 @@ public:
static api_error request_limit_exceeded(std::string msg) {
return api_error("RequestLimitExceeded", std::move(msg));
}
static api_error serialization(std::string msg) {
return api_error("SerializationException", std::move(msg));
}
static api_error table_not_found(std::string msg) {
return api_error("TableNotFoundException", std::move(msg));
}
static api_error internal(std::string msg) {
return api_error("InternalServerError", std::move(msg), http::reply::status_type::internal_server_error);
return api_error("InternalServerError", std::move(msg), reply::status_type::internal_server_error);
}
// Provide the "std::exception" interface, to make it easier to print this

File diff suppressed because it is too large Load Diff

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -22,7 +35,6 @@
#include "alternator/error.hh"
#include "stats.hh"
#include "utils/rjson.hh"
#include "utils/updateable_value.hh"
namespace db {
class system_distributed_keyspace;
@@ -61,16 +73,6 @@ public:
explicit make_jsonable(rjson::value&& value);
std::string to_json() const override;
};
/**
* Make return type for serializing the object "streamed",
* i.e. direct to HTTP output stream. Note: only useful for
* (very) large objects as there are overhead issues with this
* as well, but for massive lists of return objects this can
* help avoid large allocations/many re-allocs
*/
json::json_return_type make_streamed(rjson::value&&);
struct json_string : public json::jsonable {
std::string _value;
public:
@@ -82,10 +84,9 @@ namespace parsed {
class path;
};
const std::map<sstring, sstring>& get_tags_of_table(schema_ptr schema);
future<> update_tags(service::migration_manager& mm, schema_ptr schema, std::map<sstring, sstring>&& tags_map);
schema_ptr get_table(service::storage_proxy& proxy, const rjson::value& request);
bool is_alternator_keyspace(const sstring& ks_name);
// Wraps the db::get_tags_of_table and throws if the table is missing the tags extension.
const std::map<sstring, sstring>& get_tags_of_table_or_throw(schema_ptr schema);
// An attribute_path_map object is used to hold data for various attributes
// paths (parsed::path) in a hierarchy of attribute paths. Each attribute path
@@ -145,11 +146,6 @@ template<typename T>
using attribute_path_map = std::unordered_map<std::string, attribute_path_map_node<T>>;
using attrs_to_get_node = attribute_path_map_node<std::monostate>;
// attrs_to_get lists which top-level attribute are needed, and possibly also
// which part of the top-level attribute is really needed (when nested
// attribute paths appeared in the query).
// Most code actually uses optional<attrs_to_get>. There, a disengaged
// optional means we should get all attributes, not specific ones.
using attrs_to_get = attribute_path_map<std::monostate>;
@@ -171,16 +167,8 @@ public:
static constexpr auto KEYSPACE_NAME_PREFIX = "alternator_";
static constexpr std::string_view INTERNAL_TABLE_PREFIX = ".scylla.alternator.";
executor(gms::gossiper& gossiper,
service::storage_proxy& proxy,
service::migration_manager& mm,
db::system_distributed_keyspace& sdks,
cdc::metadata& cdc_metadata,
smp_service_group ssg,
utils::updateable_value<uint32_t> default_timeout_in_ms)
: _gossiper(gossiper), _proxy(proxy), _mm(mm), _sdks(sdks), _cdc_metadata(cdc_metadata), _ssg(ssg) {
s_default_timeout_in_ms = std::move(default_timeout_in_ms);
}
executor(gms::gossiper& gossiper, service::storage_proxy& proxy, service::migration_manager& mm, db::system_distributed_keyspace& sdks, cdc::metadata& cdc_metadata, smp_service_group ssg)
: _gossiper(gossiper), _proxy(proxy), _mm(mm), _sdks(sdks), _cdc_metadata(cdc_metadata), _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);
@@ -205,61 +193,44 @@ public:
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<request_return_type> describe_continuous_backups(client_state& client_state, service_permit permit, rjson::value request);
future<> start();
future<> stop() {
// disconnect from the value source, but keep the value unchanged.
s_default_timeout_in_ms = utils::updateable_value<uint32_t>{s_default_timeout_in_ms()};
return make_ready_future<>();
}
future<> stop() { return make_ready_future<>(); }
future<> create_keyspace(std::string_view keyspace_name);
static sstring table_name(const schema&);
static db::timeout_clock::time_point default_timeout();
static void set_default_timeout(db::timeout_clock::duration timeout);
private:
static thread_local utils::updateable_value<uint32_t> s_default_timeout_in_ms;
static db::timeout_clock::duration s_default_timeout;
public:
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:
public:
static std::optional<rjson::value> describe_single_item(schema_ptr,
const query::partition_slice&,
const cql3::selection::selection&,
const query::result&,
const std::optional<attrs_to_get>&);
static future<std::vector<rjson::value>> describe_multi_item(schema_ptr schema,
const query::partition_slice&& slice,
shared_ptr<cql3::selection::selection> selection,
foreign_ptr<lw_shared_ptr<query::result>> query_result,
shared_ptr<const std::optional<attrs_to_get>> attrs_to_get);
const attrs_to_get&);
static void describe_single_item(const cql3::selection::selection&,
const std::vector<managed_bytes_opt>&,
const std::optional<attrs_to_get>&,
const std::vector<bytes_opt>&,
const attrs_to_get&,
rjson::value&,
bool = false);
static void add_stream_options(const rjson::value& stream_spec, schema_builder&, service::storage_proxy& sp);
static void supplement_table_info(rjson::value& descr, const schema& schema, service::storage_proxy& sp);
static void supplement_table_stream_info(rjson::value& descr, const schema& schema, service::storage_proxy& sp);
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;
};
// is_big() checks approximately if the given JSON value is "bigger" than
// the given big_size number of bytes. The goal is to *quickly* detect
// oversized JSON that, for example, is too large to be serialized to a
// contiguous string - we don't need an accurate size for that. Moreover,
// as soon as we detect that the JSON is indeed "big", we can return true
// and don't need to continue calculating its exact size.
// For simplicity, we use a recursive implementation. This is fine because
// Alternator limits the depth of JSONs it reads from inputs, and doesn't
// add more than a couple of levels in its own output construction.
bool is_big(const rjson::value& val, int big_size = 100'000);
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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"
@@ -29,7 +42,7 @@
namespace alternator {
template <typename Func, typename Result = std::result_of_t<Func(expressionsParser&)>>
Result do_with_parser(std::string_view input, Func&& f) {
Result do_with_parser(std::string input, Func&& f) {
expressionsLexer::InputStreamType input_stream{
reinterpret_cast<const ANTLR_UINT8*>(input.data()),
ANTLR_ENC_UTF8,
@@ -44,7 +57,7 @@ Result do_with_parser(std::string_view input, Func&& f) {
}
parsed::update_expression
parse_update_expression(std::string_view query) {
parse_update_expression(std::string query) {
try {
return do_with_parser(query, std::mem_fn(&expressionsParser::update_expression));
} catch (...) {
@@ -53,7 +66,7 @@ parse_update_expression(std::string_view query) {
}
std::vector<parsed::path>
parse_projection_expression(std::string_view query) {
parse_projection_expression(std::string query) {
try {
return do_with_parser(query, std::mem_fn(&expressionsParser::projection_expression));
} catch (...) {
@@ -62,7 +75,7 @@ parse_projection_expression(std::string_view query) {
}
parsed::condition_expression
parse_condition_expression(std::string_view query) {
parse_condition_expression(std::string query) {
try {
return do_with_parser(query, std::mem_fn(&expressionsParser::condition_expression));
} catch (...) {
@@ -634,8 +647,7 @@ std::unordered_map<std::string_view, function_handler_type*> function_handlers {
}
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,
f._parameters[0].is_constant(), f._parameters[1].is_constant()));
return to_bool_json(check_CONTAINS(v1.IsNull() ? nullptr : &v1, v2));
}
},
};

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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/>.
*/
/*

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -26,9 +39,9 @@ public:
using runtime_error::runtime_error;
};
parsed::update_expression parse_update_expression(std::string_view query);
std::vector<parsed::path> parse_projection_expression(std::string_view query);
parsed::condition_expression parse_condition_expression(std::string_view query);
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,

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -19,7 +32,7 @@
/*
* Parsed representation of expressions and their components.
*
* Types in alternator::parsed namespace are used for holding the parse
* 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

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "utils/base64.hh"
@@ -14,14 +27,11 @@
#include "rapidjson/writer.h"
#include "concrete_types.hh"
#include "cql3/type_json.hh"
#include "mutation/position_in_partition.hh"
static logging::logger slogger("alternator-serialization");
namespace alternator {
bool is_alternator_keyspace(const sstring& ks_name);
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}},
@@ -50,115 +60,6 @@ type_representation represent_type(alternator_type atype) {
return it->second;
}
// Get the magnitude and precision of a big_decimal - as these concepts are
// defined by DynamoDB - to allow us to enforce limits on those as explained
// in ssue #6794. The "magnitude" of 9e123 is 123 and of -9e-123 is -123,
// the "precision" of 12.34e56 is the number of significant digits - 4.
//
// Unfortunately it turned out to be quite difficult to take a big_decimal and
// calculate its magnitude and precision from its scale() and unscaled_value().
// So in the following ugly implementation we calculate them from the string
// representation instead. We assume the number was already parsed
// sucessfully to a big_decimal to it follows its syntax rules.
//
// FIXME: rewrite this function to take a big_decimal, not a string.
// Maybe a snippet like this can help:
// boost::multiprecision::cpp_int digits = boost::multiprecision::log10(num.unscaled_value().convert_to<boost::multiprecision::mpf_float_50>()).convert_to<boost::multiprecision::cpp_int>() + 1;
internal::magnitude_and_precision internal::get_magnitude_and_precision(std::string_view s) {
size_t e_or_end = s.find_first_of("eE");
std::string_view base = s.substr(0, e_or_end);
if (s[0]=='-' || s[0]=='+') {
base = base.substr(1);
}
int magnitude = 0;
int precision = 0;
size_t dot_or_end = base.find_first_of(".");
size_t nonzero = base.find_first_not_of("0");
if (dot_or_end != std::string_view::npos) {
if (nonzero == dot_or_end) {
// 0.000031 => magnitude = -5 (like 3.1e-5), precision = 2.
std::string_view fraction = base.substr(dot_or_end + 1);
size_t nonzero2 = fraction.find_first_not_of("0");
if (nonzero2 != std::string_view::npos) {
magnitude = -nonzero2 - 1;
precision = fraction.size() - nonzero2;
}
} else {
// 000123.45678 => magnitude = 2, precision = 8.
magnitude = dot_or_end - nonzero - 1;
precision = base.size() - nonzero - 1;
}
// trailing zeros don't count to precision, e.g., precision
// of 1000.0, 1.0 or 1.0000 are just 1.
size_t last_significant = base.find_last_not_of(".0");
if (last_significant == std::string_view::npos) {
precision = 0;
} else if (last_significant < dot_or_end) {
// e.g., 1000.00 reduce 5 = 7 - (0+1) - 1 from precision
precision -= base.size() - last_significant - 2;
} else {
// e.g., 1235.60 reduce 5 = 7 - (5+1) from precision
precision -= base.size() - last_significant - 1;
}
} else if (nonzero == std::string_view::npos) {
// all-zero integer 000000
magnitude = 0;
precision = 0;
} else {
magnitude = base.size() - 1 - nonzero;
precision = base.size() - nonzero;
// trailing zeros don't count to precision, e.g., precision
// of 1000 is just 1.
size_t last_significant = base.find_last_not_of("0");
if (last_significant == std::string_view::npos) {
precision = 0;
} else {
// e.g., 1000 reduce 3 = 4 - (0+1)
precision -= base.size() - last_significant - 1;
}
}
if (precision && e_or_end != std::string_view::npos) {
std::string_view exponent = s.substr(e_or_end + 1);
if (exponent.size() > 4) {
// don't even bother atoi(), exponent is too large
magnitude = exponent[0]=='-' ? -9999 : 9999;
} else {
try {
magnitude += boost::lexical_cast<int32_t>(exponent);
} catch (...) {
magnitude = 9999;
}
}
}
return magnitude_and_precision {magnitude, precision};
}
// Parse a number read from user input, validating that it has a valid
// numeric format and also in the allowed magnitude and precision ranges
// (see issue #6794). Throws an api_error::validation if the validation
// failed.
static big_decimal parse_and_validate_number(std::string_view s) {
try {
big_decimal ret(s);
auto [magnitude, precision] = internal::get_magnitude_and_precision(s);
if (magnitude > 125) {
throw api_error::validation(format("Number overflow: {}. Attempting to store a number with magnitude larger than supported range.", s));
}
if (magnitude < -130) {
throw api_error::validation(format("Number underflow: {}. Attempting to store a number with magnitude lower than supported range.", s));
}
if (precision > 38) {
throw api_error::validation(format("Number too precise: {}. Attempting to store a number with more significant digits than supported.", s));
}
return ret;
} catch (const marshal_exception& e) {
throw api_error::validation(format("The parameter cannot be converted to a numeric value: {}", s));
}
}
struct from_json_visitor {
const rjson::value& v;
bytes_ostream& bo;
@@ -168,19 +69,21 @@ struct from_json_visitor {
bo.write(t.from_string(rjson::to_string_view(v)));
}
void operator()(const bytes_type_impl& t) const {
// FIXME: it's difficult at this point to get information if value was provided
// in request or comes from the storage, for now we assume it's user's fault.
bo.write(*unwrap_bytes(v, true));
bo.write(rjson::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 {
bo.write(decimal_type->decompose(parse_and_validate_number(rjson::to_string_view(v))));
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));
bo.write(from_json_object(t, v, cql_serialization_format::internal()));
}
};
@@ -271,47 +174,32 @@ bytes get_key_column_value(const rjson::value& item, const column_definition& co
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 and the value is the encoded value.
// If this type does not match the desired "type_str", an api_error::validation
// error is thrown (the "name" parameter is the name of the column which will
// mentioned in the exception message).
// If the type does match, a reference to the encoded value is returned.
static const rjson::value& get_typed_value(const rjson::value& key_typed_value, std::string_view type_str, std::string_view name, std::string_view value_name) {
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 {} {}: {}",
value_name, name, key_typed_value));
}
auto it = key_typed_value.MemberBegin();
if (rjson::to_string_view(it->name) != type_str) {
throw api_error::validation(
format("Type mismatch: expected type {} for {} {}, got type {}",
type_str, value_name, name, it->name));
}
return it->value;
}
// 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) {
auto& value = get_typed_value(key_typed_value, type_to_string(column.type), column.name_as_text(), "key column");
std::string_view value_view = rjson::to_string_view(value);
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) {
// FIXME: it's difficult at this point to get information if value was provided
// in request or comes from the storage, for now we assume it's user's fault.
return *unwrap_bytes(value, true);
} else if (column.type == decimal_type) {
return decimal_type->decompose(parse_and_validate_number(rjson::to_string_view(value)));
return rjson::base64_decode(it->value);
} else {
return column.type->from_string(value_view);
return column.type->from_string(rjson::to_string_view(it->value));
}
}
@@ -321,7 +209,7 @@ rjson::value json_key_column_value(bytes_view cell, const column_definition& col
std::string b64 = base64_encode(cell);
return rjson::from_string(b64);
} if (column.type == utf8_type) {
return rjson::from_string(reinterpret_cast<const char*>(cell.data()), cell.size());
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
@@ -362,38 +250,6 @@ clustering_key ck_from_json(const rjson::value& item, schema_ptr schema) {
return clustering_key::from_exploded(raw_ck);
}
position_in_partition pos_from_json(const rjson::value& item, schema_ptr schema) {
auto ck = ck_from_json(item, schema);
if (is_alternator_keyspace(schema->ks_name())) {
return position_in_partition::for_key(std::move(ck));
}
const auto region_item = rjson::find(item, scylla_paging_region);
const auto weight_item = rjson::find(item, scylla_paging_weight);
if (bool(region_item) != bool(weight_item)) {
throw api_error::validation("Malformed value object: region and weight has to be either both missing or both present");
}
bound_weight weight;
if (region_item) {
auto region_view = rjson::to_string_view(get_typed_value(*region_item, "S", scylla_paging_region, "key region"));
auto weight_view = rjson::to_string_view(get_typed_value(*weight_item, "N", scylla_paging_weight, "key weight"));
auto region = parse_partition_region(region_view);
if (weight_view == "-1") {
weight = bound_weight::before_all_prefixed;
} else if (weight_view == "0") {
weight = bound_weight::equal;
} else if (weight_view == "1") {
weight = bound_weight::after_all_prefixed;
} else {
throw std::runtime_error(fmt::format("Invalid value for weight: {}", weight_view));
}
return position_in_partition(region, weight, region == partition_region::clustered ? std::optional(std::move(ck)) : std::nullopt);
}
if (ck.is_empty()) {
return position_in_partition::for_partition_start();
}
return position_in_partition::for_key(std::move(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));
@@ -402,38 +258,17 @@ big_decimal unwrap_number(const rjson::value& v, std::string_view diagnostic) {
if (it->name != "N") {
throw api_error::validation(format("{}: expected number, found type '{}'", diagnostic, it->name));
}
if (!it->value.IsString()) {
// We shouldn't reach here. Callers normally validate their input
// earlier with validate_value().
throw api_error::validation(format("{}: improperly formatted number constant", diagnostic));
}
big_decimal ret = parse_and_validate_number(rjson::to_string_view(it->value));
return ret;
}
std::optional<big_decimal> try_unwrap_number(const rjson::value& v) {
if (!v.IsObject() || v.MemberCount() != 1) {
return std::nullopt;
}
auto it = v.MemberBegin();
if (it->name != "N" || !it->value.IsString()) {
return std::nullopt;
}
try {
return parse_and_validate_number(rjson::to_string_view(it->value));
} catch (api_error&) {
return std::nullopt;
}
}
std::optional<bytes> unwrap_bytes(const rjson::value& value, bool from_query) {
try {
return rjson::base64_decode(value);
} catch (...) {
if (from_query) {
throw api_error::serialization(format("Invalid base64 data"));
if (it->value.IsNumber()) {
// FIXME(sarna): should use big_decimal constructor with numeric values directly:
return big_decimal(rjson::print(it->value));
}
return std::nullopt;
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));
}
}
@@ -444,7 +279,7 @@ const std::pair<std::string, const rjson::value*> unwrap_set(const rjson::value&
auto it = v.MemberBegin();
const std::string it_key = it->name.GetString();
if (it_key != "SS" && it_key != "BS" && it_key != "NS") {
return {std::move(it_key), nullptr};
return {"", nullptr};
}
return std::make_pair(it_key, &(it->value));
}
@@ -466,7 +301,7 @@ 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();
sstring str_ret = (n1 + n2).to_string();
std::string str_ret = std::string((n1 + n2).to_string());
rjson::add(ret, "N", rjson::from_string(str_ret));
return ret;
}
@@ -475,7 +310,7 @@ 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();
sstring str_ret = (n1 - n2).to_string();
std::string str_ret = std::string((n1 - n2).to_string());
rjson::add(ret, "N", rjson::from_string(str_ret));
return ret;
}
@@ -514,7 +349,7 @@ std::optional<rjson::value> set_diff(const rjson::value& v1, const rjson::value&
auto [set1_type, set1] = unwrap_set(v1);
auto [set2_type, set2] = unwrap_set(v2);
if (set1_type != set2_type) {
throw api_error::validation(format("Set DELETE type mismatch: {} and {}", 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");

View File

@@ -3,22 +3,32 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 <optional>
#include "types/types.hh"
#include "schema/schema_fwd.hh"
#include "types.hh"
#include "schema_fwd.hh"
#include "keys.hh"
#include "utils/rjson.hh"
#include "utils/big_decimal.hh"
class position_in_partition;
namespace alternator {
enum class alternator_type : int8_t {
@@ -35,9 +45,6 @@ struct type_representation {
data_type dtype;
};
inline constexpr std::string_view scylla_paging_region(":scylla:paging:region");
inline constexpr std::string_view scylla_paging_weight(":scylla:paging:weight");
type_info type_info_from_string(std::string_view type);
type_representation represent_type(alternator_type atype);
@@ -52,21 +59,11 @@ rjson::value json_key_column_value(bytes_view cell, const column_definition& col
partition_key pk_from_json(const rjson::value& item, schema_ptr schema);
clustering_key ck_from_json(const rjson::value& item, schema_ptr schema);
position_in_partition pos_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);
// try_unwrap_number is like unwrap_number, but returns an unset optional
// when the given v does not encode a number.
std::optional<big_decimal> try_unwrap_number(const rjson::value& v);
// unwrap_bytes decodes byte value, on decoding failure it either raises api_error::serialization
// iff from_query is true or returns unset optional iff from_query is false.
// Therefore it's safe to dereference returned optional when called with from_query equal true.
std::optional<bytes> unwrap_bytes(const rjson::value& value, bool from_query);
// 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}
@@ -94,12 +91,5 @@ std::optional<rjson::value> set_diff(const rjson::value& v1, const rjson::value&
// Returns a null value if one of the arguments is not actually a list.
rjson::value list_concatenate(const rjson::value& v1, const rjson::value& v2);
namespace internal {
struct magnitude_and_precision {
int magnitude;
int precision;
};
magnitude_and_precision get_magnitude_and_precision(std::string_view);
}
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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"
@@ -13,24 +26,20 @@
#include <seastar/core/coroutine.hh>
#include <seastar/json/json_elements.hh>
#include <seastar/util/defer.hh>
#include <seastar/util/short_streams.hh>
#include "seastarx.hh"
#include "error.hh"
#include "service/qos/service_level_controller.hh"
#include "utils/rjson.hh"
#include "auth.hh"
#include <cctype>
#include "service/storage_proxy.hh"
#include "locator/snitch_base.hh"
#include "gms/gossiper.hh"
#include "utils/overloaded_functor.hh"
#include "utils/fb_utilities.hh"
#include "utils/aws_sigv4.hh"
static logging::logger slogger("alternator-server");
using namespace httpd;
using request = http::request;
using reply = http::reply;
namespace alternator {
@@ -146,7 +155,7 @@ public:
std::unique_ptr<request> req, std::unique_ptr<reply> rep) override {
handle_CORS(*req, *rep, false);
return _f_handle(std::move(req), std::move(rep)).then(
[](std::unique_ptr<reply> rep) {
[this](std::unique_ptr<reply> rep) {
rep->set_mime_type("application/x-amz-json-1.0");
rep->done();
return make_ready_future<std::unique_ptr<reply>>(std::move(rep));
@@ -155,10 +164,8 @@ public:
protected:
void generate_error_reply(reply& rep, const api_error& err) {
rjson::value results = rjson::empty_object();
rjson::add(results, "__type", rjson::from_string("com.amazonaws.dynamodb.v20120810#" + err._type));
rjson::add(results, "message", err._msg);
rep._content = rjson::print(std::move(results));
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);
}
@@ -204,9 +211,10 @@ protected:
// It's very easy to get a list of all live nodes on the cluster,
// using _gossiper().get_live_members(). But getting
// just the list of live nodes in this DC needs more elaborate code:
auto& topology = _proxy.get_token_metadata_ptr()->get_topology();
sstring local_dc = topology.get_datacenter();
std::unordered_set<gms::inet_address> local_dc_nodes = topology.get_datacenter_endpoints().at(local_dc);
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 =
_proxy.get_token_metadata_ptr()->get_topology().get_datacenter_endpoints().at(local_dc);
for (auto& ip : local_dc_nodes) {
if (_gossiper.is_alive(ip)) {
rjson::push_back(results, rjson::from_string(ip.to_sstring()));
@@ -238,7 +246,7 @@ protected:
future<std::string> server::verify_signature(const request& req, const chunked_content& content) {
if (!_enforce_authorization) {
slogger.debug("Skipping authorization");
return make_ready_future<std::string>();
return make_ready_future<std::string>("<unauthenticated request>");
}
auto host_it = req._headers.find("Host");
if (host_it == req._headers.end()) {
@@ -320,13 +328,8 @@ future<std::string> server::verify_signature(const request& req, const chunked_c
region = std::move(region),
service = std::move(service),
user_signature = std::move(user_signature)] (key_cache::value_ptr key_ptr) {
std::string signature;
try {
signature = utils::aws::get_signature(user, *key_ptr, std::string_view(host), "/", req._method,
datestamp, signed_headers_str, signed_headers_map, &content, region, service, "");
} catch (const std::exception& e) {
throw api_error::invalid_signature(e.what());
}
std::string signature = get_signature(user, *key_ptr, std::string_view(host), req._method,
datestamp, signed_headers_str, signed_headers_map, content, region, service, "");
if (signature != std::string_view(user_signature)) {
_key_cache.remove(user);
@@ -373,9 +376,7 @@ static tracing::trace_state_ptr maybe_trace_query(service::client_state& client_
tracing::add_session_param(trace_state, "alternator_op", op);
tracing::add_query(trace_state, truncated_content_view(query, buf));
tracing::begin(trace_state, format("Alternator {}", op), client_state.get_client_address());
if (!username.empty()) {
tracing::set_username(trace_state, auth::authenticated_user(username));
}
tracing::set_username(trace_state, auth::authenticated_user(username));
}
return trace_state;
}
@@ -398,7 +399,7 @@ future<executor::request_return_type> server::handle_api_request(std::unique_ptr
}
auto units = co_await std::move(units_fut);
assert(req->content_stream);
chunked_content content = co_await util::read_entire_stream(*req->content_stream);
chunked_content content = co_await httpd::read_entire_stream(*req->content_stream);
auto username = co_await verify_signature(*req, content);
if (slogger.is_enabled(log_level::trace)) {
@@ -418,11 +419,7 @@ future<executor::request_return_type> server::handle_api_request(std::unique_ptr
auto leave = defer([this] () noexcept { _pending_requests.leave(); });
//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
executor::client_state client_state = username.empty()
? service::client_state{service::client_state::internal_tag()}
: service::client_state{service::client_state::internal_tag(), _auth_service, _sl_controller, username};
co_await client_state.maybe_update_per_service_level_params();
executor::client_state client_state{executor::client_state::internal_tag()};
tracing::trace_state_ptr trace_state = maybe_trace_query(client_state, username, op, content);
tracing::trace(trace_state, op);
rjson::value json_request = co_await _json_parser.parse(std::move(content));
@@ -455,14 +452,12 @@ void server::set_routes(routes& r) {
//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, service::storage_proxy& proxy, gms::gossiper& gossiper, auth::service& auth_service, qos::service_level_controller& sl_controller)
server::server(executor& exec, service::storage_proxy& proxy, gms::gossiper& gossiper)
: _http_server("http-alternator")
, _https_server("https-alternator")
, _executor(exec)
, _proxy(proxy)
, _gossiper(gossiper)
, _auth_service(auth_service)
, _sl_controller(sl_controller)
, _key_cache(1024, 1min, slogger)
, _enforce_authorization(false)
, _enabled_servers{}
@@ -537,9 +532,6 @@ server::server(executor& exec, service::storage_proxy& proxy, gms::gossiper& gos
{"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));
}},
{"DescribeContinuousBackups", [] (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_continuous_backups(client_state, std::move(permit), std::move(json_request));
}},
} {
}
@@ -553,28 +545,36 @@ future<> server::init(net::inet_address addr, std::optional<uint16_t> port, std:
" must be specified in order to init an alternator HTTP server instance"));
}
return seastar::async([this, addr, port, https_port, creds] {
_executor.start().get();
try {
_executor.start().get();
if (port) {
set_routes(_http_server._routes);
_http_server.set_content_length_limit(server::content_length_limit);
_http_server.set_content_streaming(true);
_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_content_streaming(true);
_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));
if (port) {
set_routes(_http_server._routes);
_http_server.set_content_length_limit(server::content_length_limit);
_http_server.set_content_streaming(true);
_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_content_streaming(true);
_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")));
}
});
}
@@ -631,7 +631,7 @@ future<> server::json_parser::stop() {
const char* api_error::what() const noexcept {
if (_what_string.empty()) {
_what_string = format("{} {}: {}", static_cast<int>(_http_code), _type, _msg);
_what_string = format("{} {}: {}", _http_code, _type, _msg);
}
return _what_string.c_str();
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -15,7 +28,6 @@
#include <seastar/net/tls.hh>
#include <optional>
#include "alternator/auth.hh"
#include "service/qos/service_level_controller.hh"
#include "utils/small_vector.hh"
#include "utils/updateable_value.hh"
#include <seastar/core/units.hh>
@@ -27,16 +39,14 @@ using chunked_content = rjson::chunked_content;
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<http::request>)>;
tracing::trace_state_ptr, service_permit, rjson::value, std::unique_ptr<request>)>;
using alternator_callbacks_map = std::unordered_map<std::string_view, alternator_callback>;
httpd::http_server _http_server;
httpd::http_server _https_server;
http_server _http_server;
http_server _https_server;
executor& _executor;
service::storage_proxy& _proxy;
gms::gossiper& _gossiper;
auth::service& _auth_service;
qos::service_level_controller& _sl_controller;
key_cache _key_cache;
bool _enforce_authorization;
@@ -68,7 +78,7 @@ class server {
json_parser _json_parser;
public:
server(executor& executor, service::storage_proxy& proxy, gms::gossiper& gossiper, auth::service& service, qos::service_level_controller& sl_controller);
server(executor& executor, service::storage_proxy& proxy, gms::gossiper& gossiper);
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, utils::updateable_value<uint32_t> max_concurrent_requests);
@@ -76,8 +86,8 @@ public:
private:
void set_routes(seastar::httpd::routes& r);
// If verification succeeds, returns the authenticated user's username
future<std::string> verify_signature(const seastar::http::request&, const chunked_content&);
future<executor::request_return_type> handle_api_request(std::unique_ptr<http::request> req);
future<std::string> verify_signature(const seastar::httpd::request&, const chunked_content&);
future<executor::request_return_type> handle_api_request(std::unique_ptr<request> req);
};
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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"

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 <type_traits>
@@ -15,6 +28,7 @@
#include "utils/base64.hh"
#include "log.hh"
#include "database.hh"
#include "db/config.hh"
#include "cdc/log.hh"
@@ -27,14 +41,14 @@
#include "cql3/result_set.hh"
#include "cql3/type_json.hh"
#include "cql3/column_identifier.hh"
#include "schema/schema_builder.hh"
#include "schema_builder.hh"
#include "service/storage_proxy.hh"
#include "gms/feature.hh"
#include "gms/feature_service.hh"
#include "executor.hh"
#include "tags_extension.hh"
#include "rmw_operation.hh"
#include "data_dictionary/data_dictionary.hh"
/**
* Base template type to implement rapidjson::internal::TypeHelper<...>:s
@@ -75,8 +89,8 @@ struct rapidjson::internal::TypeHelper<ValueType, utils::UUID>
: public from_string_helper<ValueType, utils::UUID>
{};
static db_clock::time_point as_timepoint(const table_id& tid) {
return db_clock::time_point{utils::UUID_gen::unix_timestamp(tid.uuid())};
static db_clock::time_point as_timepoint(const utils::UUID& uuid) {
return db_clock::time_point{utils::UUID_gen::unix_timestamp(uuid)};
}
/**
@@ -107,9 +121,6 @@ public:
stream_arn(const UUID& uuid)
: UUID(uuid)
{}
stream_arn(const table_id& tid)
: UUID(tid.uuid())
{}
stream_arn(std::string_view v)
: UUID(v.substr(1))
{
@@ -141,46 +152,27 @@ namespace alternator {
future<alternator::executor::request_return_type> alternator::executor::list_streams(client_state& client_state, service_permit permit, rjson::value request) {
_stats.api_operations.list_streams++;
auto limit = rjson::get_opt<int>(request, "Limit").value_or(100);
auto limit = rjson::get_opt<int>(request, "Limit").value_or(std::numeric_limits<int>::max());
auto streams_start = rjson::get_opt<stream_arn>(request, "ExclusiveStartStreamArn");
auto table = find_table(_proxy, request);
auto db = _proxy.data_dictionary();
auto& db = _proxy.get_db().local();
auto& cfs = db.get_column_families();
auto i = cfs.begin();
auto e = cfs.end();
if (limit < 1) {
throw api_error::validation("Limit must be 1 or more");
}
std::vector<data_dictionary::table> cfs;
if (table) {
auto log_name = cdc::log_name(table->cf_name());
try {
cfs.emplace_back(db.find_table(table->ks_name(), log_name));
} catch (data_dictionary::no_such_column_family&) {
cfs.clear();
}
} else {
cfs = db.get_tables();
}
// # 12601 (maybe?) - sort the set of tables on ID. This should ensure we never
// generate duplicates in a paged listing here. Can obviously miss things if they
// are added between paged calls and end up with a "smaller" UUID/ARN, but that
// is to be expected.
if (std::cmp_less(limit, cfs.size()) || streams_start) {
std::sort(cfs.begin(), cfs.end(), [](const data_dictionary::table& t1, const data_dictionary::table& t2) {
return t1.schema()->id().uuid() < t2.schema()->id().uuid();
});
}
auto i = cfs.begin();
auto e = cfs.end();
// TODO: the unordered_map here is not really well suited for partial
// querying - we're sorting on local hash order, and creating a table
// between queries may or may not miss info. But that should be rare,
// and we can probably expect this to be a single call.
if (streams_start) {
i = std::find_if(i, e, [&](const data_dictionary::table& t) {
return t.schema()->id().uuid() == streams_start
&& cdc::get_base_table(db.real_database(), *t.schema())
&& is_alternator_keyspace(t.schema()->ks_name())
i = std::find_if(i, e, [&](const std::pair<utils::UUID, lw_shared_ptr<column_family>>& p) {
return p.first == streams_start
&& cdc::get_base_table(db, *p.second->schema())
&& is_alternator_keyspace(p.second->schema()->ks_name())
;
});
if (i != e) {
@@ -194,17 +186,24 @@ future<alternator::executor::request_return_type> alternator::executor::list_str
std::optional<stream_arn> last;
for (;limit > 0 && i != e; ++i) {
auto s = i->schema();
auto s = i->second->schema();
auto& ks_name = s->ks_name();
auto& cf_name = s->cf_name();
if (!is_alternator_keyspace(ks_name)) {
continue;
}
if (cdc::is_log_for_some_table(db.real_database(), ks_name, cf_name)) {
if (table && ks_name != table->ks_name()) {
continue;
}
if (cdc::is_log_for_some_table(db, ks_name, cf_name)) {
if (table && table != cdc::get_base_table(db, *s)) {
continue;
}
rjson::value new_entry = rjson::empty_object();
last = i->schema()->id();
last = i->first;
rjson::add(new_entry, "StreamArn", *last);
rjson::add(new_entry, "StreamLabel", rjson::from_string(stream_label(*s)));
rjson::add(new_entry, "TableName", rjson::from_string(cdc::base_name(table_name(*s))));
@@ -425,12 +424,10 @@ using namespace std::string_literals;
* This will be a partial overlap, but it is the best we can do.
*/
static std::chrono::seconds confidence_interval(data_dictionary::database db) {
static std::chrono::seconds confidence_interval(const database& db) {
return std::chrono::seconds(db.get_config().alternator_streams_time_window_s());
}
using namespace std::chrono_literals;
// Dynamo docs says no data shall live longer than 24h.
static constexpr auto dynamodb_streams_max_window = 24h;
@@ -445,12 +442,12 @@ future<executor::request_return_type> executor::describe_stream(client_state& cl
auto stream_arn = rjson::get<alternator::stream_arn>(request, "StreamArn");
schema_ptr schema, bs;
auto db = _proxy.data_dictionary();
auto& db = _proxy.get_db().local();
try {
auto cf = db.find_column_family(table_id(stream_arn));
auto& cf = db.find_column_family(stream_arn);
schema = cf.schema();
bs = cdc::get_base_table(db.real_database(), *schema);
bs = cdc::get_base_table(_proxy.get_db().local(), *schema);
} catch (...) {
}
@@ -508,7 +505,7 @@ future<executor::request_return_type> executor::describe_stream(client_state& cl
// filter out cdc generations older than the table or now() - cdc::ttl (typically dynamodb_streams_max_window - 24h)
auto low_ts = std::max(as_timepoint(schema->id()), db_clock::now() - ttl);
return _sdks.cdc_get_versioned_streams(low_ts, { normal_token_owners }).then([db, shard_start, limit, ret = std::move(ret), stream_desc = std::move(stream_desc)] (std::map<db_clock::time_point, cdc::streams_version> topologies) mutable {
return _sdks.cdc_get_versioned_streams(low_ts, { normal_token_owners }).then([this, &db, shard_start, limit, ret = std::move(ret), stream_desc = std::move(stream_desc)] (std::map<db_clock::time_point, cdc::streams_version> topologies) mutable {
auto e = topologies.end();
auto prev = e;
@@ -729,18 +726,18 @@ future<executor::request_return_type> executor::get_shard_iterator(client_state&
}
auto stream_arn = rjson::get<alternator::stream_arn>(request, "StreamArn");
auto db = _proxy.data_dictionary();
auto& db = _proxy.get_db().local();
schema_ptr schema = nullptr;
std::optional<shard_id> sid;
try {
auto cf = db.find_column_family(table_id(stream_arn));
auto& cf = db.find_column_family(stream_arn);
schema = cf.schema();
sid = rjson::get<shard_id>(request, "ShardId");
} catch (...) {
}
if (!schema || !cdc::get_base_table(db.real_database(), *schema) || !is_alternator_keyspace(schema->ks_name())) {
if (!schema || !cdc::get_base_table(db, *schema) || !is_alternator_keyspace(schema->ks_name())) {
throw api_error::resource_not_found("Invalid StreamArn");
}
if (!sid) {
@@ -817,17 +814,17 @@ future<executor::request_return_type> executor::get_records(client_state& client
throw api_error::validation("Limit must be 1 or more");
}
auto db = _proxy.data_dictionary();
auto& db = _proxy.get_db().local();
schema_ptr schema, base;
try {
auto log_table = db.find_column_family(table_id(iter.table));
auto& log_table = db.find_column_family(iter.table);
schema = log_table.schema();
base = cdc::get_base_table(db.real_database(), *schema);
base = cdc::get_base_table(db, *schema);
} catch (...) {
}
if (!schema || !base || !is_alternator_keyspace(schema->ks_name())) {
throw api_error::resource_not_found(fmt::to_string(iter.table));
throw api_error::resource_not_found(boost::lexical_cast<std::string>(iter.table));
}
tracing::add_table_name(trace_state, schema->ks_name(), schema->cf_name());
@@ -850,14 +847,14 @@ future<executor::request_return_type> executor::get_records(client_state& client
static const bytes op_column_name = cdc::log_meta_column_name_bytes("operation");
static const bytes eor_column_name = cdc::log_meta_column_name_bytes("end_of_batch");
std::optional<attrs_to_get> key_names = boost::copy_range<attrs_to_get>(
auto key_names = boost::copy_range<attrs_to_get>(
boost::range::join(std::move(base->partition_key_columns()), std::move(base->clustering_key_columns()))
| boost::adaptors::transformed([&] (const column_definition& cdef) {
return std::make_pair<std::string, attrs_to_get_node>(cdef.name_as_text(), {}); })
);
// Include all base table columns as values (in case pre or post is enabled).
// This will include attributes not stored in the frozen map column
std::optional<attrs_to_get> attr_names = boost::copy_range<attrs_to_get>(base->regular_columns()
auto attr_names = boost::copy_range<attrs_to_get>(base->regular_columns()
// this will include the :attrs column, which we will also force evaluating.
// But not having this set empty forces out any cdc columns from actual result
| boost::adaptors::transformed([] (const column_definition& cdef) {
@@ -894,11 +891,11 @@ future<executor::request_return_type> executor::get_records(client_state& client
++mul;
}
auto command = ::make_lw_shared<query::read_command>(schema->id(), schema->version(), partition_slice, _proxy.get_max_result_size(partition_slice),
query::tombstone_limit(_proxy.get_tombstone_limit()), query::row_limit(limit * mul));
query::row_limit(limit * mul));
return _proxy.query(schema, std::move(command), std::move(partition_ranges), cl, service::storage_proxy::coordinator_query_options(default_timeout(), std::move(permit), client_state)).then(
[this, schema, partition_slice = std::move(partition_slice), selection = std::move(selection), start_time = std::move(start_time), limit, key_names = std::move(key_names), attr_names = std::move(attr_names), type, iter, high_ts] (service::storage_proxy::coordinator_query_result qr) mutable {
cql3::selection::result_set_builder builder(*selection, gc_clock::now());
cql3::selection::result_set_builder builder(*selection, gc_clock::now(), cql_serialization_format::latest());
query::result_view::consume(*qr.query_result, partition_slice, cql3::selection::result_set_builder::visitor(builder, *schema, *selection));
auto result_set = builder.build();
@@ -1026,7 +1023,7 @@ future<executor::request_return_type> executor::get_records(client_state& client
// ugh. figure out if we are and end-of-shard
auto normal_token_owners = _proxy.get_token_metadata_ptr()->count_normal_token_owners();
return _sdks.cdc_current_generation_timestamp({ normal_token_owners }).then([this, iter, high_ts, start_time, ret = std::move(ret)](db_clock::time_point ts) mutable {
auto& shard = iter.shard;
@@ -1044,27 +1041,24 @@ future<executor::request_return_type> executor::get_records(client_state& client
rjson::add(ret, "NextShardIterator", iter);
}
_stats.api_operations.get_records_latency.add(std::chrono::steady_clock::now() - start_time);
if (is_big(ret)) {
return make_ready_future<executor::request_return_type>(make_streamed(std::move(ret)));
}
return make_ready_future<executor::request_return_type>(make_jsonable(std::move(ret)));
});
});
}
void executor::add_stream_options(const rjson::value& stream_specification, schema_builder& builder, service::storage_proxy& sp) {
void executor::add_stream_options(const rjson::value& stream_specification, schema_builder& builder) const {
auto stream_enabled = rjson::find(stream_specification, "StreamEnabled");
if (!stream_enabled || !stream_enabled->IsBool()) {
throw api_error::validation("StreamSpecification needs boolean StreamEnabled");
}
if (stream_enabled->GetBool()) {
auto db = sp.data_dictionary();
auto& db = _proxy.get_db().local();
if (!db.features().cdc) {
if (!db.features().cluster_supports_cdc()) {
throw api_error::validation("StreamSpecification: streams (CDC) feature not enabled in cluster.");
}
if (!db.features().alternator_streams) {
if (!db.features().cluster_supports_alternator_streams()) {
throw api_error::validation("StreamSpecification: alternator streams feature not enabled in cluster.");
}
@@ -1096,11 +1090,11 @@ void executor::add_stream_options(const rjson::value& stream_specification, sche
}
}
void executor::supplement_table_stream_info(rjson::value& descr, const schema& schema, service::storage_proxy& sp) {
void executor::supplement_table_stream_info(rjson::value& descr, const schema& schema) const {
auto& opts = schema.cdc_options();
if (opts.enabled()) {
auto db = sp.data_dictionary();
auto cf = db.find_table(schema.ks_name(), cdc::log_name(schema.cf_name()));
auto& db = _proxy.get_db().local();
auto& cf = db.find_column_family(schema.ks_name(), cdc::log_name(schema.cf_name()));
stream_arn arn(cf.schema()->id());
rjson::add(descr, "LatestStreamArn", arn);
rjson::add(descr, "LatestStreamLabel", rjson::from_string(stream_label(*cf.schema())));

View File

@@ -3,16 +3,29 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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/schema.hh"
#include "schema.hh"
#include "db/extensions.hh"
namespace db {
namespace alternator {
class tags_extension : public schema_extension {
public:
@@ -37,9 +50,4 @@ private:
std::map<sstring, sstring> _tags;
};
// Information whether the view updates are synchronous is stored using the
// SYNCHRONOUS_VIEW_UPDATES_TAG_KEY tag. Value of this tag is a stored as a
// serialized boolean value ("true" or "false")
static const sstring SYNCHRONOUS_VIEW_UPDATES_TAG_KEY("system:synchronous_view_updates");
}

View File

@@ -3,56 +3,30 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 <chrono>
#include <cstdint>
#include <exception>
#include <optional>
#include <seastar/core/sstring.hh>
#include <seastar/core/coroutine.hh>
#include <seastar/core/sleep.hh>
#include <seastar/core/future.hh>
#include <seastar/core/lowres_clock.hh>
#include <seastar/coroutine/maybe_yield.hh>
#include <boost/multiprecision/cpp_int.hpp>
#include "exceptions/exceptions.hh"
#include "gms/gossiper.hh"
#include "gms/inet_address.hh"
#include "inet_address_vectors.hh"
#include "locator/abstract_replication_strategy.hh"
#include "log.hh"
#include "gc_clock.hh"
#include "replica/database.hh"
#include "service_permit.hh"
#include "timestamp.hh"
#include "executor.hh"
#include "service/storage_proxy.hh"
#include "service/pager/paging_state.hh"
#include "service/pager/query_pagers.hh"
#include "gms/feature_service.hh"
#include "sstables/types.hh"
#include "mutation/mutation.hh"
#include "types/types.hh"
#include "types/map.hh"
#include "database.hh"
#include "utils/rjson.hh"
#include "utils/big_decimal.hh"
#include "utils/fb_utilities.hh"
#include "cql3/selection/selection.hh"
#include "cql3/values.hh"
#include "cql3/query_options.hh"
#include "cql3/column_identifier.hh"
#include "alternator/executor.hh"
#include "alternator/controller.hh"
#include "alternator/serialization.hh"
#include "dht/sharder.hh"
#include "db/config.hh"
#include "db/tags/utils.hh"
#include "ttl.hh"
static logging::logger tlogger("alternator_ttl");
namespace alternator {
@@ -67,8 +41,8 @@ static const sstring TTL_TAG_KEY("system:ttl_attribute");
future<executor::request_return_type> executor::update_time_to_live(client_state& client_state, service_permit permit, rjson::value request) {
_stats.api_operations.update_time_to_live++;
if (!_proxy.data_dictionary().features().alternator_ttl) {
co_return api_error::unknown_operation("UpdateTimeToLive not yet supported. Experimental support is available if the 'alternator-ttl' experimental feature is enabled on all nodes.");
if (!_proxy.get_db().local().features().cluster_supports_alternator_ttl()) {
co_return api_error::unknown_operation("UpdateTimeToLive not yet supported. Experimental support is available if the 'alternator_ttl' experimental feature is enabled on all nodes.");
}
schema_ptr schema = get_table(_proxy, request);
@@ -94,25 +68,24 @@ future<executor::request_return_type> executor::update_time_to_live(client_state
}
sstring attribute_name(v->GetString(), v->GetStringLength());
co_await db::modify_tags(_mm, schema->ks_name(), schema->cf_name(), [&](std::map<sstring, sstring>& tags_map) {
if (enabled) {
if (tags_map.contains(TTL_TAG_KEY)) {
throw api_error::validation("TTL is already enabled");
}
tags_map[TTL_TAG_KEY] = attribute_name;
} else {
auto i = tags_map.find(TTL_TAG_KEY);
if (i == tags_map.end()) {
throw api_error::validation("TTL is already disabled");
} else if (i->second != attribute_name) {
throw api_error::validation(format(
"Requested to disable TTL on attribute {}, but a different attribute {} is enabled.",
attribute_name, i->second));
}
tags_map.erase(TTL_TAG_KEY);
std::map<sstring, sstring> tags_map = get_tags_of_table(schema);
if (enabled) {
if (tags_map.contains(TTL_TAG_KEY)) {
co_return api_error::validation("TTL is already enabled");
}
});
tags_map[TTL_TAG_KEY] = attribute_name;
} else {
auto i = tags_map.find(TTL_TAG_KEY);
if (i == tags_map.end()) {
co_return api_error::validation("TTL is already disabled");
} else if (i->second != attribute_name) {
co_return api_error::validation(format(
"Requested to disable TTL on attribute {}, but a different attribute {} is enabled.",
attribute_name, i->second));
}
tags_map.erase(TTL_TAG_KEY);
}
co_await update_tags(_mm, schema, std::move(tags_map));
// Prepare the response, which contains a TimeToLiveSpecification
// basically identical to the request's
rjson::value response = rjson::empty_object();
@@ -123,7 +96,7 @@ future<executor::request_return_type> executor::update_time_to_live(client_state
future<executor::request_return_type> executor::describe_time_to_live(client_state& client_state, service_permit permit, rjson::value request) {
_stats.api_operations.describe_time_to_live++;
schema_ptr schema = get_table(_proxy, request);
std::map<sstring, sstring> tags_map = get_tags_of_table_or_throw(schema);
std::map<sstring, sstring> tags_map = get_tags_of_table(schema);
rjson::value desc = rjson::empty_object();
auto i = tags_map.find(TTL_TAG_KEY);
if (i == tags_map.end()) {
@@ -137,713 +110,4 @@ future<executor::request_return_type> executor::describe_time_to_live(client_sta
co_return make_jsonable(std::move(response));
}
// expiration_service is a sharded service responsible for cleaning up expired
// items in all tables with per-item expiration enabled. Currently, this means
// Alternator tables with TTL configured via a UpdateTimeToLive request.
//
// Here is a brief overview of how the expiration service works:
//
// An expiration thread on each shard periodically scans the items (i.e.,
// rows) owned by this shard, looking for items whose chosen expiration-time
// attribute indicates they are expired, and deletes those items.
// The expiration-time "attribute" can be either an actual Scylla column
// (must be numeric) or an Alternator "attribute" - i.e., an element in
// the ATTRS_COLUMN_NAME map<utf8,bytes> column where the numeric expiration
// time is encoded in DynamoDB's JSON encoding inside the bytes value.
// To avoid scanning the same items RF times in RF replicas, only one node is
// responsible for scanning a token range at a time. Normally, this is the
// node owning this range as a "primary range" (the first node in the ring
// with this range), but when this node is down, the secondary owner (the
// second in the ring) may take over.
// An expiration thread is reponsible for all tables which need expiration
// scans. Currently, the different tables are scanned sequentially (not in
// parallel).
// The expiration thread scans item using CL=QUORUM to ensures that it reads
// a consistent expiration-time attribute. This means that the items are read
// locally and in addition QUORUM-1 additional nodes (one additional node
// when RF=3) need to read the data and send digests.
// When the expiration thread decides that an item has expired and wants
// to delete it, it does it using a CL=QUORUM write. This allows this
// deletion to be visible for consistent (quorum) reads. The deletion,
// like user deletions, will also appear on the CDC log and therefore
// Alternator Streams if enabled - currently as ordinary deletes (the
// userIdentity flag is currently missing this is issue #11523).
expiration_service::expiration_service(data_dictionary::database db, service::storage_proxy& proxy, gms::gossiper& g)
: _db(db)
, _proxy(proxy)
, _gossiper(g)
{
}
// Convert the big_decimal used to represent expiration time to an integer.
// Any fractional part is dropped. If the number is negative or invalid,
// 0 is returned, and if it's too high, the maximum unsigned long is returned.
static unsigned long bigdecimal_to_ul(const big_decimal& bd) {
// The big_decimal format has an integer mantissa of arbitrary length
// "unscaled_value" and then a (power of 10) exponent "scale".
if (bd.unscaled_value() <= 0) {
return 0;
}
if (bd.scale() == 0) {
// The fast path, when the expiration time is an integer, scale==0.
return static_cast<unsigned long>(bd.unscaled_value());
}
// Because the mantissa can be of arbitrary length, we work on it
// as a string. TODO: find a less ugly algorithm.
auto str = bd.unscaled_value().str();
if (bd.scale() > 0) {
int len = str.length();
if (len < bd.scale()) {
return 0;
}
str = str.substr(0, len-bd.scale());
} else {
if (bd.scale() < -20) {
return std::numeric_limits<unsigned long>::max();
}
for (int i = 0; i < -bd.scale(); i++) {
str.push_back('0');
}
}
// strtoul() returns ULONG_MAX if the number is too large, or 0 if not
// a number.
return strtoul(str.c_str(), nullptr, 10);
}
// The following is_expired() functions all check if an item with the given
// expiration time has expired, according to the DynamoDB API rules.
// The rules are:
// 1. If the expiration time attribute's value is not a number type,
// the item is not expired.
// 2. The expiration time is measured in seconds since the UNIX epoch.
// 3. If the expiration time is more than 5 years in the past, it is assumed
// to be malformed and ignored - and the item does not expire.
static bool is_expired(gc_clock::time_point expiration_time, gc_clock::time_point now) {
return expiration_time <= now &&
expiration_time > now - std::chrono::years(5);
}
static bool is_expired(const big_decimal& expiration_time, gc_clock::time_point now) {
unsigned long t = bigdecimal_to_ul(expiration_time);
// We assume - and the assumption turns out to be correct - that the
// epoch of gc_clock::time_point and the one used by the DynamoDB protocol
// are the same (the UNIX epoch in UTC). The resolution (seconds) is also
// the same.
return is_expired(gc_clock::time_point(gc_clock::duration(std::chrono::seconds(t))), now);
}
static bool is_expired(const rjson::value& expiration_time, gc_clock::time_point now) {
std::optional<big_decimal> n = try_unwrap_number(expiration_time);
return n && is_expired(*n, now);
}
// expire_item() expires an item - i.e., deletes it as appropriate for
// expiration - with CL=QUORUM and (FIXME!) in a way Alternator Streams
// understands it is an expiration event - not a user-initiated deletion.
static future<> expire_item(service::storage_proxy& proxy,
const service::query_state& qs,
const std::vector<managed_bytes_opt>& row,
schema_ptr schema,
api::timestamp_type ts) {
// Prepare the row key to delete
// NOTICE: the order of columns is guaranteed by the fact that selection::wildcard
// is used, which indicates that columns appear in the order defined by
// schema::all_columns_in_select_order() - partition key columns goes first,
// immediately followed by clustering key columns
std::vector<bytes> exploded_pk;
const unsigned pk_size = schema->partition_key_size();
const unsigned ck_size = schema->clustering_key_size();
for (unsigned c = 0; c < pk_size; ++c) {
const auto& row_c = row[c];
if (!row_c) {
// This shouldn't happen - all key columns must have values.
// But if it ever happens, let's just *not* expire the item.
// FIXME: log or increment a metric if this happens.
return make_ready_future<>();
}
exploded_pk.push_back(to_bytes(*row_c));
}
auto pk = partition_key::from_exploded(exploded_pk);
mutation m(schema, pk);
// If there's no clustering key, a tombstone should be created directly
// on a partition, not on a clustering row - otherwise it will look like
// an open-ended range tombstone, which will crash on KA/LA sstable format.
// See issue #6035
if (ck_size == 0) {
m.partition().apply(tombstone(ts, gc_clock::now()));
} else {
std::vector<bytes> exploded_ck;
for (unsigned c = pk_size; c < pk_size + ck_size; ++c) {
const auto& row_c = row[c];
if (!row_c) {
// This shouldn't happen - all key columns must have values.
// But if it ever happens, let's just *not* expire the item.
// FIXME: log or increment a metric if this happens.
return make_ready_future<>();
}
exploded_ck.push_back(to_bytes(*row_c));
}
auto ck = clustering_key::from_exploded(exploded_ck);
m.partition().clustered_row(*schema, ck).apply(tombstone(ts, gc_clock::now()));
}
std::vector<mutation> mutations;
mutations.push_back(std::move(m));
return proxy.mutate(std::move(mutations),
db::consistency_level::LOCAL_QUORUM,
executor::default_timeout(), // FIXME - which timeout?
qs.get_trace_state(), qs.get_permit(),
db::allow_per_partition_rate_limit::no);
}
static size_t random_offset(size_t min, size_t max) {
static thread_local std::default_random_engine re{std::random_device{}()};
std::uniform_int_distribution<size_t> dist(min, max);
return dist(re);
}
// Get a list of secondary token ranges for the given node, and the primary
// node responsible for each of these token ranges.
// A "secondary range" is a range of tokens where for each token, the second
// node (in ring order) out of the RF replicas that hold this token is the
// given node.
// In the expiration scanner, we want to scan a secondary range but only if
// this range's primary node is down. For this we need to return not just
// a list of this node's secondary ranges - but also the primary owner of
// each of those ranges.
static std::vector<std::pair<dht::token_range, gms::inet_address>> get_secondary_ranges(
const locator::effective_replication_map_ptr& erm,
gms::inet_address ep) {
const auto& tm = *erm->get_token_metadata_ptr();
const auto& sorted_tokens = tm.sorted_tokens();
std::vector<std::pair<dht::token_range, gms::inet_address>> ret;
if (sorted_tokens.empty()) {
on_internal_error(tlogger, "Token metadata is empty");
}
auto prev_tok = sorted_tokens.back();
for (const auto& tok : sorted_tokens) {
inet_address_vector_replica_set eps = erm->get_natural_endpoints(tok);
if (eps.size() <= 1 || eps[1] != ep) {
prev_tok = tok;
continue;
}
// Add the range (prev_tok, tok] to ret. However, if the range wraps
// around, split it to two non-wrapping ranges.
if (prev_tok < tok) {
ret.emplace_back(
dht::token_range{
dht::token_range::bound(prev_tok, false),
dht::token_range::bound(tok, true)},
eps[0]);
} else {
ret.emplace_back(
dht::token_range{
dht::token_range::bound(prev_tok, false),
std::nullopt},
eps[0]);
ret.emplace_back(
dht::token_range{
std::nullopt,
dht::token_range::bound(tok, true)},
eps[0]);
}
prev_tok = tok;
}
return ret;
}
// A class for iterating over all the token ranges *owned* by this shard.
// To avoid code duplication, it is a template with two distinct cases -
// <primary> and <secondary>:
//
// In the <primary> case, we consider a token *owned* by this shard if:
// 1. This node is a replica for this token.
// 2. Moreover, this node is the *primary* replica of the token (i.e., the
// first replica in the ring).
// 3. In this node, this shard is responsible for this token.
// We will use this definition of which shard in the cluster owns which tokens
// to split the expiration scanner's work between all the shards of the
// system.
//
// In the <secondary> case, we consider a token *owned* by this shard if:
// 1. This node is the *secondary* replica for this token (i.e., the second
// replica in the ring).
// 2. The primary replica for this token is currently marked down.
// 3. In this node, this shard is responsible for this token.
// We use the <secondary> case to handle the possibility that some of the
// nodes in the system are down. A dead node will not be expiring
// the tokens owned by it, so we want the secondary owner to take over its
// primary ranges.
//
// FIXME: need to decide how to choose primary ranges in multi-DC setup!
// We could call get_primary_ranges_within_dc() below instead of get_primary_ranges().
// NOTICE: Iteration currently starts from a random token range in order to improve
// the chances of covering all ranges during a scan when restarts occur.
// A more deterministic way would be to regularly persist the scanning state,
// but that incurs overhead that we want to avoid if not needed.
enum primary_or_secondary_t {primary, secondary};
template<primary_or_secondary_t primary_or_secondary>
class token_ranges_owned_by_this_shard {
// ranges_holder_primary holds just the primary ranges themselves
class ranges_holder_primary {
const dht::token_range_vector _token_ranges;
public:
ranges_holder_primary(const locator::vnode_effective_replication_map_ptr& erm, gms::gossiper& g, gms::inet_address ep)
: _token_ranges(erm->get_primary_ranges(ep)) {}
std::size_t size() const { return _token_ranges.size(); }
const dht::token_range& operator[](std::size_t i) const {
return _token_ranges[i];
}
bool should_skip(std::size_t i) const {
return false;
}
};
// ranges_holder<secondary> holds the secondary token ranges plus each
// range's primary owner, needed to implement should_skip().
class ranges_holder_secondary {
std::vector<std::pair<dht::token_range, gms::inet_address>> _token_ranges;
gms::gossiper& _gossiper;
public:
ranges_holder_secondary(const locator::effective_replication_map_ptr& erm, gms::gossiper& g, gms::inet_address ep)
: _token_ranges(get_secondary_ranges(erm, ep))
, _gossiper(g) {}
std::size_t size() const { return _token_ranges.size(); }
const dht::token_range& operator[](std::size_t i) const {
return _token_ranges[i].first;
}
// range i should be skipped if its primary owner is alive.
bool should_skip(std::size_t i) const {
return _gossiper.is_alive(_token_ranges[i].second);
}
};
schema_ptr _s;
// _token_ranges will contain a list of token ranges owned by this node.
// We'll further need to split each such range to the pieces owned by
// the current shard, using _intersecter.
using ranges_holder = std::conditional_t<
primary_or_secondary == primary_or_secondary_t::primary,
ranges_holder_primary,
ranges_holder_secondary>;
const ranges_holder _token_ranges;
// NOTICE: _range_idx is used modulo _token_ranges size when accessing
// the data to ensure that it doesn't go out of bounds
size_t _range_idx;
size_t _end_idx;
std::optional<dht::selective_token_range_sharder> _intersecter;
public:
token_ranges_owned_by_this_shard(replica::database& db, gms::gossiper& g, schema_ptr s)
: _s(s)
, _token_ranges(db.find_keyspace(s->ks_name()).get_effective_replication_map(),
g, utils::fb_utilities::get_broadcast_address())
, _range_idx(random_offset(0, _token_ranges.size() - 1))
, _end_idx(_range_idx + _token_ranges.size())
{
tlogger.debug("Generating token ranges starting from base range {} of {}", _range_idx, _token_ranges.size());
}
// Return the next token_range owned by this shard, or nullopt when the
// iteration ends.
std::optional<dht::token_range> next() {
// We may need three or more iterations in the following loop if a
// vnode doesn't intersect with the given shard at all (such a small
// vnode is unlikely, but possible). The loop cannot be infinite
// because each iteration of the loop advances _range_idx.
for (;;) {
if (_intersecter) {
std::optional<dht::token_range> ret = _intersecter->next();
if (ret) {
return ret;
}
// done with this range, go to next one
++_range_idx;
_intersecter = std::nullopt;
}
if (_range_idx == _end_idx) {
return std::nullopt;
}
// If should_skip(), the range should be skipped. This happens for
// a secondary range whose primary owning node is still alive.
while (_token_ranges.should_skip(_range_idx % _token_ranges.size())) {
++_range_idx;
if (_range_idx == _end_idx) {
return std::nullopt;
}
}
_intersecter.emplace(_s->get_sharder(), _token_ranges[_range_idx % _token_ranges.size()], this_shard_id());
}
}
// Same as next(), just return a partition_range instead of token_range
std::optional<dht::partition_range> next_partition_range() {
std::optional<dht::token_range> ret = next();
if (ret) {
return dht::to_partition_range(*ret);
} else {
return std::nullopt;
}
}
};
// Precomputed information needed to perform a scan on partition ranges
struct scan_ranges_context {
schema_ptr s;
bytes column_name;
std::optional<std::string> member;
::shared_ptr<cql3::selection::selection> selection;
std::unique_ptr<service::query_state> query_state_ptr;
std::unique_ptr<cql3::query_options> query_options;
::lw_shared_ptr<query::read_command> command;
scan_ranges_context(schema_ptr s, service::storage_proxy& proxy, bytes column_name, std::optional<std::string> member)
: s(s)
, column_name(column_name)
, member(member)
{
// FIXME: don't read the entire items - read only parts of it.
// We must read the key columns (to be able to delete) and also
// the requested attribute. If the requested attribute is a map's
// member we may be forced to read the entire map - but it would
// be good if we can read only the single item of the map - it
// should be possible (and a must for issue #7751!).
lw_shared_ptr<service::pager::paging_state> paging_state = nullptr;
auto regular_columns = boost::copy_range<query::column_id_vector>(
s->regular_columns() | boost::adaptors::transformed([] (const column_definition& cdef) { return cdef.id; }));
selection = cql3::selection::selection::wildcard(s);
query::partition_slice::option_set opts = selection->get_query_options();
opts.set<query::partition_slice::option::allow_short_read>();
// It is important that the scan bypass cache to avoid polluting it:
opts.set<query::partition_slice::option::bypass_cache>();
std::vector<query::clustering_range> ck_bounds{query::clustering_range::make_open_ended_both_sides()};
auto partition_slice = query::partition_slice(std::move(ck_bounds), {}, std::move(regular_columns), opts);
command = ::make_lw_shared<query::read_command>(s->id(), s->version(), partition_slice, proxy.get_max_result_size(partition_slice), query::tombstone_limit(proxy.get_tombstone_limit()));
executor::client_state client_state{executor::client_state::internal_tag()};
tracing::trace_state_ptr trace_state;
// NOTICE: empty_service_permit is used because the TTL service has fixed parallelism
query_state_ptr = std::make_unique<service::query_state>(client_state, trace_state, empty_service_permit());
// FIXME: What should we do on multi-DC? Will we run the expiration on the same ranges on all
// DCs or only once for each range? If the latter, we need to change the CLs in the
// scanner and deleter.
db::consistency_level cl = db::consistency_level::LOCAL_QUORUM;
query_options = std::make_unique<cql3::query_options>(cl, std::vector<cql3::raw_value>{});
query_options = std::make_unique<cql3::query_options>(std::move(query_options), std::move(paging_state));
}
};
// Scan data in a list of token ranges in one table, looking for expired
// items and deleting them.
// Because of issue #9167, partition_ranges must have a single partition
// range for this code to work correctly.
static future<> scan_table_ranges(
service::storage_proxy& proxy,
const scan_ranges_context& scan_ctx,
dht::partition_range_vector&& partition_ranges,
abort_source& abort_source,
named_semaphore& page_sem,
expiration_service::stats& expiration_stats)
{
const schema_ptr& s = scan_ctx.s;
assert (partition_ranges.size() == 1); // otherwise issue #9167 will cause incorrect results.
auto p = service::pager::query_pagers::pager(proxy, s, scan_ctx.selection, *scan_ctx.query_state_ptr,
*scan_ctx.query_options, scan_ctx.command, std::move(partition_ranges), nullptr);
while (!p->is_exhausted()) {
if (abort_source.abort_requested()) {
co_return;
}
auto units = co_await get_units(page_sem, 1);
// We don't need to limit page size in number of rows because there is
// a builtin limit of the page's size in bytes. Setting this limit to
// 1 is useful for debugging the paging code with moderate-size data.
uint32_t limit = std::numeric_limits<uint32_t>::max();
// Read a page, and if that times out, try again after a small sleep.
// If we didn't catch the timeout exception, it would cause the scan
// be aborted and only be restarted at the next scanning period.
// If we retry too many times, give up and restart the scan later.
std::unique_ptr<cql3::result_set> rs;
for (int retries=0; ; retries++) {
try {
// FIXME: which timeout?
rs = co_await p->fetch_page(limit, gc_clock::now(), executor::default_timeout());
break;
} catch(exceptions::read_timeout_exception&) {
tlogger.warn("expiration scanner read timed out, will retry: {}",
std::current_exception());
}
// If we didn't break out of this loop, add a minimal sleep
if (retries >= 10) {
// Don't get stuck forever asking the same page, maybe there's
// a bug or a real problem in several replicas. Give up on
// this scan an retry the scan from a random position later,
// in the next scan period.
throw runtime_exception("scanner thread failed after too many timeouts for the same page");
}
co_await sleep_abortable(std::chrono::seconds(1), abort_source);
}
auto rows = rs->rows();
auto meta = rs->get_metadata().get_names();
std::optional<unsigned> expiration_column;
for (unsigned i = 0; i < meta.size(); i++) {
const cql3::column_specification& col = *meta[i];
if (col.name->name() == scan_ctx.column_name) {
expiration_column = i;
break;
}
}
if (!expiration_column) {
continue;
}
for (const auto& row : rows) {
const managed_bytes_opt& cell = row[*expiration_column];
if (!cell) {
continue;
}
auto v = meta[*expiration_column]->type->deserialize(*cell);
bool expired = false;
// FIXME: don't recalculate "now" all the time
auto now = gc_clock::now();
if (scan_ctx.member) {
// In this case, the expiration-time attribute we're
// looking for is a member in a map, saved serialized
// into bytes using Alternator's serialization (basically
// a JSON serialized into bytes)
// FIXME: is it possible to find a specific member of a map
// without iterating through it like we do here and compare
// the key?
for (const auto& entry : value_cast<map_type_impl::native_type>(v)) {
std::string attr_name = value_cast<sstring>(entry.first);
if (value_cast<sstring>(entry.first) == *scan_ctx.member) {
bytes value = value_cast<bytes>(entry.second);
rjson::value json = deserialize_item(value);
expired = is_expired(json, now);
break;
}
}
} else {
// For a real column to contain an expiration time, it
// must be a numeric type.
// FIXME: Currently we only support decimal_type (which is
// what Alternator uses), but other numeric types can be
// supported as well to make this feature more useful in CQL.
// Note that kind::decimal is also checked above.
big_decimal n = value_cast<big_decimal>(v);
expired = is_expired(n, now);
}
if (expired) {
expiration_stats.items_deleted++;
// FIXME: maybe don't recalculate new_timestamp() all the time
// FIXME: if expire_item() throws on timeout, we need to retry it.
auto ts = api::new_timestamp();
co_await expire_item(proxy, *scan_ctx.query_state_ptr, row, s, ts);
}
}
// FIXME: once in a while, persist p->state(), so on reboot
// we don't start from scratch.
}
}
// scan_table() scans, in one table, data "owned" by this shard, looking for
// expired items and deleting them.
// We consider each node to "own" its primary token ranges, i.e., the tokens
// that this node is their first replica in the ring. Inside the node, each
// shard "owns" subranges of the node's token ranges - according to the node's
// sharding algorithm.
// When a node goes down, the token ranges owned by it will not be scanned
// and items in those token ranges will not expire, so in the future (FIXME)
// this function should additionally work on token ranges whose primary owner
// is down and this node is the range's secondary owner.
// If the TTL (expiration-time scanning) feature is not enabled for this
// table, scan_table() returns false without doing anything. Remember that the
// TTL feature may be enabled later so this function will need to be called
// again when the feature is enabled.
// Currently this function scans the entire table (or, rather the parts owned
// by this shard) at full rate, once. In the future (FIXME) we should consider
// how to pace this scan, how and when to repeat it, how to interleave or
// parallelize scanning of multiple tables, and how to continue scans after a
// reboot.
static future<bool> scan_table(
service::storage_proxy& proxy,
data_dictionary::database db,
gms::gossiper& gossiper,
schema_ptr s,
abort_source& abort_source,
named_semaphore& page_sem,
expiration_service::stats& expiration_stats)
{
// Check if an expiration-time attribute is enabled for this table.
// If not, just return false immediately.
// FIXME: the setting of the TTL may change in the middle of a long scan!
std::optional<std::string> attribute_name = db::find_tag(*s, TTL_TAG_KEY);
if (!attribute_name) {
co_return false;
}
// attribute_name may be one of the schema's columns (in Alternator, this
// means it's a key column), or an element in Alternator's attrs map
// encoded in Alternator's JSON encoding.
// FIXME: To make this less Alternators-specific, we should encode in the
// single key's value three things:
// 1. The name of a column
// 2. Optionally if column is a map, a member in the map
// 3. The deserializer for the value: CQL or Alternator (JSON).
// The deserializer can be guessed: If the given column or map item is
// numeric, it can be used directly. If it is a "bytes" type, it needs to
// be deserialized using Alternator's deserializer.
bytes column_name = to_bytes(*attribute_name);
const column_definition *cd = s->get_column_definition(column_name);
std::optional<std::string> member;
if (!cd) {
member = std::move(attribute_name);
column_name = bytes(executor::ATTRS_COLUMN_NAME);
cd = s->get_column_definition(column_name);
tlogger.info("table {} TTL enabled with attribute {} in {}", s->cf_name(), *member, executor::ATTRS_COLUMN_NAME);
} else {
tlogger.info("table {} TTL enabled with attribute {}", s->cf_name(), *attribute_name);
}
if (!cd) {
tlogger.info("table {} TTL column is missing, not scanning", s->cf_name());
co_return false;
}
data_type column_type = cd->type;
// Verify that the column has the right type: If "member" exists
// the column must be a map, and if it doesn't, the column must
// (currently) be a decimal_type. If the column has the wrong type
// nothing can get expired in this table, and it's pointless to
// scan it.
if ((member && column_type->get_kind() != abstract_type::kind::map) ||
(!member && column_type->get_kind() != abstract_type::kind::decimal)) {
tlogger.info("table {} TTL column has unsupported type, not scanning", s->cf_name());
co_return false;
}
expiration_stats.scan_table++;
// FIXME: need to pace the scan, not do it all at once.
scan_ranges_context scan_ctx{s, proxy, std::move(column_name), std::move(member)};
token_ranges_owned_by_this_shard<primary> my_ranges(db.real_database(), gossiper, s);
while (std::optional<dht::partition_range> range = my_ranges.next_partition_range()) {
// Note that because of issue #9167 we need to run a separate
// query on each partition range, and can't pass several of
// them into one partition_range_vector.
dht::partition_range_vector partition_ranges;
partition_ranges.push_back(std::move(*range));
// FIXME: if scanning a single range fails, including network errors,
// we fail the entire scan (and rescan from the beginning). Need to
// reconsider this. Saving the scan position might be a good enough
// solution for this problem.
co_await scan_table_ranges(proxy, scan_ctx, std::move(partition_ranges), abort_source, page_sem, expiration_stats);
}
// If each node only scans its own primary ranges, then when any node is
// down part of the token range will not get scanned. This can be viewed
// as acceptable (when the comes back online, it will resume its scan),
// but as noted in issue #9787, we can allow more prompt expiration
// by tasking another node to take over scanning of the dead node's primary
// ranges. What we do here is that this node will also check expiration
// on its *secondary* ranges - but only those whose primary owner is down.
token_ranges_owned_by_this_shard<secondary> my_secondary_ranges(db.real_database(), gossiper, s);
while (std::optional<dht::partition_range> range = my_secondary_ranges.next_partition_range()) {
expiration_stats.secondary_ranges_scanned++;
dht::partition_range_vector partition_ranges;
partition_ranges.push_back(std::move(*range));
co_await scan_table_ranges(proxy, scan_ctx, std::move(partition_ranges), abort_source, page_sem, expiration_stats);
}
co_return true;
}
future<> expiration_service::run() {
// FIXME: don't just tight-loop, think about timing, pace, and
// store position in durable storage, etc.
// FIXME: think about working on different tables in parallel.
// also need to notice when a new table is added, a table is
// deleted or when ttl is enabled or disabled for a table!
for (;;) {
auto start = lowres_clock::now();
// _db.tables() may change under our feet during a
// long-living loop, so we must keep our own copy of the list of
// schemas.
std::vector<schema_ptr> schemas;
for (auto cf : _db.get_tables()) {
schemas.push_back(cf.schema());
}
for (schema_ptr s : schemas) {
co_await coroutine::maybe_yield();
if (shutting_down()) {
co_return;
}
try {
co_await scan_table(_proxy, _db, _gossiper, s, _abort_source, _page_sem, _expiration_stats);
} catch (...) {
// The scan of a table may fail in the middle for many
// reasons, including network failure and even the table
// being removed. We'll continue scanning this table later
// (if it still exists). In any case it's important to catch
// the exception and not let the scanning service die for
// good.
// If the table has been deleted, it is expected that the scan
// will fail at some point, and even a warning is excessive.
if (_db.has_schema(s->ks_name(), s->cf_name())) {
tlogger.warn("table {}.{} expiration scan failed: {}",
s->ks_name(), s->cf_name(), std::current_exception());
} else {
tlogger.info("expiration scan failed when table {}.{} was deleted",
s->ks_name(), s->cf_name());
}
}
}
_expiration_stats.scan_passes++;
// The TTL scanner runs above once over all tables, at full steam.
// After completing such a scan, we sleep until it's time start
// another scan. TODO: If the scan went too fast, we can slow it down
// in the next iteration by reducing the scanner's scheduling-group
// share (if using a separate scheduling group), or introduce
// finer-grain sleeps into the scanning code.
std::chrono::milliseconds scan_duration(std::chrono::duration_cast<std::chrono::milliseconds>(lowres_clock::now() - start));
std::chrono::milliseconds period(long(_db.get_config().alternator_ttl_period_in_seconds() * 1000));
if (scan_duration < period) {
try {
tlogger.info("sleeping {} seconds until next period", (period - scan_duration).count()/1000.0);
co_await seastar::sleep_abortable(period - scan_duration, _abort_source);
} catch(seastar::sleep_aborted&) {}
} else {
tlogger.warn("scan took {} seconds, longer than period - not sleeping", scan_duration.count()/1000.0);
}
}
}
future<> expiration_service::start() {
// Called by main() on each shard to start the expiration-service
// thread. Just runs run() in the background and allows stop().
if (_db.features().alternator_ttl) {
if (!shutting_down()) {
_end = run().handle_exception([] (std::exception_ptr ep) {
tlogger.error("expiration_service failed: {}", ep);
});
}
}
return make_ready_future<>();
}
future<> expiration_service::stop() {
if (_abort_source.abort_requested()) {
throw std::logic_error("expiration_service::stop() called a second time");
}
_abort_source.request_abort();
if (!_end) {
// if _end is was not set, start() was never called
return make_ready_future<>();
}
return std::move(*_end);
}
expiration_service::stats::stats() {
_metrics.add_group("expiration", {
seastar::metrics::make_total_operations("scan_passes", scan_passes,
seastar::metrics::description("number of passes over the database")),
seastar::metrics::make_total_operations("scan_table", scan_table,
seastar::metrics::description("number of table scans (counting each scan of each table that enabled expiration)")),
seastar::metrics::make_total_operations("items_deleted", items_deleted,
seastar::metrics::description("number of items deleted after expiration")),
seastar::metrics::make_total_operations("secondary_ranges_scanned", secondary_ranges_scanned,
seastar::metrics::description("number of token ranges scanned by this node while their primary owner was down")),
});
}
} // namespace alternator

View File

@@ -1,80 +0,0 @@
/*
* Copyright 2021-present ScyllaDB
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
*/
#pragma once
#include "seastarx.hh"
#include <seastar/core/sharded.hh>
#include <seastar/core/abort_source.hh>
#include <seastar/core/semaphore.hh>
#include "data_dictionary/data_dictionary.hh"
namespace gms {
class gossiper;
}
namespace replica {
class database;
}
namespace service {
class storage_proxy;
}
namespace alternator {
// expiration_service is a sharded service responsible for cleaning up expired
// items in all tables with per-item expiration enabled. Currently, this means
// Alternator tables with TTL configured via a UpdateTimeToLeave request.
class expiration_service final : public seastar::peering_sharded_service<expiration_service> {
public:
// Object holding per-shard statistics related to the expiration service.
// While this object is alive, these metrics are also registered to be
// visible by the metrics REST API, with the "expiration_" prefix.
class stats {
public:
stats();
uint64_t scan_passes = 0;
uint64_t scan_table = 0;
uint64_t items_deleted = 0;
uint64_t secondary_ranges_scanned = 0;
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;
};
private:
data_dictionary::database _db;
service::storage_proxy& _proxy;
gms::gossiper& _gossiper;
// _end is set by start(), and resolves when the the background service
// started by it ends. To ask the background service to end, _abort_source
// should be triggered. stop() below uses both _abort_source and _end.
std::optional<future<>> _end;
abort_source _abort_source;
// Ensures that at most 1 page of scan results at a time is processed by the TTL service
named_semaphore _page_sem{1, named_semaphore_exception_factory{"alternator_ttl"}};
bool shutting_down() { return _abort_source.abort_requested(); }
stats _expiration_stats;
public:
// sharded_service<expiration_service>::start() creates this object on
// all shards, so calls this constructor on each shard. Later, the
// additional start() function should be invoked on all shards.
expiration_service(data_dictionary::database, service::storage_proxy&, gms::gossiper&);
future<> start();
future<> run();
// sharded_service<expiration_service>::stop() calls the following stop()
// method on each shard. This stop() asks the service on this shard to
// shut down as quickly as it can. The returned future indicates when the
// service is no longer running.
// stop() may be called even before start(), but may only be called once -
// calling it twice will result in an exception.
future<> stop();
};
} // namespace alternator

View File

@@ -1,15 +0,0 @@
version: 1
applications:
- frontend:
phases:
build:
commands:
- make setupenv
- make dirhtml
artifacts:
baseDirectory: _build/dirhtml
files:
- '**/*'
cache:
paths: []
appRoot: docs

View File

@@ -1,70 +0,0 @@
# Generate C++ sources from Swagger definitions
set(swagger_files
api-doc/authorization_cache.json
api-doc/cache_service.json
api-doc/collectd.json
api-doc/column_family.json
api-doc/commitlog.json
api-doc/compaction_manager.json
api-doc/config.json
api-doc/endpoint_snitch_info.json
api-doc/error_injection.json
api-doc/failure_detector.json
api-doc/gossiper.json
api-doc/hinted_handoff.json
api-doc/lsa.json
api-doc/messaging_service.json
api-doc/storage_proxy.json
api-doc/storage_service.json
api-doc/stream_manager.json
api-doc/system.json
api-doc/task_manager.json
api-doc/task_manager_test.json
api-doc/utils.json)
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_${fname}_files
IN_FILE "${CMAKE_CURRENT_SOURCE_DIR}/${f}"
OUT_DIR "${scylla_gen_build_dir}/api/${dir}")
list(APPEND swagger_gen_files "${scylla_swagger_gen_${fname}_files}")
endforeach()
add_library(api)
target_sources(api
PRIVATE
api.cc
cache_service.cc
collectd.cc
column_family.cc
commitlog.cc
compaction_manager.cc
config.cc
endpoint_snitch.cc
error_injection.cc
authorization_cache.cc
failure_detector.cc
gossiper.cc
hinted_handoff.cc
lsa.cc
messaging_service.cc
storage_proxy.cc
storage_service.cc
stream_manager.cc
system.cc
task_manager.cc
task_manager_test.cc
${swagger_gen_files})
target_include_directories(api
PUBLIC
${CMAKE_SOURCE_DIR}
${scylla_gen_build_dir})
target_link_libraries(api
idl
wasmtime_bindings
Seastar::seastar
xxHash::xxhash)

View File

@@ -1,29 +0,0 @@
{
"apiVersion":"0.0.1",
"swaggerVersion":"1.2",
"basePath":"{{Protocol}}://{{Host}}",
"resourcePath":"/authorization_cache",
"produces":[
"application/json"
],
"apis":[
{
"path":"/authorization_cache/reset",
"operations":[
{
"method":"POST",
"summary":"Reset cache",
"type":"void",
"nickname":"authorization_cache_reset",
"produces":[
"application/json"
],
"parameters":[
]
}
]
}
],
"models":{
}
}

View File

@@ -437,68 +437,6 @@
}
]
},
{
"path":"/column_family/tombstone_gc/{name}",
"operations":[
{
"method":"GET",
"summary":"Check if tombstone GC is enabled for a given table",
"type":"boolean",
"nickname":"get_tombstone_gc",
"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 tombstone GC for a given table",
"type":"void",
"nickname":"enable_tombstone_gc",
"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 tombstone GC for a given table",
"type":"void",
"nickname":"disable_tombstone_gc",
"produces":[
"application/json"
],
"parameters":[
{
"name":"name",
"description":"The table name in keyspace:name format",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
}
]
}
]
},
{
"path":"/column_family/estimate_keys/{name}",
"operations":[

View File

@@ -102,47 +102,7 @@
"parameters":[
{
"name":"type",
"description":"The type of compaction to stop. Can be one of: COMPACTION | CLEANUP | SCRUB | UPGRADE | RESHAPE",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"query"
}
]
}
]
},
{
"path":"/compaction_manager/stop_keyspace_compaction/{keyspace}",
"operations":[
{
"method":"POST",
"summary":"Stop all running compaction-like tasks in the given keyspace and tables having the provided type.",
"type":"void",
"nickname":"stop_keyspace_compaction",
"produces":[
"application/json"
],
"parameters":[
{
"name":"keyspace",
"description":"The keyspace to stop compaction in",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
},
{
"name":"tables",
"description":"Comma-separated tables to stop compaction in",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
},
{
"name":"type",
"description":"The type of compaction to stop. Can be one of: COMPACTION | CLEANUP | SCRUB | UPGRADE | RESHAPE",
"description":"the type of compaction to stop. Can be one of: - COMPACTION - VALIDATION - CLEANUP - SCRUB - INDEX_BUILD",
"required":true,
"allowMultiple":false,
"type":"string",

View File

@@ -624,7 +624,7 @@
},
{
"name":"kn",
"description":"Keyspace(s) to snapshot. Multiple keyspaces can be provided using a comma-separated list. If omitted, snapshot all keyspaces.",
"description":"Comma seperated keyspaces name to snapshot",
"required":false,
"allowMultiple":false,
"type":"string",
@@ -632,7 +632,7 @@
},
{
"name":"cf",
"description":"Table(s) to snapshot. Multiple tables (in a single keyspace) can be provided using a comma-separated list. If omitted, snapshot all tables in the given keyspace(s).",
"description":"the column family to snapshot",
"required":false,
"allowMultiple":false,
"type":"string",
@@ -667,7 +667,7 @@
},
{
"name":"kn",
"description":"Comma-separated keyspaces name that their snapshot will be deleted",
"description":"Comma seperated keyspaces name that their snapshot will be deleted",
"required":false,
"allowMultiple":false,
"type":"string",
@@ -723,7 +723,7 @@
},
{
"name":"cf",
"description":"Comma-separated column family names",
"description":"Comma seperated column family names",
"required":false,
"allowMultiple":false,
"type":"string",
@@ -755,39 +755,7 @@
},
{
"name":"cf",
"description":"Comma-separated column family names",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
}
]
}
]
},
{
"path":"/storage_service/keyspace_offstrategy_compaction/{keyspace}",
"operations":[
{
"method":"POST",
"summary":"Perform offstrategy compaction, if needed, in a single keyspace",
"type":"boolean",
"nickname":"perform_keyspace_offstrategy_compaction",
"produces":[
"application/json"
],
"parameters":[
{
"name":"keyspace",
"description":"The keyspace to operate on",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
},
{
"name":"cf",
"description":"Comma-separated table names",
"description":"Comma seperated column family names",
"required":false,
"allowMultiple":false,
"type":"string",
@@ -839,19 +807,6 @@
],
"paramType":"query"
},
{
"name":"quarantine_mode",
"description":"Controls whether to scrub quarantined sstables (default INCLUDE)",
"required":false,
"allowMultiple":false,
"type":"string",
"enum":[
"INCLUDE",
"EXCLUDE",
"ONLY"
],
"paramType":"query"
},
{
"name":"keyspace",
"description":"The keyspace to query about",
@@ -862,7 +817,7 @@
},
{
"name":"cf",
"description":"Comma-separated column family names",
"description":"Comma seperated column family names",
"required":false,
"allowMultiple":false,
"type":"string",
@@ -902,7 +857,7 @@
},
{
"name":"cf",
"description":"Comma-separated column family names",
"description":"Comma seperated column family names",
"required":false,
"allowMultiple":false,
"type":"string",
@@ -934,7 +889,7 @@
},
{
"name":"cf",
"description":"Comma-separated column family names",
"description":"Comma seperated column family names",
"required":false,
"allowMultiple":false,
"type":"string",
@@ -1228,7 +1183,7 @@
"operations":[
{
"method":"POST",
"summary":"Removes a node from the cluster. Replicated data that logically belonged to this node is redistributed among the remaining nodes.",
"summary":"Removes token (and all data associated with enpoint that had it) from the ring",
"type":"void",
"nickname":"remove_node",
"produces":[
@@ -1245,7 +1200,7 @@
},
{
"name":"ignore_nodes",
"description":"Comma-separated list of dead nodes to ignore in removenode operation. Use the same method for all nodes to ignore: either Host IDs or ip addresses.",
"description":"List of dead nodes to ingore in removenode operation",
"required":false,
"allowMultiple":false,
"type":"string",
@@ -2073,7 +2028,7 @@
},
{
"name":"cf",
"description":"Comma-separated column family names",
"description":"Comma seperated column family names",
"required":false,
"allowMultiple":false,
"type":"string",
@@ -2100,66 +2055,7 @@
},
{
"name":"cf",
"description":"Comma-separated column family names",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
}
]
}
]
},
{
"path":"/storage_service/tombstone_gc/{keyspace}",
"operations":[
{
"method":"POST",
"summary":"Enable tombstone GC",
"type":"void",
"nickname":"enable_tombstone_gc",
"produces":[
"application/json"
],
"parameters":[
{
"name":"keyspace",
"description":"The keyspace",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
},
{
"name":"cf",
"description":"Comma-separated column family names",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
}
]
},
{
"method":"DELETE",
"summary":"Disable tombstone GC",
"type":"void",
"nickname":"disable_tombstone_gc",
"produces":[
"application/json"
],
"parameters":[
{
"name":"keyspace",
"description":"The keyspace",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
},
{
"name":"cf",
"description":"Comma-separated column family names",
"description":"Comma seperated column family names",
"required":false,
"allowMultiple":false,
"type":"string",
@@ -2690,7 +2586,7 @@
"description":"File creation time"
},
"generation":{
"type":"string",
"type":"long",
"description":"SSTable generation"
},
"level":{
@@ -2700,7 +2596,7 @@
"version":{
"type":"string",
"enum":[
"ka", "la", "mc", "md", "me"
"ka", "la", "mc", "md"
],
"description":"SSTable version"
},

View File

@@ -52,45 +52,6 @@
}
]
},
{
"path":"/system/log",
"operations":[
{
"method":"POST",
"summary":"Write a message to the Scylla log",
"type":"void",
"nickname":"write_log_message",
"produces":[
"application/json"
],
"parameters":[
{
"name":"message",
"description":"The message to write to the log",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"query"
},
{
"name":"level",
"description":"The logging level to use",
"required":true,
"allowMultiple":false,
"type":"string",
"enum":[
"error",
"warn",
"info",
"debug",
"trace"
],
"paramType":"query"
}
]
}
]
},
{
"path":"/system/drop_sstable_caches",
"operations":[

View File

@@ -1,329 +0,0 @@
{
"apiVersion":"0.0.1",
"swaggerVersion":"1.2",
"basePath":"{{Protocol}}://{{Host}}",
"resourcePath":"/task_manager",
"produces":[
"application/json"
],
"apis":[
{
"path":"/task_manager/list_modules",
"operations":[
{
"method":"GET",
"summary":"Get all modules names",
"type":"array",
"items":{
"type":"string"
},
"nickname":"get_modules",
"produces":[
"application/json"
],
"parameters":[
]
}
]
},
{
"path":"/task_manager/list_module_tasks/{module}",
"operations":[
{
"method":"GET",
"summary":"Get a list of tasks",
"type":"array",
"items":{
"type":"task_stats"
},
"nickname":"get_tasks",
"produces":[
"application/json"
],
"parameters":[
{
"name":"module",
"description":"The module to query about",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
},
{
"name":"internal",
"description":"Boolean flag indicating whether internal tasks should be shown (false by default)",
"required":false,
"allowMultiple":false,
"type":"boolean",
"paramType":"query"
},
{
"name":"keyspace",
"description":"The keyspace to query about",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
},
{
"name":"table",
"description":"The table to query about",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
}
]
}
]
},
{
"path":"/task_manager/task_status/{task_id}",
"operations":[
{
"method":"GET",
"summary":"Get task status",
"type":"task_status",
"nickname":"get_task_status",
"produces":[
"application/json"
],
"parameters":[
{
"name":"task_id",
"description":"The uuid of a task to query about",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
}
]
}
]
},
{
"path":"/task_manager/abort_task/{task_id}",
"operations":[
{
"method":"POST",
"summary":"Abort running task and its descendants",
"type":"void",
"nickname":"abort_task",
"produces":[
"application/json"
],
"parameters":[
{
"name":"task_id",
"description":"The uuid of a task to abort",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
}
]
}
]
},
{
"path":"/task_manager/wait_task/{task_id}",
"operations":[
{
"method":"GET",
"summary":"Wait for a task to complete",
"type":"task_status",
"nickname":"wait_task",
"produces":[
"application/json"
],
"parameters":[
{
"name":"task_id",
"description":"The uuid of a task to wait for",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
}
]
}
]
},
{
"path":"/task_manager/task_status_recursive/{task_id}",
"operations":[
{
"method":"GET",
"summary":"Get statuses of the task and all its descendants",
"type":"array",
"items":{
"type":"task_status"
},
"nickname":"get_task_status_recursively",
"produces":[
"application/json"
],
"parameters":[
{
"name":"task_id",
"description":"The uuid of a task to query about",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
}
]
}
]
},
{
"path":"/task_manager/ttl",
"operations":[
{
"method":"POST",
"summary":"Set ttl in seconds and get last value",
"type":"long",
"nickname":"get_and_update_ttl",
"produces":[
"application/json"
],
"parameters":[
{
"name":"ttl",
"description":"The number of seconds for which the tasks will be kept in memory after it finishes",
"required":true,
"allowMultiple":false,
"type":"long",
"paramType":"query"
}
]
}
]
}
],
"models":{
"task_stats" :{
"id": "task_stats",
"description":"A task statistics object",
"properties":{
"task_id":{
"type":"string",
"description":"The uuid of a task"
},
"state":{
"type":"string",
"enum":[
"created",
"running",
"done",
"failed"
],
"description":"The state of a task"
},
"type":{
"type":"string",
"description":"The description of the task"
},
"keyspace":{
"type":"string",
"description":"The keyspace the task is working on (if applicable)"
},
"table":{
"type":"string",
"description":"The table the task is working on (if applicable)"
},
"entity":{
"type":"string",
"description":"Task-specific entity description"
},
"sequence_number":{
"type":"long",
"description":"The running sequence number of the task"
}
}
},
"task_status":{
"id":"task_status",
"description":"A task status object",
"properties":{
"id":{
"type":"string",
"description":"The uuid of the task"
},
"type":{
"type":"string",
"description":"The description of the task"
},
"state":{
"type":"string",
"enum":[
"created",
"running",
"done",
"failed"
],
"description":"The state of the task"
},
"is_abortable":{
"type":"boolean",
"description":"Boolean flag indicating whether the task can be aborted"
},
"start_time":{
"type":"datetime",
"description":"The start time of the task"
},
"end_time":{
"type":"datetime",
"description":"The end time of the task (unspecified when the task is not completed)"
},
"error":{
"type":"string",
"description":"Error string, if the task failed"
},
"parent_id":{
"type":"string",
"description":"The uuid of the parent task"
},
"sequence_number":{
"type":"long",
"description":"The running sequence number of the task"
},
"shard":{
"type":"long",
"description":"The number of a shard the task is running on"
},
"keyspace":{
"type":"string",
"description":"The keyspace the task is working on (if applicable)"
},
"table":{
"type":"string",
"description":"The table the task is working on (if applicable)"
},
"entity":{
"type":"string",
"description":"Task-specific entity description"
},
"progress_units":{
"type":"string",
"description":"A description of the progress units"
},
"progress_total":{
"type":"double",
"description":"The total number of units to complete for the task"
},
"progress_completed":{
"type":"double",
"description":"The number of units completed so far"
},
"children_ids":{
"type":"array",
"items":{
"type":"string"
},
"description":"Task IDs of children of this task"
}
}
}
}
}

View File

@@ -1,153 +0,0 @@
{
"apiVersion":"0.0.1",
"swaggerVersion":"1.2",
"basePath":"{{Protocol}}://{{Host}}",
"resourcePath":"/task_manager_test",
"produces":[
"application/json"
],
"apis":[
{
"path":"/task_manager_test/test_module",
"operations":[
{
"method":"POST",
"summary":"Register test module in task manager",
"type":"void",
"nickname":"register_test_module",
"produces":[
"application/json"
],
"parameters":[
]
},
{
"method":"DELETE",
"summary":"Unregister test module in task manager",
"type":"void",
"nickname":"unregister_test_module",
"produces":[
"application/json"
],
"parameters":[
]
}
]
},
{
"path":"/task_manager_test/test_task",
"operations":[
{
"method":"POST",
"summary":"Register test task",
"type":"string",
"nickname":"register_test_task",
"produces":[
"application/json"
],
"parameters":[
{
"name":"task_id",
"description":"The uuid of a task to register",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
},
{
"name":"shard",
"description":"The shard of the task",
"required":false,
"allowMultiple":false,
"type":"long",
"paramType":"query"
},
{
"name":"parent_id",
"description":"The uuid of a parent task",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
},
{
"name":"keyspace",
"description":"The keyspace the task is working on",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
},
{
"name":"table",
"description":"The table the task is working on",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
},
{
"name":"entity",
"description":"Task-specific entity description",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
}
]
},
{
"method":"DELETE",
"summary":"Unregister test task",
"type":"void",
"nickname":"unregister_test_task",
"produces":[
"application/json"
],
"parameters":[
{
"name":"task_id",
"description":"The uuid of a task to register",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"query"
}
]
}
]
},
{
"path":"/task_manager_test/finish_test_task/{task_id}",
"operations":[
{
"method":"POST",
"summary":"Finish test task",
"type":"void",
"nickname":"finish_test_task",
"produces":[
"application/json"
],
"parameters":[
{
"name":"task_id",
"description":"The uuid of a task to finish",
"required":true,
"allowMultiple":false,
"type":"string",
"paramType":"path"
},
{
"name":"error",
"description":"The error with which task fails (if it does)",
"required":false,
"allowMultiple":false,
"type":"string",
"paramType":"query"
}
]
}
]
}
]
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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.hh"
@@ -24,23 +37,19 @@
#include "compaction_manager.hh"
#include "hinted_handoff.hh"
#include "error_injection.hh"
#include "authorization_cache.hh"
#include <seastar/http/exception.hh>
#include "stream_manager.hh"
#include "system.hh"
#include "api/config.hh"
#include "task_manager.hh"
#include "task_manager_test.hh"
logging::logger apilog("api");
namespace api {
using namespace seastar::httpd;
static std::unique_ptr<reply> exception_reply(std::exception_ptr eptr) {
try {
std::rethrow_exception(eptr);
} catch (const replica::no_such_keyspace& ex) {
} catch (const no_such_keyspace& ex) {
throw bad_param_exception(ex.what());
}
// We never going to get here
@@ -100,9 +109,9 @@ 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, sharded<service::storage_service>& ss, sharded<gms::gossiper>& g, sharded<cdc::generation_service>& cdc_gs, sharded<db::system_keyspace>& sys_ks) {
return register_api(ctx, "storage_service", "The storage service API", [&ss, &g, &cdc_gs, &sys_ks] (http_context& ctx, routes& r) {
set_storage_service(ctx, r, ss, g.local(), cdc_gs, sys_ks);
future<> set_server_storage_service(http_context& ctx, sharded<service::storage_service>& ss, sharded<gms::gossiper>& g, sharded<cdc::generation_service>& cdc_gs) {
return register_api(ctx, "storage_service", "The storage service API", [&ss, &g, &cdc_gs] (http_context& ctx, routes& r) {
set_storage_service(ctx, r, ss, g.local(), cdc_gs);
});
}
@@ -130,17 +139,6 @@ future<> unset_server_repair(http_context& ctx) {
return ctx.http_server.set_routes([&ctx] (routes& r) { unset_repair(ctx, r); });
}
future<> set_server_authorization_cache(http_context &ctx, sharded<auth::service> &auth_service) {
return register_api(ctx, "authorization_cache",
"The authorization cache API", [&auth_service] (http_context &ctx, routes &r) {
set_authorization_cache(ctx, r, auth_service);
});
}
future<> unset_server_authorization_cache(http_context& ctx) {
return ctx.http_server.set_routes([&ctx] (routes& r) { unset_authorization_cache(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); });
}
@@ -149,14 +147,8 @@ 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, sharded<locator::snitch_ptr>& snitch) {
return register_api(ctx, "endpoint_snitch_info", "The endpoint snitch info API", [&snitch] (http_context& ctx, routes& r) {
set_endpoint_snitch(ctx, r, snitch);
});
}
future<> unset_server_snitch(http_context& ctx) {
return ctx.http_server.set_routes([&ctx] (routes& r) { unset_endpoint_snitch(ctx, r); });
future<> set_server_snitch(http_context& ctx) {
return register_api(ctx, "endpoint_snitch_info", "The endpoint snitch info API", set_endpoint_snitch);
}
future<> set_server_gossip(http_context& ctx, sharded<gms::gossiper>& g) {
@@ -166,15 +158,9 @@ future<> set_server_gossip(http_context& ctx, sharded<gms::gossiper>& g) {
});
}
future<> set_server_load_sstable(http_context& ctx, sharded<db::system_keyspace>& sys_ks) {
future<> set_server_load_sstable(http_context& ctx) {
return register_api(ctx, "column_family",
"The column family API", [&sys_ks] (http_context& ctx, routes& r) {
set_column_family(ctx, r, sys_ks);
});
}
future<> unset_server_load_sstable(http_context& ctx) {
return ctx.http_server.set_routes([&ctx] (routes& r) { unset_column_family(ctx, r); });
"The column family API", set_column_family);
}
future<> set_server_messaging_service(http_context& ctx, sharded<netw::messaging_service>& ms) {
@@ -194,19 +180,9 @@ future<> set_server_storage_proxy(http_context& ctx, sharded<service::storage_se
});
}
future<> unset_server_storage_proxy(http_context& ctx) {
return ctx.http_server.set_routes([&ctx] (routes& r) { unset_storage_proxy(ctx, r); });
}
future<> set_server_stream_manager(http_context& ctx, sharded<streaming::stream_manager>& sm) {
future<> set_server_stream_manager(http_context& ctx) {
return register_api(ctx, "stream_manager",
"The stream manager API", [&sm] (http_context& ctx, routes& r) {
set_stream_manager(ctx, r, sm);
});
}
future<> unset_server_stream_manager(http_context& ctx) {
return ctx.http_server.set_routes([&ctx] (routes& r) { unset_stream_manager(ctx, r); });
"The stream manager API", set_stream_manager);
}
future<> set_server_cache(http_context& ctx) {
@@ -264,56 +240,5 @@ future<> set_server_done(http_context& ctx) {
});
}
future<> set_server_task_manager(http_context& ctx, lw_shared_ptr<db::config> cfg) {
auto rb = std::make_shared < api_registry_builder > (ctx.api_doc);
return ctx.http_server.set_routes([rb, &ctx, &cfg = *cfg](routes& r) {
rb->register_function(r, "task_manager",
"The task manager API");
set_task_manager(ctx, r, cfg);
});
}
#ifndef SCYLLA_BUILD_MODE_RELEASE
future<> set_server_task_manager_test(http_context& ctx) {
auto rb = std::make_shared < api_registry_builder > (ctx.api_doc);
return ctx.http_server.set_routes([rb, &ctx](routes& r) mutable {
rb->register_function(r, "task_manager_test",
"The task manager test API");
set_task_manager_test(ctx, r);
});
}
#endif
void req_params::process(const request& req) {
// Process mandatory parameters
for (auto& [name, ent] : params) {
if (!ent.is_mandatory) {
continue;
}
try {
ent.value = req.param[name];
} catch (std::out_of_range&) {
throw httpd::bad_param_exception(fmt::format("Mandatory parameter '{}' was not provided", name));
}
}
// Process optional parameters
for (auto& [name, value] : req.query_parameters) {
try {
auto& ent = params.at(name);
if (ent.is_mandatory) {
throw httpd::bad_param_exception(fmt::format("Parameter '{}' is expected to be provided as part of the request url", name));
}
ent.value = value;
} catch (std::out_of_range&) {
throw httpd::bad_param_exception(fmt::format("Unsupported optional parameter '{}'", name));
}
}
}
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -27,7 +40,7 @@ template<class T>
std::vector<sstring> container_to_vec(const T& container) {
std::vector<sstring> res;
for (auto i : container) {
res.push_back(fmt::to_string(i));
res.push_back(boost::lexical_cast<std::string>(i));
}
return res;
}
@@ -47,8 +60,8 @@ template<class T, class MAP>
std::vector<T>& map_to_key_value(const MAP& map, std::vector<T>& res) {
for (auto i : map) {
T val;
val.key = fmt::to_string(i.first);
val.value = fmt::to_string(i.second);
val.key = boost::lexical_cast<std::string>(i.first);
val.value = boost::lexical_cast<std::string>(i.second);
res.push_back(val);
}
return res;
@@ -65,7 +78,7 @@ template <typename MAP>
std::vector<sstring> map_keys(const MAP& map) {
std::vector<sstring> res;
for (const auto& i : map) {
res.push_back(fmt::to_string(i.first));
res.push_back(boost::lexical_cast<std::string>(i.first));
}
return res;
}
@@ -81,6 +94,13 @@ inline std::vector<sstring> split(const sstring& text, const char* separator) {
return boost::split(tokens, text, boost::is_any_of(separator));
}
/**
* Split a column family parameter
*/
inline std::vector<sstring> split_cf(const sstring& cf) {
return split(cf, ",");
}
/**
* A helper function to sum values on an a distributed object that
* has a get_stats method.
@@ -137,14 +157,6 @@ future<json::json_return_type> sum_timer_stats(distributed<T>& d, utils::timed_
});
}
template<class T, class F>
future<json::json_return_type> sum_timer_stats(distributed<T>& d, utils::timed_rate_moving_average_summary_and_histogram F::*f) {
return d.map_reduce0([f](const T& p) {return (p.get_stats().*f).rate();}, utils::rate_moving_average_and_histogram(),
std::plus<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));
});
}
inline int64_t min_int64(int64_t a, int64_t b) {
return std::min(a,b);
}
@@ -189,7 +201,7 @@ struct basic_ratio_holder : public json::jsonable {
typedef basic_ratio_holder<double> ratio_holder;
typedef basic_ratio_holder<int64_t> integral_ratio_holder;
class unimplemented_exception : public httpd::base_exception {
class unimplemented_exception : public base_exception {
public:
unimplemented_exception()
: base_exception("API call is not supported yet", reply::status_type::internal_server_error) {
@@ -238,74 +250,13 @@ public:
value = T{boost::lexical_cast<Base>(param)};
}
} catch (boost::bad_lexical_cast&) {
throw httpd::bad_param_exception(format("{} ({}): type error - should be {}", name, param, boost::units::detail::demangle(typeid(Base).name())));
throw bad_param_exception(format("{} ({}): type error - should be {}", name, param, boost::units::detail::demangle(typeid(Base).name())));
}
}
operator T() const { return value; }
};
using mandatory = bool_class<struct mandatory_tag>;
class req_params {
public:
struct def {
std::optional<sstring> value;
mandatory is_mandatory = mandatory::no;
def(std::optional<sstring> value_ = std::nullopt, mandatory is_mandatory_ = mandatory::no)
: value(std::move(value_))
, is_mandatory(is_mandatory_)
{ }
def(mandatory is_mandatory_)
: is_mandatory(is_mandatory_)
{ }
};
private:
std::unordered_map<sstring, def> params;
public:
req_params(std::initializer_list<std::pair<sstring, def>> l) {
for (const auto& [name, ent] : l) {
add(std::move(name), std::move(ent));
}
}
void add(sstring name, def ent) {
params.emplace(std::move(name), std::move(ent));
}
void process(const request& req);
const std::optional<sstring>& get(const char* name) const {
return params.at(name).value;
}
template <typename T = sstring>
const std::optional<T> get_as(const char* name) const {
return get(name);
}
template <typename T = sstring>
requires std::same_as<T, bool>
const std::optional<bool> get_as(const char* name) const {
auto value = get(name);
if (!value) {
return std::nullopt;
}
std::transform(value->begin(), value->end(), value->begin(), ::tolower);
if (value == "true" || value == "yes" || value == "1") {
return true;
}
if (value == "false" || value == "no" || value == "0") {
return false;
}
throw boost::bad_lexical_cast{};
}
};
httpd::utils_json::estimated_histogram time_to_json_histogram(const utils::time_estimated_histogram& val);
utils_json::estimated_histogram time_to_json_histogram(const utils::time_estimated_histogram& val);
}

View File

@@ -3,20 +3,29 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 <seastar/http/httpd.hh>
#include <seastar/core/future.hh>
#include "replica/database_fwd.hh"
#include "tasks/task_manager.hh"
#include "database_fwd.hh"
#include "seastarx.hh"
using request = http::request;
using reply = http::reply;
namespace service {
class load_meter;
@@ -27,15 +36,10 @@ class storage_service;
class sstables_loader;
namespace streaming {
class stream_manager;
}
namespace locator {
class token_metadata;
class shared_token_metadata;
class snitch_ptr;
} // namespace locator
@@ -47,7 +51,6 @@ class config;
namespace view {
class view_builder;
}
class system_keyspace;
}
namespace netw { class messaging_service; }
class repair_service;
@@ -59,24 +62,21 @@ class gossiper;
}
namespace auth { class service; }
namespace api {
struct http_context {
sstring api_dir;
sstring api_doc;
httpd::http_server_control http_server;
distributed<replica::database>& db;
distributed<database>& db;
distributed<service::storage_proxy>& sp;
service::load_meter& lmeter;
const sharded<locator::shared_token_metadata>& shared_token_metadata;
sharded<tasks::task_manager>& tm;
http_context(distributed<replica::database>& _db,
http_context(distributed<database>& _db,
distributed<service::storage_proxy>& _sp,
service::load_meter& _lm, const sharded<locator::shared_token_metadata>& _stm, sharded<tasks::task_manager>& _tm)
: db(_db), sp(_sp), lmeter(_lm), shared_token_metadata(_stm), tm(_tm) {
service::load_meter& _lm, const sharded<locator::shared_token_metadata>& _stm)
: db(_db), sp(_sp), lmeter(_lm), shared_token_metadata(_stm) {
}
const locator::token_metadata& get_token_metadata();
@@ -84,9 +84,8 @@ struct http_context {
future<> set_server_init(http_context& ctx);
future<> set_server_config(http_context& ctx, const db::config& cfg);
future<> set_server_snitch(http_context& ctx, sharded<locator::snitch_ptr>& snitch);
future<> unset_server_snitch(http_context& ctx);
future<> set_server_storage_service(http_context& ctx, sharded<service::storage_service>& ss, sharded<gms::gossiper>& g, sharded<cdc::generation_service>& cdc_gs, sharded<db::system_keyspace>& sys_ks);
future<> set_server_snitch(http_context& ctx);
future<> set_server_storage_service(http_context& ctx, sharded<service::storage_service>& ss, sharded<gms::gossiper>& g, sharded<cdc::generation_service>& cdc_gs);
future<> set_server_sstables_loader(http_context& ctx, sharded<sstables_loader>& sst_loader);
future<> unset_server_sstables_loader(http_context& ctx);
future<> set_server_view_builder(http_context& ctx, sharded<db::view::view_builder>& vb);
@@ -97,26 +96,19 @@ future<> set_transport_controller(http_context& ctx, cql_transport::controller&
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_authorization_cache(http_context& ctx, sharded<auth::service> &auth_service);
future<> unset_server_authorization_cache(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, sharded<gms::gossiper>& g);
future<> set_server_load_sstable(http_context& ctx, sharded<db::system_keyspace>& sys_ks);
future<> unset_server_load_sstable(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_storage_proxy(http_context& ctx, sharded<service::storage_service>& ss);
future<> unset_server_storage_proxy(http_context& ctx);
future<> set_server_stream_manager(http_context& ctx, sharded<streaming::stream_manager>& sm);
future<> unset_server_stream_manager(http_context& ctx);
future<> set_server_stream_manager(http_context& ctx);
future<> set_hinted_handoff(http_context& ctx, sharded<gms::gossiper>& g);
future<> unset_hinted_handoff(http_context& ctx);
future<> set_server_gossip_settle(http_context& ctx, sharded<gms::gossiper>& g);
future<> set_server_cache(http_context& ctx);
future<> set_server_compaction_manager(http_context& ctx);
future<> set_server_done(http_context& ctx);
future<> set_server_task_manager(http_context& ctx, lw_shared_ptr<db::config> cfg);
future<> set_server_task_manager_test(http_context& ctx);
}

View File

@@ -1,34 +0,0 @@
/*
* Copyright (C) 2022-present ScyllaDB
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
*/
#include "api/api-doc/authorization_cache.json.hh"
#include "api/authorization_cache.hh"
#include "api/api.hh"
#include "auth/common.hh"
namespace api {
using namespace json;
using namespace seastar::httpd;
void set_authorization_cache(http_context& ctx, routes& r, sharded<auth::service> &auth_service) {
httpd::authorization_cache_json::authorization_cache_reset.set(r, [&auth_service] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
co_await auth_service.invoke_on_all([] (auth::service& auth) -> future<> {
auth.reset_authorization_cache();
return make_ready_future<>();
});
co_return json_void();
});
}
void unset_authorization_cache(http_context& ctx, routes& r) {
httpd::authorization_cache_json::authorization_cache_reset.unset(r);
}
}

View File

@@ -1,18 +0,0 @@
/*
* Copyright (C) 2022-present ScyllaDB
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
*/
#pragma once
#include "api.hh"
namespace api {
void set_authorization_cache(http_context& ctx, httpd::routes& r, sharded<auth::service> &auth_service);
void unset_authorization_cache(http_context& ctx, httpd::routes& r);
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "cache_service.hh"
@@ -12,128 +25,127 @@
namespace api {
using namespace json;
using namespace seastar::httpd;
namespace cs = httpd::cache_service_json;
void set_cache_service(http_context& ctx, routes& r) {
cs::get_row_cache_save_period_in_seconds.set(r, [](std::unique_ptr<http::request> req) {
cs::get_row_cache_save_period_in_seconds.set(r, [](std::unique_ptr<request> req) {
// We never save the cache
// Origin uses 0 for never
return make_ready_future<json::json_return_type>(0);
});
cs::set_row_cache_save_period_in_seconds.set(r, [](std::unique_ptr<http::request> req) {
cs::set_row_cache_save_period_in_seconds.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
auto period = req->get_query_param("period");
return make_ready_future<json::json_return_type>(json_void());
});
cs::get_key_cache_save_period_in_seconds.set(r, [](std::unique_ptr<http::request> req) {
cs::get_key_cache_save_period_in_seconds.set(r, [](std::unique_ptr<request> req) {
// We never save the cache
// Origin uses 0 for never
return make_ready_future<json::json_return_type>(0);
});
cs::set_key_cache_save_period_in_seconds.set(r, [](std::unique_ptr<http::request> req) {
cs::set_key_cache_save_period_in_seconds.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
auto period = req->get_query_param("period");
return make_ready_future<json::json_return_type>(json_void());
});
cs::get_counter_cache_save_period_in_seconds.set(r, [](std::unique_ptr<http::request> req) {
cs::get_counter_cache_save_period_in_seconds.set(r, [](std::unique_ptr<request> req) {
// We never save the cache
// Origin uses 0 for never
return make_ready_future<json::json_return_type>(0);
});
cs::set_counter_cache_save_period_in_seconds.set(r, [](std::unique_ptr<http::request> req) {
cs::set_counter_cache_save_period_in_seconds.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
auto ccspis = req->get_query_param("ccspis");
return make_ready_future<json::json_return_type>(json_void());
});
cs::get_row_cache_keys_to_save.set(r, [](std::unique_ptr<http::request> req) {
cs::get_row_cache_keys_to_save.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
return make_ready_future<json::json_return_type>(0);
});
cs::set_row_cache_keys_to_save.set(r, [](std::unique_ptr<http::request> req) {
cs::set_row_cache_keys_to_save.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
auto rckts = req->get_query_param("rckts");
return make_ready_future<json::json_return_type>(json_void());
});
cs::get_key_cache_keys_to_save.set(r, [](std::unique_ptr<http::request> req) {
cs::get_key_cache_keys_to_save.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
return make_ready_future<json::json_return_type>(0);
});
cs::set_key_cache_keys_to_save.set(r, [](std::unique_ptr<http::request> req) {
cs::set_key_cache_keys_to_save.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
auto kckts = req->get_query_param("kckts");
return make_ready_future<json::json_return_type>(json_void());
});
cs::get_counter_cache_keys_to_save.set(r, [](std::unique_ptr<http::request> req) {
cs::get_counter_cache_keys_to_save.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
return make_ready_future<json::json_return_type>(0);
});
cs::set_counter_cache_keys_to_save.set(r, [](std::unique_ptr<http::request> req) {
cs::set_counter_cache_keys_to_save.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
auto cckts = req->get_query_param("cckts");
return make_ready_future<json::json_return_type>(json_void());
});
cs::invalidate_key_cache.set(r, [](std::unique_ptr<http::request> req) {
cs::invalidate_key_cache.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
return make_ready_future<json::json_return_type>(json_void());
});
cs::invalidate_counter_cache.set(r, [](std::unique_ptr<http::request> req) {
cs::invalidate_counter_cache.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
return make_ready_future<json::json_return_type>(json_void());
});
cs::set_row_cache_capacity_in_mb.set(r, [](std::unique_ptr<http::request> req) {
cs::set_row_cache_capacity_in_mb.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
auto capacity = req->get_query_param("capacity");
return make_ready_future<json::json_return_type>(json_void());
});
cs::set_key_cache_capacity_in_mb.set(r, [](std::unique_ptr<http::request> req) {
cs::set_key_cache_capacity_in_mb.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
auto period = req->get_query_param("period");
return make_ready_future<json::json_return_type>(json_void());
});
cs::set_counter_cache_capacity_in_mb.set(r, [](std::unique_ptr<http::request> req) {
cs::set_counter_cache_capacity_in_mb.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
auto capacity = req->get_query_param("capacity");
return make_ready_future<json::json_return_type>(json_void());
});
cs::save_caches.set(r, [](std::unique_ptr<http::request> req) {
cs::save_caches.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
return make_ready_future<json::json_return_type>(json_void());
});
cs::get_key_capacity.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_key_capacity.set(r, [] (std::unique_ptr<request> req) {
// TBD
// FIXME
// we don't support keys cache,
@@ -141,7 +153,7 @@ void set_cache_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(0);
});
cs::get_key_hits.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_key_hits.set(r, [] (std::unique_ptr<request> req) {
// TBD
// FIXME
// we don't support keys cache,
@@ -149,7 +161,7 @@ void set_cache_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(0);
});
cs::get_key_requests.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_key_requests.set(r, [] (std::unique_ptr<request> req) {
// TBD
// FIXME
// we don't support keys cache,
@@ -157,7 +169,7 @@ void set_cache_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(0);
});
cs::get_key_hit_rate.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_key_hit_rate.set(r, [] (std::unique_ptr<request> req) {
// TBD
// FIXME
// we don't support keys cache,
@@ -165,21 +177,21 @@ void set_cache_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(0);
});
cs::get_key_hits_moving_avrage.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_key_hits_moving_avrage.set(r, [&ctx] (std::unique_ptr<request> req) {
// TBD
// FIXME
// See above
return make_ready_future<json::json_return_type>(meter_to_json(utils::rate_moving_average()));
});
cs::get_key_requests_moving_avrage.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_key_requests_moving_avrage.set(r, [&ctx] (std::unique_ptr<request> req) {
// TBD
// FIXME
// See above
return make_ready_future<json::json_return_type>(meter_to_json(utils::rate_moving_average()));
});
cs::get_key_size.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_key_size.set(r, [] (std::unique_ptr<request> req) {
// TBD
// FIXME
// we don't support keys cache,
@@ -187,7 +199,7 @@ void set_cache_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(0);
});
cs::get_key_entries.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_key_entries.set(r, [] (std::unique_ptr<request> req) {
// TBD
// FIXME
// we don't support keys cache,
@@ -195,68 +207,68 @@ void set_cache_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(0);
});
cs::get_row_capacity.set(r, [&ctx] (std::unique_ptr<http::request> req) {
return ctx.db.map_reduce0([](replica::database& db) -> uint64_t {
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);
});
});
cs::get_row_hits.set(r, [&ctx] (std::unique_ptr<http::request> req) {
return map_reduce_cf(ctx, uint64_t(0), [](const replica::column_family& cf) {
cs::get_row_hits.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf(ctx, uint64_t(0), [](const column_family& cf) {
return cf.get_row_cache().stats().hits.count();
}, std::plus<uint64_t>());
});
cs::get_row_requests.set(r, [&ctx] (std::unique_ptr<http::request> req) {
return map_reduce_cf(ctx, uint64_t(0), [](const replica::column_family& cf) {
cs::get_row_requests.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf(ctx, uint64_t(0), [](const column_family& cf) {
return cf.get_row_cache().stats().hits.count() + cf.get_row_cache().stats().misses.count();
}, std::plus<uint64_t>());
});
cs::get_row_hit_rate.set(r, [&ctx] (std::unique_ptr<http::request> req) {
return map_reduce_cf(ctx, ratio_holder(), [](const replica::column_family& cf) {
cs::get_row_hit_rate.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf(ctx, ratio_holder(), [](const column_family& cf) {
return ratio_holder(cf.get_row_cache().stats().hits.count() + cf.get_row_cache().stats().misses.count(),
cf.get_row_cache().stats().hits.count());
}, std::plus<ratio_holder>());
});
cs::get_row_hits_moving_avrage.set(r, [&ctx] (std::unique_ptr<http::request> req) {
return map_reduce_cf_raw(ctx, utils::rate_moving_average(), [](const replica::column_family& cf) {
cs::get_row_hits_moving_avrage.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf_raw(ctx, utils::rate_moving_average(), [](const column_family& cf) {
return cf.get_row_cache().stats().hits.rate();
}, std::plus<utils::rate_moving_average>()).then([](const utils::rate_moving_average& m) {
return make_ready_future<json::json_return_type>(meter_to_json(m));
});
});
cs::get_row_requests_moving_avrage.set(r, [&ctx] (std::unique_ptr<http::request> req) {
return map_reduce_cf_raw(ctx, utils::rate_moving_average(), [](const replica::column_family& cf) {
cs::get_row_requests_moving_avrage.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf_raw(ctx, utils::rate_moving_average(), [](const column_family& cf) {
return cf.get_row_cache().stats().hits.rate() + cf.get_row_cache().stats().misses.rate();
}, std::plus<utils::rate_moving_average>()).then([](const utils::rate_moving_average& m) {
return make_ready_future<json::json_return_type>(meter_to_json(m));
});
});
cs::get_row_size.set(r, [&ctx] (std::unique_ptr<http::request> req) {
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([](replica::database& db) -> uint64_t {
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);
});
});
cs::get_row_entries.set(r, [&ctx] (std::unique_ptr<http::request> req) {
return ctx.db.map_reduce0([](replica::database& db) -> 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);
});
});
cs::get_counter_capacity.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_counter_capacity.set(r, [] (std::unique_ptr<request> req) {
// TBD
// FIXME
// we don't support counter cache,
@@ -264,7 +276,7 @@ void set_cache_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(0);
});
cs::get_counter_hits.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_counter_hits.set(r, [] (std::unique_ptr<request> req) {
// TBD
// FIXME
// we don't support counter cache,
@@ -272,7 +284,7 @@ void set_cache_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(0);
});
cs::get_counter_requests.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_counter_requests.set(r, [] (std::unique_ptr<request> req) {
// TBD
// FIXME
// we don't support counter cache,
@@ -280,7 +292,7 @@ void set_cache_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(0);
});
cs::get_counter_hit_rate.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_counter_hit_rate.set(r, [] (std::unique_ptr<request> req) {
// TBD
// FIXME
// we don't support counter cache,
@@ -288,21 +300,21 @@ void set_cache_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(0);
});
cs::get_counter_hits_moving_avrage.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_counter_hits_moving_avrage.set(r, [&ctx] (std::unique_ptr<request> req) {
// TBD
// FIXME
// See above
return make_ready_future<json::json_return_type>(meter_to_json(utils::rate_moving_average()));
});
cs::get_counter_requests_moving_avrage.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_counter_requests_moving_avrage.set(r, [&ctx] (std::unique_ptr<request> req) {
// TBD
// FIXME
// See above
return make_ready_future<json::json_return_type>(meter_to_json(utils::rate_moving_average()));
});
cs::get_counter_size.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_counter_size.set(r, [] (std::unique_ptr<request> req) {
// TBD
// FIXME
// we don't support counter cache,
@@ -310,7 +322,7 @@ void set_cache_service(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(0);
});
cs::get_counter_entries.set(r, [] (std::unique_ptr<http::request> req) {
cs::get_counter_entries.set(r, [] (std::unique_ptr<request> req) {
// TBD
// FIXME
// we don't support counter cache,

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -12,6 +25,6 @@
namespace api {
void set_cache_service(http_context& ctx, httpd::routes& r);
void set_cache_service(http_context& ctx, routes& r);
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "collectd.hh"
@@ -29,11 +42,8 @@ static auto transformer(const std::vector<collectd_value>& values) {
case scollectd::data_type::GAUGE:
collected_value.values.push(v.d());
break;
case scollectd::data_type::COUNTER:
collected_value.values.push(v.ui());
break;
case scollectd::data_type::REAL_COUNTER:
collected_value.values.push(v.d());
case scollectd::data_type::DERIVE:
collected_value.values.push(v.i());
break;
default:
collected_value.values.push(v.ui());
@@ -52,7 +62,7 @@ static const char* str_to_regex(const sstring& v) {
}
void set_collectd(http_context& ctx, routes& r) {
cd::get_collectd.set(r, [](std::unique_ptr<request> req) {
cd::get_collectd.set(r, [&ctx](std::unique_ptr<request> req) {
auto id = ::make_shared<scollectd::type_instance_id>(req->param["pluginid"],
req->get_query_param("instance"), req->get_query_param("type"),

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -12,6 +25,6 @@
namespace api {
void set_collectd(http_context& ctx, httpd::routes& r);
void set_collectd(http_context& ctx, routes& r);
}

File diff suppressed because it is too large Load Diff

View File

@@ -3,37 +3,45 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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"
#include "api/api-doc/column_family.json.hh"
#include "replica/database.hh"
#include "database.hh"
#include <seastar/core/future-util.hh>
#include <any>
namespace db {
class system_keyspace;
}
namespace api {
void set_column_family(http_context& ctx, httpd::routes& r, sharded<db::system_keyspace>& sys_ks);
void unset_column_family(http_context& ctx, httpd::routes& r);
void set_column_family(http_context& ctx, routes& r);
const table_id& get_uuid(const sstring& name, const replica::database& db);
future<> foreach_column_family(http_context& ctx, const sstring& name, std::function<void(replica::column_family&)> f);
const utils::UUID& get_uuid(const sstring& name, const database& db);
future<> foreach_column_family(http_context& ctx, const sstring& name, std::function<void(column_family&)> f);
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>(replica::database&)>;
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>)>;
return ctx.db.map_reduce0(mapper_type([mapper, uuid](replica::database& db) {
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)))));
@@ -60,15 +68,15 @@ 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 replica::column_family&)> f);
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>(replica::column_family&)> mapper;
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()(replica::database& db) const {
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<table_id, seastar::lw_shared_ptr<replica::table>>& i) {
return do_for_each(db.get_column_families(), [res, this](const std::pair<utils::UUID, seastar::lw_shared_ptr<table>>& i) {
*res = reducer(std::move(*res), mapper(*i.second.get()));
}).then([res] {
return std::move(*res);
@@ -79,9 +87,9 @@ struct map_reduce_column_families_locally {
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>(replica::column_family&)>;
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>)>;
auto wrapped_mapper = mapper_type([mapper = std::move(mapper)] (replica::column_family& cf) mutable {
auto wrapped_mapper = mapper_type([mapper = std::move(mapper)] (column_family& cf) mutable {
return std::make_unique<std::any>(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 {
@@ -103,10 +111,10 @@ 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 replica::column_family_stats::*f);
int64_t column_family_stats::*f);
future<json::json_return_type> get_cf_stats(http_context& ctx,
int64_t replica::column_family_stats::*f);
int64_t column_family_stats::*f);
std::tuple<sstring, sstring> parse_fully_qualified_cf_name(sstring name);

View File

@@ -3,23 +3,35 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "commitlog.hh"
#include "db/commitlog/commitlog.hh"
#include "api/api-doc/commitlog.json.hh"
#include "replica/database.hh"
#include "database.hh"
#include <vector>
namespace api {
using namespace seastar::httpd;
template<typename T>
static auto acquire_cl_metric(http_context& ctx, std::function<T (db::commitlog*)> func) {
typedef T ret_type;
return ctx.db.map_reduce0([func = std::move(func)](replica::database& db) {
return ctx.db.map_reduce0([func = std::move(func)](database& db) {
if (db.commitlog() == nullptr) {
return make_ready_future<ret_type>();
}
@@ -35,7 +47,7 @@ void set_commitlog(http_context& ctx, routes& r) {
auto res = make_shared<std::vector<sstring>>();
return ctx.db.map_reduce([res](std::vector<sstring> names) {
res->insert(res->end(), names.begin(), names.end());
}, [](replica::database& db) {
}, [](database& db) {
if (db.commitlog() == nullptr) {
return make_ready_future<std::vector<sstring>>(std::vector<sstring>());
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -12,6 +25,6 @@
namespace api {
void set_commitlog(http_context& ctx, httpd::routes& r);
void set_commitlog(http_context& ctx, routes& r);
}

View File

@@ -3,30 +3,38 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 <seastar/core/coroutine.hh>
#include "compaction_manager.hh"
#include "compaction/compaction_manager.hh"
#include "api/api-doc/compaction_manager.json.hh"
#include "db/system_keyspace.hh"
#include "column_family.hh"
#include "unimplemented.hh"
#include "storage_service.hh"
#include <utility>
namespace api {
namespace cm = httpd::compaction_manager_json;
using namespace json;
using namespace seastar::httpd;
static future<json::json_return_type> get_cm_stats(http_context& ctx,
int64_t compaction_manager::stats::*f) {
return ctx.db.map_reduce0([f](replica::database& db) {
return ctx.db.map_reduce0([f](database& db) {
return db.get_compaction_manager().get_stats().*f;
}, int64_t(0), std::plus<int64_t>()).then([](const int64_t& res) {
return make_ready_future<json::json_return_type>(res);
@@ -42,9 +50,10 @@ static std::unordered_map<std::pair<sstring, sstring>, uint64_t, utils::tuple_ha
return std::move(a);
}
void set_compaction_manager(http_context& ctx, routes& r) {
cm::get_compactions.set(r, [&ctx] (std::unique_ptr<http::request> req) {
return ctx.db.map_reduce0([](replica::database& db) {
cm::get_compactions.set(r, [&ctx] (std::unique_ptr<request> req) {
return ctx.db.map_reduce0([](database& db) {
std::vector<cm::summary> summaries;
const compaction_manager& cm = db.get_compaction_manager();
@@ -65,12 +74,12 @@ void set_compaction_manager(http_context& ctx, routes& r) {
});
});
cm::get_pending_tasks_by_table.set(r, [&ctx] (std::unique_ptr<http::request> req) {
return ctx.db.map_reduce0([](replica::database& db) {
return do_with(std::unordered_map<std::pair<sstring, sstring>, uint64_t, utils::tuple_hash>(), [&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<table_id, seastar::lw_shared_ptr<replica::table>>& i) -> future<> {
replica::table& cf = *i.second.get();
tasks[std::make_pair(cf.schema()->ks_name(), cf.schema()->cf_name())] = cf.estimate_pending_compactions();
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);
@@ -91,71 +100,52 @@ void set_compaction_manager(http_context& ctx, routes& r) {
});
});
cm::force_user_defined_compaction.set(r, [] (std::unique_ptr<http::request> req) {
cm::force_user_defined_compaction.set(r, [] (std::unique_ptr<request> req) {
//TBD
// FIXME
warn(unimplemented::cause::API);
return make_ready_future<json::json_return_type>(json_void());
});
cm::stop_compaction.set(r, [&ctx] (std::unique_ptr<http::request> req) {
cm::stop_compaction.set(r, [&ctx] (std::unique_ptr<request> req) {
auto type = req->get_query_param("type");
return ctx.db.invoke_on_all([type] (replica::database& db) {
return ctx.db.invoke_on_all([type] (database& db) {
auto& cm = db.get_compaction_manager();
return cm.stop_compaction(type);
cm.stop_compaction(type);
}).then([] {
return make_ready_future<json::json_return_type>(json_void());
});
});
cm::stop_keyspace_compaction.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
auto ks_name = validate_keyspace(ctx, req->param);
auto table_names = parse_tables(ks_name, ctx, req->query_parameters, "tables");
if (table_names.empty()) {
table_names = map_keys(ctx.db.local().find_keyspace(ks_name).metadata().get()->cf_meta_data());
}
auto type = req->get_query_param("type");
co_await ctx.db.invoke_on_all([&ks_name, &table_names, type] (replica::database& db) {
auto& cm = db.get_compaction_manager();
return parallel_for_each(table_names, [&db, &cm, &ks_name, type] (sstring& table_name) {
auto& t = db.find_column_family(ks_name, table_name);
return t.parallel_foreach_table_state([&] (compaction::table_state& ts) {
return cm.stop_compaction(type, &ts);
});
});
});
co_return json_void();
});
cm::get_pending_tasks.set(r, [&ctx] (std::unique_ptr<http::request> req) {
return map_reduce_cf(ctx, int64_t(0), [](replica::column_family& cf) {
return cf.estimate_pending_compactions();
cm::get_pending_tasks.set(r, [&ctx] (std::unique_ptr<request> req) {
return map_reduce_cf(ctx, int64_t(0), [](column_family& cf) {
return cf.get_compaction_strategy().estimated_pending_compactions(cf);
}, std::plus<int64_t>());
});
cm::get_completed_tasks.set(r, [&ctx] (std::unique_ptr<http::request> req) {
cm::get_completed_tasks.set(r, [&ctx] (std::unique_ptr<request> req) {
return get_cm_stats(ctx, &compaction_manager::stats::completed_tasks);
});
cm::get_total_compactions_completed.set(r, [] (std::unique_ptr<http::request> req) {
cm::get_total_compactions_completed.set(r, [] (std::unique_ptr<request> req) {
// FIXME
// We are currently dont have an API for compaction
// so returning a 0 as the number of total compaction is ok
return make_ready_future<json::json_return_type>(0);
});
cm::get_bytes_compacted.set(r, [] (std::unique_ptr<http::request> req) {
cm::get_bytes_compacted.set(r, [] (std::unique_ptr<request> req) {
//TBD
// FIXME
warn(unimplemented::cause::API);
return make_ready_future<json::json_return_type>(0);
});
cm::get_compaction_history.set(r, [&ctx] (std::unique_ptr<http::request> req) {
std::function<future<>(output_stream<char>&&)> f = [&ctx](output_stream<char>&& s) {
return do_with(output_stream<char>(std::move(s)), true, [&ctx] (output_stream<char>& s, bool& first){
return s.write("[").then([&ctx, &s, &first] {
return ctx.db.local().get_compaction_manager().get_compaction_history([&s, &first](const db::compaction_history_entry& entry) mutable {
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);
@@ -185,7 +175,7 @@ void set_compaction_manager(http_context& ctx, routes& r) {
return make_ready_future<json::json_return_type>(std::move(f));
});
cm::get_compaction_info.set(r, [] (std::unique_ptr<http::request> req) {
cm::get_compaction_info.set(r, [] (std::unique_ptr<request> req) {
//TBD
// FIXME
warn(unimplemented::cause::API);

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -12,6 +25,6 @@
namespace api {
void set_compaction_manager(http_context& ctx, httpd::routes& r);
void set_compaction_manager(http_context& ctx, routes& r);
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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/config.hh"
@@ -13,7 +26,6 @@
#include <boost/algorithm/string/replace.hpp>
namespace api {
using namespace seastar::httpd;
template<class T>
json::json_return_type get_json_return_type(const T& val) {

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -13,5 +26,5 @@
namespace api {
void set_config(std::shared_ptr<httpd::api_registry_builder20> rb, http_context& ctx, httpd::routes& r, const db::config& cfg);
void set_config(std::shared_ptr<api_registry_builder20> rb, http_context& ctx, routes& r, const db::config& cfg);
}

View File

@@ -3,67 +3,46 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "locator/token_metadata.hh"
#include "locator/snitch_base.hh"
#include "locator/production_snitch_base.hh"
#include "endpoint_snitch.hh"
#include "api/api-doc/endpoint_snitch_info.json.hh"
#include "api/api-doc/storage_service.json.hh"
#include "utils/fb_utilities.hh"
namespace api {
using namespace seastar::httpd;
void set_endpoint_snitch(http_context& ctx, routes& r, sharded<locator::snitch_ptr>& snitch) {
void set_endpoint_snitch(http_context& ctx, routes& r) {
static auto host_or_broadcast = [](const_req req) {
auto host = req.get_query_param("host");
return host.empty() ? gms::inet_address(utils::fb_utilities::get_broadcast_address()) : gms::inet_address(host);
};
httpd::endpoint_snitch_info_json::get_datacenter.set(r, [&ctx](const_req req) {
auto& topology = ctx.shared_token_metadata.local().get()->get_topology();
auto ep = host_or_broadcast(req);
if (!topology.has_endpoint(ep)) {
// Cannot return error here, nodetool status can race, request
// info about just-left node and not handle it nicely
return locator::endpoint_dc_rack::default_location.dc;
}
return topology.get_datacenter(ep);
httpd::endpoint_snitch_info_json::get_datacenter.set(r, [](const_req req) {
return locator::i_endpoint_snitch::get_local_snitch_ptr()->get_datacenter(host_or_broadcast(req));
});
httpd::endpoint_snitch_info_json::get_rack.set(r, [&ctx](const_req req) {
auto& topology = ctx.shared_token_metadata.local().get()->get_topology();
auto ep = host_or_broadcast(req);
if (!topology.has_endpoint(ep)) {
// Cannot return error here, nodetool status can race, request
// info about just-left node and not handle it nicely
return locator::endpoint_dc_rack::default_location.rack;
}
return topology.get_rack(ep);
httpd::endpoint_snitch_info_json::get_rack.set(r, [](const_req req) {
return locator::i_endpoint_snitch::get_local_snitch_ptr()->get_rack(host_or_broadcast(req));
});
httpd::endpoint_snitch_info_json::get_snitch_name.set(r, [&snitch] (const_req req) {
return snitch.local()->get_name();
httpd::endpoint_snitch_info_json::get_snitch_name.set(r, [] (const_req req) {
return locator::i_endpoint_snitch::get_local_snitch_ptr()->get_name();
});
httpd::storage_service_json::update_snitch.set(r, [&snitch](std::unique_ptr<request> req) {
locator::snitch_config cfg;
cfg.name = req->get_query_param("ep_snitch_class_name");
return locator::i_endpoint_snitch::reset_snitch(snitch, cfg).then([] {
return make_ready_future<json::json_return_type>(json::json_void());
});
});
}
void unset_endpoint_snitch(http_context& ctx, routes& r) {
httpd::endpoint_snitch_info_json::get_datacenter.unset(r);
httpd::endpoint_snitch_info_json::get_rack.unset(r);
httpd::endpoint_snitch_info_json::get_snitch_name.unset(r);
httpd::storage_service_json::update_snitch.unset(r);
}
}

View File

@@ -3,20 +3,28 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 locator {
class snitch_ptr;
}
namespace api {
void set_endpoint_snitch(http_context& ctx, httpd::routes& r, sharded<locator::snitch_ptr>&);
void unset_endpoint_snitch(http_context& ctx, httpd::routes& r);
void set_endpoint_snitch(http_context& ctx, routes& r);
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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"
@@ -12,10 +25,9 @@
#include <seastar/http/exception.hh>
#include "log.hh"
#include "utils/error_injection.hh"
#include <seastar/core/future-util.hh>
#include "seastar/core/future-util.hh"
namespace api {
using namespace seastar::httpd;
namespace hf = httpd::error_injection_json;

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -12,6 +25,6 @@
namespace api {
void set_error_injection(http_context& ctx, httpd::routes& r);
void set_error_injection(http_context& ctx, routes& r);
}

View File

@@ -3,36 +3,48 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "failure_detector.hh"
#include "api/api-doc/failure_detector.json.hh"
#include "gms/failure_detector.hh"
#include "gms/application_state.hh"
#include "gms/gossiper.hh"
namespace api {
using namespace seastar::httpd;
namespace fd = httpd::failure_detector_json;
void set_failure_detector(http_context& ctx, routes& r, gms::gossiper& g) {
fd::get_all_endpoint_states.set(r, [&g](std::unique_ptr<request> req) {
std::vector<fd::endpoint_state> res;
for (auto i : g.get_endpoint_states()) {
for (auto i : g.endpoint_state_map) {
fd::endpoint_state val;
val.addrs = fmt::to_string(i.first);
val.addrs = boost::lexical_cast<std::string>(i.first);
val.is_alive = i.second.is_alive();
val.generation = i.second.get_heart_beat_state().get_generation().value();
val.version = i.second.get_heart_beat_state().get_heart_beat_version().value();
val.generation = i.second.get_heart_beat_state().get_generation();
val.version = i.second.get_heart_beat_state().get_heart_beat_version();
val.update_time = i.second.get_update_timestamp().time_since_epoch().count();
for (auto a : i.second.get_application_state_map()) {
fd::version_value version_val;
// We return the enum index and not it's name to stay compatible to origin
// method that the state index are static but the name can be changed.
version_val.application_state = static_cast<std::underlying_type<gms::application_state>::type>(a.first);
version_val.value = a.second.value();
version_val.version = a.second.version().value();
version_val.value = a.second.value;
version_val.version = a.second.version;
val.application_state.push(version_val);
}
res.push_back(val);
@@ -41,49 +53,54 @@ void set_failure_detector(http_context& ctx, routes& r, gms::gossiper& g) {
});
fd::get_up_endpoint_count.set(r, [&g](std::unique_ptr<request> req) {
int res = g.get_up_endpoint_count();
return make_ready_future<json::json_return_type>(res);
return gms::get_up_endpoint_count(g).then([](int res) {
return make_ready_future<json::json_return_type>(res);
});
});
fd::get_down_endpoint_count.set(r, [&g](std::unique_ptr<request> req) {
int res = g.get_down_endpoint_count();
return make_ready_future<json::json_return_type>(res);
return gms::get_down_endpoint_count(g).then([](int res) {
return make_ready_future<json::json_return_type>(res);
});
});
fd::get_phi_convict_threshold.set(r, [] (std::unique_ptr<request> req) {
return make_ready_future<json::json_return_type>(8);
return gms::get_phi_convict_threshold().then([](double res) {
return make_ready_future<json::json_return_type>(res);
});
});
fd::get_simple_states.set(r, [&g] (std::unique_ptr<request> req) {
std::map<sstring, sstring> nodes_status;
for (auto& entry : g.get_endpoint_states()) {
nodes_status.emplace(entry.first.to_sstring(), entry.second.is_alive() ? "UP" : "DOWN");
}
return make_ready_future<json::json_return_type>(map_to_key_value<fd::mapper>(nodes_status));
return gms::get_simple_states(g).then([](const std::map<sstring, sstring>& map) {
return make_ready_future<json::json_return_type>(map_to_key_value<fd::mapper>(map));
});
});
fd::set_phi_convict_threshold.set(r, [](std::unique_ptr<request> req) {
// TBD
unimplemented();
std::ignore = atof(req->get_query_param("phi").c_str());
return make_ready_future<json::json_return_type>("");
double phi = atof(req->get_query_param("phi").c_str());
return gms::set_phi_convict_threshold(phi).then([]() {
return make_ready_future<json::json_return_type>("");
});
});
fd::get_endpoint_state.set(r, [&g] (std::unique_ptr<request> req) {
auto* state = g.get_endpoint_state_for_endpoint_ptr(gms::inet_address(req->param["addr"]));
if (!state) {
return make_ready_future<json::json_return_type>(format("unknown endpoint {}", req->param["addr"]));
}
std::stringstream ss;
g.append_endpoint_state(ss, *state);
return make_ready_future<json::json_return_type>(sstring(ss.str()));
return get_endpoint_state(g, req->param["addr"]).then([](const sstring& state) {
return make_ready_future<json::json_return_type>(state);
});
});
fd::get_endpoint_phi_values.set(r, [](std::unique_ptr<request> req) {
// We no longer have a phi failure detector,
// just returning the empty value is good enough.
std::vector<fd::endpoint_phi_value> res;
return make_ready_future<json::json_return_type>(res);
return gms::get_arrival_samples().then([](std::map<gms::inet_address, gms::arrival_window> map) {
std::vector<fd::endpoint_phi_value> res;
auto now = gms::arrival_window::clk::now();
for (auto& p : map) {
fd::endpoint_phi_value val;
val.endpoint = p.first.to_sstring();
val.phi = p.second.phi(now);
res.emplace_back(std::move(val));
}
return make_ready_future<json::json_return_type>(res);
});
});
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -18,6 +31,6 @@ class gossiper;
namespace api {
void set_failure_detector(http_context& ctx, httpd::routes& r, gms::gossiper& g);
void set_failure_detector(http_context& ctx, routes& r, gms::gossiper& g);
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "gossiper.hh"
@@ -11,7 +24,6 @@
#include "gms/gossiper.hh"
namespace api {
using namespace seastar::httpd;
using namespace json;
void set_gossiper(http_context& ctx, routes& r, gms::gossiper& g) {
@@ -20,11 +32,9 @@ void set_gossiper(http_context& ctx, routes& r, gms::gossiper& g) {
return container_to_vec(res);
});
httpd::gossiper_json::get_live_endpoint.set(r, [&g] (std::unique_ptr<request> req) {
return g.get_live_members_synchronized().then([] (auto res) {
return make_ready_future<json::json_return_type>(container_to_vec(res));
});
httpd::gossiper_json::get_live_endpoint.set(r, [&g] (const_req req) {
auto res = g.get_live_members();
return container_to_vec(res);
});
httpd::gossiper_json::get_endpoint_downtime.set(r, [&g] (const_req req) {
@@ -32,21 +42,21 @@ void set_gossiper(http_context& ctx, routes& r, gms::gossiper& g) {
return g.get_endpoint_downtime(ep);
});
httpd::gossiper_json::get_current_generation_number.set(r, [&g] (std::unique_ptr<http::request> req) {
httpd::gossiper_json::get_current_generation_number.set(r, [&g] (std::unique_ptr<request> req) {
gms::inet_address ep(req->param["addr"]);
return g.get_current_generation_number(ep).then([] (gms::generation_type res) {
return make_ready_future<json::json_return_type>(res.value());
return g.get_current_generation_number(ep).then([] (int res) {
return make_ready_future<json::json_return_type>(res);
});
});
httpd::gossiper_json::get_current_heart_beat_version.set(r, [&g] (std::unique_ptr<http::request> req) {
httpd::gossiper_json::get_current_heart_beat_version.set(r, [&g] (std::unique_ptr<request> req) {
gms::inet_address ep(req->param["addr"]);
return g.get_current_heart_beat_version(ep).then([] (gms::version_type res) {
return make_ready_future<json::json_return_type>(res.value());
return g.get_current_heart_beat_version(ep).then([] (int res) {
return make_ready_future<json::json_return_type>(res);
});
});
httpd::gossiper_json::assassinate_endpoint.set(r, [&g](std::unique_ptr<http::request> req) {
httpd::gossiper_json::assassinate_endpoint.set(r, [&g](std::unique_ptr<request> req) {
if (req->get_query_param("unsafe") != "True") {
return g.assassinate_endpoint(req->param["addr"]).then([] {
return make_ready_future<json::json_return_type>(json_void());
@@ -57,7 +67,7 @@ void set_gossiper(http_context& ctx, routes& r, gms::gossiper& g) {
});
});
httpd::gossiper_json::force_remove_endpoint.set(r, [&g](std::unique_ptr<http::request> req) {
httpd::gossiper_json::force_remove_endpoint.set(r, [&g](std::unique_ptr<request> req) {
gms::inet_address ep(req->param["addr"]);
return g.force_remove_endpoint(ep).then([] {
return make_ready_future<json::json_return_type>(json_void());

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -18,6 +31,6 @@ class gossiper;
namespace api {
void set_gossiper(http_context& ctx, httpd::routes& r, gms::gossiper& g);
void set_gossiper(http_context& ctx, routes& r, gms::gossiper& g);
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 <algorithm>
@@ -19,11 +32,10 @@
namespace api {
using namespace json;
using namespace seastar::httpd;
namespace hh = httpd::hinted_handoff_json;
void set_hinted_handoff(http_context& ctx, routes& r, gms::gossiper& g) {
hh::create_hints_sync_point.set(r, [&ctx, &g] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
hh::create_hints_sync_point.set(r, [&ctx, &g] (std::unique_ptr<request> req) -> future<json::json_return_type> {
auto parse_hosts_list = [&g] (sstring arg) {
std::vector<sstring> hosts_str = split(arg, ",");
std::vector<gms::inet_address> hosts;
@@ -53,7 +65,7 @@ void set_hinted_handoff(http_context& ctx, routes& r, gms::gossiper& g) {
});
});
hh::get_hints_sync_point.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
hh::get_hints_sync_point.set(r, [&ctx] (std::unique_ptr<request> req) -> future<json::json_return_type> {
db::hints::sync_point sync_point;
const sstring encoded = req->get_query_param("id");
try {
@@ -94,42 +106,42 @@ void set_hinted_handoff(http_context& ctx, routes& r, gms::gossiper& g) {
});
});
hh::list_endpoints_pending_hints.set(r, [] (std::unique_ptr<http::request> req) {
hh::list_endpoints_pending_hints.set(r, [] (std::unique_ptr<request> req) {
//TBD
unimplemented();
std::vector<sstring> res;
return make_ready_future<json::json_return_type>(res);
});
hh::truncate_all_hints.set(r, [] (std::unique_ptr<http::request> req) {
hh::truncate_all_hints.set(r, [] (std::unique_ptr<request> req) {
//TBD
unimplemented();
sstring host = req->get_query_param("host");
return make_ready_future<json::json_return_type>(json_void());
});
hh::schedule_hint_delivery.set(r, [] (std::unique_ptr<http::request> req) {
hh::schedule_hint_delivery.set(r, [] (std::unique_ptr<request> req) {
//TBD
unimplemented();
sstring host = req->get_query_param("host");
return make_ready_future<json::json_return_type>(json_void());
});
hh::pause_hints_delivery.set(r, [] (std::unique_ptr<http::request> req) {
hh::pause_hints_delivery.set(r, [] (std::unique_ptr<request> req) {
//TBD
unimplemented();
sstring pause = req->get_query_param("pause");
return make_ready_future<json::json_return_type>(json_void());
});
hh::get_create_hint_count.set(r, [] (std::unique_ptr<http::request> req) {
hh::get_create_hint_count.set(r, [] (std::unique_ptr<request> req) {
//TBD
unimplemented();
sstring host = req->get_query_param("host");
return make_ready_future<json::json_return_type>(0);
});
hh::get_not_stored_hints_count.set(r, [] (std::unique_ptr<http::request> req) {
hh::get_not_stored_hints_count.set(r, [] (std::unique_ptr<request> req) {
//TBD
unimplemented();
sstring host = req->get_query_param("host");

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -18,7 +31,7 @@ class gossiper;
namespace api {
void set_hinted_handoff(http_context& ctx, httpd::routes& r, gms::gossiper& g);
void unset_hinted_handoff(http_context& ctx, httpd::routes& r);
void set_hinted_handoff(http_context& ctx, routes& r, gms::gossiper& g);
void unset_hinted_handoff(http_context& ctx, routes& r);
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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/lsa.json.hh"
@@ -13,17 +26,16 @@
#include <seastar/http/exception.hh>
#include "utils/logalloc.hh"
#include "log.hh"
#include "replica/database.hh"
#include "database.hh"
namespace api {
using namespace seastar::httpd;
static logging::logger alogger("lsa-api");
void set_lsa(http_context& ctx, routes& r) {
httpd::lsa_json::lsa_compact.set(r, [&ctx](std::unique_ptr<request> req) {
alogger.info("Triggering compaction");
return ctx.db.invoke_on_all([] (replica::database&) {
return ctx.db.invoke_on_all([] (database&) {
logalloc::shard_tracker().reclaim(std::numeric_limits<size_t>::max());
}).then([] {
return json::json_return_type(json::json_void());

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -12,6 +25,6 @@
namespace api {
void set_lsa(http_context& ctx, httpd::routes& r);
void set_lsa(http_context& ctx, routes& r);
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "messaging_service.hh"
@@ -13,7 +26,6 @@
#include <iostream>
#include <sstream>
using namespace seastar::httpd;
using namespace httpd::messaging_service_json;
using namespace netw;
@@ -29,7 +41,7 @@ std::vector<message_counter> map_to_message_counters(
std::vector<message_counter> res;
for (auto i : map) {
res.push_back(message_counter());
res.back().key = fmt::to_string(i.first);
res.back().key = boost::lexical_cast<sstring>(i.first);
res.back().value = i.second;
}
return res;

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -14,7 +27,7 @@ namespace netw { class messaging_service; }
namespace api {
void set_messaging_service(http_context& ctx, httpd::routes& r, sharded<netw::messaging_service>& ms);
void unset_messaging_service(http_context& ctx, httpd::routes& r);
void set_messaging_service(http_context& ctx, routes& r, sharded<netw::messaging_service>& ms);
void unset_messaging_service(http_context& ctx, routes& r);
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "storage_proxy.hh"
@@ -13,19 +26,15 @@
#include "service/storage_service.hh"
#include "db/config.hh"
#include "utils/histogram.hh"
#include "replica/database.hh"
#include <seastar/core/scheduling_specific.hh>
#include "database.hh"
#include "seastar/core/scheduling_specific.hh"
namespace api {
namespace sp = httpd::storage_proxy_json;
using proxy = service::storage_proxy;
using namespace seastar::httpd;
using namespace json;
utils::time_estimated_histogram timed_rate_moving_average_summary_merge(utils::time_estimated_histogram a, const utils::timed_rate_moving_average_summary_and_histogram& b) {
return a.merge(b.histogram());
}
/**
* This function implement a two dimentional map reduce where
@@ -59,10 +68,10 @@ future<V> two_dimensional_map_reduce(distributed<service::storage_proxy>& d,
* @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, typename C>
template<typename V, typename Reducer, typename F>
future<V> two_dimensional_map_reduce(distributed<service::storage_proxy>& d,
C F::*f, Reducer reducer, V initial_value) {
return two_dimensional_map_reduce(d, [f] (F& stats) -> V {
V F::*f, Reducer reducer, V initial_value) {
return two_dimensional_map_reduce(d, [f] (F& stats) {
return stats.*f;
}, reducer, initial_value);
}
@@ -116,10 +125,10 @@ utils_json::estimated_histogram time_to_json_histogram(const utils::time_estimat
return res;
}
static future<json::json_return_type> sum_estimated_histogram(http_context& ctx, utils::timed_rate_moving_average_summary_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).histogram();
}, utils::time_estimated_histogram_merge, utils::time_estimated_histogram()).then([](const utils::time_estimated_histogram& val) {
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));
});
}
@@ -134,7 +143,7 @@ static future<json::json_return_type> sum_estimated_histogram(http_context& ctx
});
}
static future<json::json_return_type> total_latency(http_context& ctx, utils::timed_rate_moving_average_summary_and_histogram service::storage_proxy_stats::stats::*f) {
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) {
@@ -154,7 +163,7 @@ static future<json::json_return_type> total_latency(http_context& ctx, utils::t
template<typename F>
future<json::json_return_type>
sum_histogram_stats_storage_proxy(distributed<proxy>& d,
utils::timed_rate_moving_average_summary_and_histogram F::*f) {
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()).
@@ -174,7 +183,7 @@ sum_histogram_stats_storage_proxy(distributed<proxy>& d,
template<typename F>
future<json::json_return_type>
sum_timer_stats_storage_proxy(distributed<proxy>& d,
utils::timed_rate_moving_average_summary_and_histogram F::*f) {
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();
@@ -185,75 +194,75 @@ sum_timer_stats_storage_proxy(distributed<proxy>& d,
}
void set_storage_proxy(http_context& ctx, routes& r, sharded<service::storage_service>& ss) {
sp::get_total_hints.set(r, [](std::unique_ptr<http::request> req) {
sp::get_total_hints.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
return make_ready_future<json::json_return_type>(0);
});
sp::get_hinted_handoff_enabled.set(r, [&ctx](std::unique_ptr<http::request> req) {
const auto& filter = ctx.sp.local().get_hints_host_filter();
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::set_hinted_handoff_enabled.set(r, [&ctx](std::unique_ptr<http::request> req) {
sp::set_hinted_handoff_enabled.set(r, [](std::unique_ptr<request> req) {
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 ctx.sp.invoke_on_all([filter = std::move(filter)] (service::storage_proxy& sp) {
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());
});
});
sp::get_hinted_handoff_enabled_by_dc.set(r, [&ctx](std::unique_ptr<http::request> req) {
sp::get_hinted_handoff_enabled_by_dc.set(r, [](std::unique_ptr<request> req) {
std::vector<sstring> res;
const auto& filter = ctx.sp.local().get_hints_host_filter();
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));
return make_ready_future<json::json_return_type>(res);
});
sp::set_hinted_handoff_enabled_by_dc_list.set(r, [&ctx](std::unique_ptr<http::request> req) {
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 ctx.sp.invoke_on_all([filter = std::move(filter)] (service::storage_proxy& sp) {
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());
});
});
sp::get_max_hint_window.set(r, [](std::unique_ptr<http::request> req) {
sp::get_max_hint_window.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
return make_ready_future<json::json_return_type>(0);
});
sp::set_max_hint_window.set(r, [](std::unique_ptr<http::request> req) {
sp::set_max_hint_window.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto enable = req->get_query_param("ms");
return make_ready_future<json::json_return_type>(json_void());
});
sp::get_max_hints_in_progress.set(r, [](std::unique_ptr<http::request> req) {
sp::get_max_hints_in_progress.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
return make_ready_future<json::json_return_type>(1);
});
sp::set_max_hints_in_progress.set(r, [](std::unique_ptr<http::request> req) {
sp::set_max_hints_in_progress.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto enable = req->get_query_param("qs");
return make_ready_future<json::json_return_type>(json_void());
});
sp::get_hints_in_progress.set(r, [](std::unique_ptr<http::request> req) {
sp::get_hints_in_progress.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
return make_ready_future<json::json_return_type>(0);
@@ -263,7 +272,7 @@ void set_storage_proxy(http_context& ctx, routes& r, sharded<service::storage_se
return ctx.db.local().get_config().request_timeout_in_ms()/1000.0;
});
sp::set_rpc_timeout.set(r, [](std::unique_ptr<http::request> req) {
sp::set_rpc_timeout.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto enable = req->get_query_param("timeout");
@@ -274,7 +283,7 @@ void set_storage_proxy(http_context& ctx, routes& r, sharded<service::storage_se
return ctx.db.local().get_config().read_request_timeout_in_ms()/1000.0;
});
sp::set_read_rpc_timeout.set(r, [](std::unique_ptr<http::request> req) {
sp::set_read_rpc_timeout.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto enable = req->get_query_param("timeout");
@@ -285,7 +294,7 @@ void set_storage_proxy(http_context& ctx, routes& r, sharded<service::storage_se
return ctx.db.local().get_config().write_request_timeout_in_ms()/1000.0;
});
sp::set_write_rpc_timeout.set(r, [](std::unique_ptr<http::request> req) {
sp::set_write_rpc_timeout.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto enable = req->get_query_param("timeout");
@@ -296,7 +305,7 @@ void set_storage_proxy(http_context& ctx, routes& r, sharded<service::storage_se
return ctx.db.local().get_config().counter_write_request_timeout_in_ms()/1000.0;
});
sp::set_counter_write_rpc_timeout.set(r, [](std::unique_ptr<http::request> req) {
sp::set_counter_write_rpc_timeout.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto enable = req->get_query_param("timeout");
@@ -307,7 +316,7 @@ void set_storage_proxy(http_context& ctx, routes& r, sharded<service::storage_se
return ctx.db.local().get_config().cas_contention_timeout_in_ms()/1000.0;
});
sp::set_cas_contention_timeout.set(r, [](std::unique_ptr<http::request> req) {
sp::set_cas_contention_timeout.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto enable = req->get_query_param("timeout");
@@ -318,7 +327,7 @@ void set_storage_proxy(http_context& ctx, routes& r, sharded<service::storage_se
return ctx.db.local().get_config().range_request_timeout_in_ms()/1000.0;
});
sp::set_range_rpc_timeout.set(r, [](std::unique_ptr<http::request> req) {
sp::set_range_rpc_timeout.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto enable = req->get_query_param("timeout");
@@ -329,32 +338,32 @@ void set_storage_proxy(http_context& ctx, routes& r, sharded<service::storage_se
return ctx.db.local().get_config().truncate_request_timeout_in_ms()/1000.0;
});
sp::set_truncate_rpc_timeout.set(r, [](std::unique_ptr<http::request> req) {
sp::set_truncate_rpc_timeout.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
auto enable = req->get_query_param("timeout");
return make_ready_future<json::json_return_type>(json_void());
});
sp::reload_trigger_classes.set(r, [](std::unique_ptr<http::request> req) {
sp::reload_trigger_classes.set(r, [](std::unique_ptr<request> req) {
//TBD
unimplemented();
return make_ready_future<json::json_return_type>(json_void());
});
sp::get_read_repair_attempted.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_read_repair_repaired_blocking.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_read_repair_repaired_background.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_schema_versions.set(r, [&ss](std::unique_ptr<http::request> req) {
sp::get_schema_versions.set(r, [&ss](std::unique_ptr<request> req) {
return ss.local().describe_schema_versions().then([] (auto result) {
std::vector<sp::mapper_list> res;
for (auto e : result) {
@@ -367,122 +376,122 @@ void set_storage_proxy(http_context& ctx, routes& r, sharded<service::storage_se
});
});
sp::get_cas_read_timeouts.set(r, [&ctx](std::unique_ptr<http::request> req) {
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_unavailables.set(r, [&ctx](std::unique_ptr<http::request> req) {
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_write_timeouts.set(r, [&ctx](std::unique_ptr<http::request> req) {
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_unavailables.set(r, [&ctx](std::unique_ptr<http::request> req) {
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_metrics_unfinished_commit.set(r, [&ctx](std::unique_ptr<http::request> req) {
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_contention.set(r, [&ctx](std::unique_ptr<http::request> req) {
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_condition_not_met.set(r, [&ctx](std::unique_ptr<http::request> req) {
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_failed_read_round_optimization.set(r, [&ctx](std::unique_ptr<http::request> req) {
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, [&ctx](std::unique_ptr<http::request> req) {
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, [&ctx](std::unique_ptr<http::request> req) {
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_read_metrics_timeouts.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_read_metrics_unavailables.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_range_metrics_timeouts.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_range_metrics_unavailables.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_write_metrics_timeouts.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_write_metrics_unavailables.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_read_metrics_timeouts_rates.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_read_metrics_unavailables_rates.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_range_metrics_timeouts_rates.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_range_metrics_unavailables_rates.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_write_metrics_timeouts_rates.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_write_metrics_unavailables_rates.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_range_metrics_latency_histogram_depricated.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_write_metrics_latency_histogram_depricated.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_read_metrics_latency_histogram_depricated.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_range_metrics_latency_histogram.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_write_metrics_latency_histogram.set(r, [&ctx](std::unique_ptr<http::request> req) {
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<http::request> req) {
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<http::request> req) {
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, [](std::unique_ptr<http::request> req) {
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
@@ -490,101 +499,32 @@ void set_storage_proxy(http_context& ctx, routes& r, sharded<service::storage_se
return make_ready_future<json::json_return_type>(get_empty_moving_average());
});
sp::get_read_metrics_latency_histogram.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_read_estimated_histogram.set(r, [&ctx](std::unique_ptr<http::request> req) {
return sum_estimated_histogram(ctx, &service::storage_proxy_stats::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);
});
sp::get_read_latency.set(r, [&ctx](std::unique_ptr<http::request> req) {
sp::get_read_latency.set(r, [&ctx](std::unique_ptr<request> req) {
return total_latency(ctx, &service::storage_proxy_stats::stats::read);
});
sp::get_write_estimated_histogram.set(r, [&ctx](std::unique_ptr<http::request> req) {
return sum_estimated_histogram(ctx, &service::storage_proxy_stats::stats::write);
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);
});
sp::get_write_latency.set(r, [&ctx](std::unique_ptr<http::request> req) {
sp::get_write_latency.set(r, [&ctx](std::unique_ptr<request> req) {
return total_latency(ctx, &service::storage_proxy_stats::stats::write);
});
sp::get_range_estimated_histogram.set(r, [&ctx](std::unique_ptr<http::request> req) {
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);
});
sp::get_range_latency.set(r, [&ctx](std::unique_ptr<http::request> req) {
sp::get_range_latency.set(r, [&ctx](std::unique_ptr<request> req) {
return total_latency(ctx, &service::storage_proxy_stats::stats::range);
});
}
void unset_storage_proxy(http_context& ctx, routes& r) {
sp::get_total_hints.unset(r);
sp::get_hinted_handoff_enabled.unset(r);
sp::set_hinted_handoff_enabled.unset(r);
sp::get_hinted_handoff_enabled_by_dc.unset(r);
sp::set_hinted_handoff_enabled_by_dc_list.unset(r);
sp::get_max_hint_window.unset(r);
sp::set_max_hint_window.unset(r);
sp::get_max_hints_in_progress.unset(r);
sp::set_max_hints_in_progress.unset(r);
sp::get_hints_in_progress.unset(r);
sp::get_rpc_timeout.unset(r);
sp::set_rpc_timeout.unset(r);
sp::get_read_rpc_timeout.unset(r);
sp::set_read_rpc_timeout.unset(r);
sp::get_write_rpc_timeout.unset(r);
sp::set_write_rpc_timeout.unset(r);
sp::get_counter_write_rpc_timeout.unset(r);
sp::set_counter_write_rpc_timeout.unset(r);
sp::get_cas_contention_timeout.unset(r);
sp::set_cas_contention_timeout.unset(r);
sp::get_range_rpc_timeout.unset(r);
sp::set_range_rpc_timeout.unset(r);
sp::get_truncate_rpc_timeout.unset(r);
sp::set_truncate_rpc_timeout.unset(r);
sp::reload_trigger_classes.unset(r);
sp::get_read_repair_attempted.unset(r);
sp::get_read_repair_repaired_blocking.unset(r);
sp::get_read_repair_repaired_background.unset(r);
sp::get_schema_versions.unset(r);
sp::get_cas_read_timeouts.unset(r);
sp::get_cas_read_unavailables.unset(r);
sp::get_cas_write_timeouts.unset(r);
sp::get_cas_write_unavailables.unset(r);
sp::get_cas_write_metrics_unfinished_commit.unset(r);
sp::get_cas_write_metrics_contention.unset(r);
sp::get_cas_write_metrics_condition_not_met.unset(r);
sp::get_cas_write_metrics_failed_read_round_optimization.unset(r);
sp::get_cas_read_metrics_unfinished_commit.unset(r);
sp::get_cas_read_metrics_contention.unset(r);
sp::get_read_metrics_timeouts.unset(r);
sp::get_read_metrics_unavailables.unset(r);
sp::get_range_metrics_timeouts.unset(r);
sp::get_range_metrics_unavailables.unset(r);
sp::get_write_metrics_timeouts.unset(r);
sp::get_write_metrics_unavailables.unset(r);
sp::get_read_metrics_timeouts_rates.unset(r);
sp::get_read_metrics_unavailables_rates.unset(r);
sp::get_range_metrics_timeouts_rates.unset(r);
sp::get_range_metrics_unavailables_rates.unset(r);
sp::get_write_metrics_timeouts_rates.unset(r);
sp::get_write_metrics_unavailables_rates.unset(r);
sp::get_range_metrics_latency_histogram_depricated.unset(r);
sp::get_write_metrics_latency_histogram_depricated.unset(r);
sp::get_read_metrics_latency_histogram_depricated.unset(r);
sp::get_range_metrics_latency_histogram.unset(r);
sp::get_write_metrics_latency_histogram.unset(r);
sp::get_cas_write_metrics_latency_histogram.unset(r);
sp::get_cas_read_metrics_latency_histogram.unset(r);
sp::get_view_write_metrics_latency_histogram.unset(r);
sp::get_read_metrics_latency_histogram.unset(r);
sp::get_read_estimated_histogram.unset(r);
sp::get_read_latency.unset(r);
sp::get_write_estimated_histogram.unset(r);
sp::get_write_latency.unset(r);
sp::get_range_estimated_histogram.unset(r);
sp::get_range_latency.unset(r);
}
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -15,7 +28,6 @@ namespace service { class storage_service; }
namespace api {
void set_storage_proxy(http_context& ctx, httpd::routes& r, sharded<service::storage_service>& ss);
void unset_storage_proxy(http_context& ctx, httpd::routes& r);
void set_storage_proxy(http_context& ctx, routes& r, sharded<service::storage_service>& ss);
}

File diff suppressed because it is too large Load Diff

View File

@@ -3,13 +3,24 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 <iostream>
#include <seastar/core/sharded.hh>
#include "api.hh"
#include "db/data_listeners.hh"
@@ -21,7 +32,6 @@ class snapshot_ctl;
namespace view {
class view_builder;
}
class system_keyspace;
}
namespace netw { class messaging_service; }
class repair_service;
@@ -36,52 +46,19 @@ class gossiper;
namespace api {
// verify that the keyspace is found, otherwise a bad_param_exception exception is thrown
// containing the description of the respective keyspace error.
sstring validate_keyspace(http_context& ctx, sstring ks_name);
// verify that the keyspace parameter is found, otherwise a bad_param_exception exception is thrown
// containing the description of the respective keyspace error.
sstring validate_keyspace(http_context& ctx, const httpd::parameters& param);
// splits a request parameter assumed to hold a comma-separated list of table names
// verify that the tables are found, otherwise a bad_param_exception exception is thrown
// containing the description of the respective no_such_column_family error.
// Returns an empty vector if no parameter was found.
// If the parameter is found and empty, returns a list of all table names in the keyspace.
std::vector<sstring> parse_tables(const sstring& ks_name, http_context& ctx, const std::unordered_map<sstring, sstring>& query_params, sstring param_name);
struct table_info {
sstring name;
table_id id;
};
// splits a request parameter assumed to hold a comma-separated list of table names
// verify that the tables are found, otherwise a bad_param_exception exception is thrown
// containing the description of the respective no_such_column_family error.
// Returns a vector of all table infos given by the parameter, or
// if the parameter is not found or is empty, returns a list of all table infos in the keyspace.
std::vector<table_info> parse_table_infos(const sstring& ks_name, http_context& ctx, const std::unordered_map<sstring, sstring>& query_params, sstring param_name);
void set_storage_service(http_context& ctx, httpd::routes& r, sharded<service::storage_service>& ss, gms::gossiper& g, sharded<cdc::generation_service>& cdc_gs, sharded<db::system_keyspace>& sys_ls);
void set_sstables_loader(http_context& ctx, httpd::routes& r, sharded<sstables_loader>& sst_loader);
void unset_sstables_loader(http_context& ctx, httpd::routes& r);
void set_view_builder(http_context& ctx, httpd::routes& r, sharded<db::view::view_builder>& vb);
void unset_view_builder(http_context& ctx, httpd::routes& r);
void set_repair(http_context& ctx, httpd::routes& r, sharded<repair_service>& repair);
void unset_repair(http_context& ctx, httpd::routes& r);
void set_transport_controller(http_context& ctx, httpd::routes& r, cql_transport::controller& ctl);
void unset_transport_controller(http_context& ctx, httpd::routes& r);
void set_rpc_controller(http_context& ctx, httpd::routes& r, thrift_controller& ctl);
void unset_rpc_controller(http_context& ctx, httpd::routes& r);
void set_snapshot(http_context& ctx, httpd::routes& r, sharded<db::snapshot_ctl>& snap_ctl);
void unset_snapshot(http_context& ctx, httpd::routes& r);
void set_storage_service(http_context& ctx, routes& r, sharded<service::storage_service>& ss, gms::gossiper& g, sharded<cdc::generation_service>& cdc_gs);
void set_sstables_loader(http_context& ctx, routes& r, sharded<sstables_loader>& sst_loader);
void unset_sstables_loader(http_context& ctx, routes& r);
void set_view_builder(http_context& ctx, routes& r, sharded<db::view::view_builder>& vb);
void unset_view_builder(http_context& ctx, routes& r);
void set_repair(http_context& ctx, routes& r, sharded<repair_service>& repair);
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);
seastar::future<json::json_return_type> run_toppartitions_query(db::toppartitions_query& q, http_context &ctx, bool legacy_request = false);
} // namespace api
namespace std {
std::ostream& operator<<(std::ostream& os, const api::table_info& ti);
} // namespace std
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "stream_manager.hh"
@@ -14,7 +27,6 @@
#include "gms/gossiper.hh"
namespace api {
using namespace seastar::httpd;
namespace hs = httpd::stream_manager_json;
@@ -22,7 +34,7 @@ static void set_summaries(const std::vector<streaming::stream_summary>& from,
json::json_list<hs::stream_summary>& to) {
if (!from.empty()) {
hs::stream_summary res;
res.cf_id = fmt::to_string(from.front().cf_id);
res.cf_id = boost::lexical_cast<std::string>(from.front().cf_id);
// For each stream_session, we pretend we are sending/receiving one
// file, to make it compatible with nodetool.
res.files = 1;
@@ -39,7 +51,7 @@ static hs::progress_info get_progress_info(const streaming::progress_info& info)
res.current_bytes = info.current_bytes;
res.direction = info.dir;
res.file_name = info.file_name;
res.peer = fmt::to_string(info.peer);
res.peer = boost::lexical_cast<std::string>(info.peer);
res.session_index = 0;
res.total_bytes = info.total_bytes;
return res;
@@ -62,7 +74,7 @@ static hs::stream_state get_state(
state.plan_id = result_future.plan_id.to_sstring();
for (auto info : result_future.get_coordinator().get()->get_all_session_info()) {
hs::stream_info si;
si.peer = fmt::to_string(info.peer);
si.peer = boost::lexical_cast<std::string>(info.peer);
si.session_index = 0;
si.state = info.state;
si.connecting = si.peer;
@@ -75,13 +87,13 @@ static hs::stream_state get_state(
return state;
}
void set_stream_manager(http_context& ctx, routes& r, sharded<streaming::stream_manager>& sm) {
void set_stream_manager(http_context& ctx, routes& r) {
hs::get_current_streams.set(r,
[&sm] (std::unique_ptr<request> req) {
return sm.invoke_on_all([] (auto& sm) {
[] (std::unique_ptr<request> req) {
return streaming::get_stream_manager().invoke_on_all([] (auto& sm) {
return sm.update_all_progress_info();
}).then([&sm] {
return sm.map_reduce0([](streaming::stream_manager& stream) {
}).then([] {
return streaming::get_stream_manager().map_reduce0([](streaming::stream_manager& stream) {
std::vector<hs::stream_state> res;
for (auto i : stream.get_initiated_streams()) {
res.push_back(get_state(*i.second.get()));
@@ -97,17 +109,17 @@ void set_stream_manager(http_context& ctx, routes& r, sharded<streaming::stream_
});
});
hs::get_all_active_streams_outbound.set(r, [&sm](std::unique_ptr<request> req) {
return sm.map_reduce0([](streaming::stream_manager& stream) {
hs::get_all_active_streams_outbound.set(r, [](std::unique_ptr<request> req) {
return streaming::get_stream_manager().map_reduce0([](streaming::stream_manager& stream) {
return stream.get_initiated_streams().size();
}, 0, std::plus<int64_t>()).then([](int64_t res) {
return make_ready_future<json::json_return_type>(res);
});
});
hs::get_total_incoming_bytes.set(r, [&sm](std::unique_ptr<request> req) {
hs::get_total_incoming_bytes.set(r, [](std::unique_ptr<request> req) {
gms::inet_address peer(req->param["peer"]);
return sm.map_reduce0([peer](streaming::stream_manager& sm) {
return streaming::get_stream_manager().map_reduce0([peer](streaming::stream_manager& sm) {
return sm.get_progress_on_all_shards(peer).then([] (auto sbytes) {
return sbytes.bytes_received;
});
@@ -116,8 +128,8 @@ void set_stream_manager(http_context& ctx, routes& r, sharded<streaming::stream_
});
});
hs::get_all_total_incoming_bytes.set(r, [&sm](std::unique_ptr<request> req) {
return sm.map_reduce0([](streaming::stream_manager& sm) {
hs::get_all_total_incoming_bytes.set(r, [](std::unique_ptr<request> req) {
return streaming::get_stream_manager().map_reduce0([](streaming::stream_manager& sm) {
return sm.get_progress_on_all_shards().then([] (auto sbytes) {
return sbytes.bytes_received;
});
@@ -126,9 +138,9 @@ void set_stream_manager(http_context& ctx, routes& r, sharded<streaming::stream_
});
});
hs::get_total_outgoing_bytes.set(r, [&sm](std::unique_ptr<request> req) {
hs::get_total_outgoing_bytes.set(r, [](std::unique_ptr<request> req) {
gms::inet_address peer(req->param["peer"]);
return sm.map_reduce0([peer] (streaming::stream_manager& sm) {
return streaming::get_stream_manager().map_reduce0([peer] (streaming::stream_manager& sm) {
return sm.get_progress_on_all_shards(peer).then([] (auto sbytes) {
return sbytes.bytes_sent;
});
@@ -137,8 +149,8 @@ void set_stream_manager(http_context& ctx, routes& r, sharded<streaming::stream_
});
});
hs::get_all_total_outgoing_bytes.set(r, [&sm](std::unique_ptr<request> req) {
return sm.map_reduce0([](streaming::stream_manager& sm) {
hs::get_all_total_outgoing_bytes.set(r, [](std::unique_ptr<request> req) {
return streaming::get_stream_manager().map_reduce0([](streaming::stream_manager& sm) {
return sm.get_progress_on_all_shards().then([] (auto sbytes) {
return sbytes.bytes_sent;
});
@@ -148,13 +160,4 @@ void set_stream_manager(http_context& ctx, routes& r, sharded<streaming::stream_
});
}
void unset_stream_manager(http_context& ctx, routes& r) {
hs::get_current_streams.unset(r);
hs::get_all_active_streams_outbound.unset(r);
hs::get_total_incoming_bytes.unset(r);
hs::get_all_total_incoming_bytes.unset(r);
hs::get_total_outgoing_bytes.unset(r);
hs::get_all_total_outgoing_bytes.unset(r);
}
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -12,7 +25,6 @@
namespace api {
void set_stream_manager(http_context& ctx, httpd::routes& r, sharded<streaming::stream_manager>& sm);
void unset_stream_manager(http_context& ctx, httpd::routes& r);
void set_stream_manager(http_context& ctx, routes& r);
}

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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/system.json.hh"
@@ -12,12 +25,11 @@
#include <seastar/core/reactor.hh>
#include <seastar/http/exception.hh>
#include "log.hh"
#include "replica/database.hh"
#include "database.hh"
extern logging::logger apilog;
namespace api {
using namespace seastar::httpd;
namespace hs = httpd::system_json;
@@ -62,19 +74,9 @@ void set_system(http_context& ctx, routes& r) {
return json::json_void();
});
hs::write_log_message.set(r, [](const_req req) {
try {
logging::log_level level = boost::lexical_cast<logging::log_level>(std::string(req.get_query_param("level")));
apilog.log(level, "/system/log: {}", std::string(req.get_query_param("message")));
} catch (boost::bad_lexical_cast& e) {
throw bad_param_exception("Unknown logging level " + req.get_query_param("level"));
}
return json::json_void();
});
hs::drop_sstable_caches.set(r, [&ctx](std::unique_ptr<request> req) {
apilog.info("Dropping sstable caches");
return ctx.db.invoke_on_all([] (replica::database& db) {
return ctx.db.invoke_on_all([] (database& db) {
return db.drop_caches();
}).then([] {
apilog.info("Caches dropped");

View File

@@ -3,7 +3,20 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
@@ -12,6 +25,6 @@
namespace api {
void set_system(http_context& ctx, httpd::routes& r);
void set_system(http_context& ctx, routes& r);
}

View File

@@ -1,236 +0,0 @@
/*
* Copyright (C) 2022-present ScyllaDB
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
*/
#include <seastar/core/coroutine.hh>
#include "task_manager.hh"
#include "api/api-doc/task_manager.json.hh"
#include "db/system_keyspace.hh"
#include "column_family.hh"
#include "unimplemented.hh"
#include "storage_service.hh"
#include <utility>
#include <boost/range/adaptors.hpp>
namespace api {
namespace tm = httpd::task_manager_json;
using namespace json;
using namespace seastar::httpd;
inline bool filter_tasks(tasks::task_manager::task_ptr task, std::unordered_map<sstring, sstring>& query_params) {
return (!query_params.contains("keyspace") || query_params["keyspace"] == task->get_status().keyspace) &&
(!query_params.contains("table") || query_params["table"] == task->get_status().table);
}
struct full_task_status {
tasks::task_manager::task::status task_status;
std::string type;
tasks::task_manager::task::progress progress;
std::string module;
tasks::task_id parent_id;
tasks::is_abortable abortable;
std::vector<std::string> children_ids;
};
struct task_stats {
task_stats(tasks::task_manager::task_ptr task)
: task_id(task->id().to_sstring())
, state(task->get_status().state)
, type(task->type())
, keyspace(task->get_status().keyspace)
, table(task->get_status().table)
, entity(task->get_status().entity)
, sequence_number(task->get_status().sequence_number)
{ }
sstring task_id;
tasks::task_manager::task_state state;
std::string type;
std::string keyspace;
std::string table;
std::string entity;
uint64_t sequence_number;
};
tm::task_status make_status(full_task_status status) {
auto start_time = db_clock::to_time_t(status.task_status.start_time);
auto end_time = db_clock::to_time_t(status.task_status.end_time);
::tm st, et;
::gmtime_r(&end_time, &et);
::gmtime_r(&start_time, &st);
tm::task_status res{};
res.id = status.task_status.id.to_sstring();
res.type = status.type;
res.state = status.task_status.state;
res.is_abortable = bool(status.abortable);
res.start_time = st;
res.end_time = et;
res.error = status.task_status.error;
res.parent_id = status.parent_id.to_sstring();
res.sequence_number = status.task_status.sequence_number;
res.shard = status.task_status.shard;
res.keyspace = status.task_status.keyspace;
res.table = status.task_status.table;
res.entity = status.task_status.entity;
res.progress_units = status.task_status.progress_units;
res.progress_total = status.progress.total;
res.progress_completed = status.progress.completed;
res.children_ids = std::move(status.children_ids);
return res;
}
future<full_task_status> retrieve_status(const tasks::task_manager::foreign_task_ptr& task) {
if (task.get() == nullptr) {
co_return coroutine::return_exception(httpd::bad_param_exception("Task not found"));
}
auto progress = co_await task->get_progress();
full_task_status s;
s.task_status = task->get_status();
s.type = task->type();
s.parent_id = task->get_parent_id();
s.abortable = task->is_abortable();
s.module = task->get_module_name();
s.progress.completed = progress.completed;
s.progress.total = progress.total;
std::vector<std::string> ct{task->get_children().size()};
boost::transform(task->get_children(), ct.begin(), [] (const auto& child) {
return child->id().to_sstring();
});
s.children_ids = std::move(ct);
co_return s;
}
void set_task_manager(http_context& ctx, routes& r, db::config& cfg) {
tm::get_modules.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
std::vector<std::string> v = boost::copy_range<std::vector<std::string>>(ctx.tm.local().get_modules() | boost::adaptors::map_keys);
co_return v;
});
tm::get_tasks.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
using chunked_stats = utils::chunked_vector<task_stats>;
auto internal = tasks::is_internal{req_param<bool>(*req, "internal", false)};
std::vector<chunked_stats> res = co_await ctx.tm.map([&req, internal] (tasks::task_manager& tm) {
chunked_stats local_res;
auto module = tm.find_module(req->param["module"]);
const auto& filtered_tasks = module->get_tasks() | boost::adaptors::filtered([&params = req->query_parameters, internal] (const auto& task) {
return (internal || !task.second->is_internal()) && filter_tasks(task.second, params);
});
for (auto& [task_id, task] : filtered_tasks) {
local_res.push_back(task_stats{task});
}
return local_res;
});
std::function<future<>(output_stream<char>&&)> f = [r = std::move(res)] (output_stream<char>&& os) -> future<> {
auto s = std::move(os);
auto res = std::move(r);
co_await s.write("[");
std::string delim = "";
for (auto& v: res) {
for (auto& stats: v) {
co_await s.write(std::exchange(delim, ", "));
tm::task_stats ts;
ts = stats;
co_await formatter::write(s, ts);
}
}
co_await s.write("]");
co_await s.close();
};
co_return std::move(f);
});
tm::get_task_status.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
auto id = tasks::task_id{utils::UUID{req->param["task_id"]}};
auto task = co_await tasks::task_manager::invoke_on_task(ctx.tm, id, std::function([] (tasks::task_manager::task_ptr task) -> future<tasks::task_manager::foreign_task_ptr> {
auto state = task->get_status().state;
if (state == tasks::task_manager::task_state::done || state == tasks::task_manager::task_state::failed) {
task->unregister_task();
}
co_return std::move(task);
}));
auto s = co_await retrieve_status(task);
co_return make_status(s);
});
tm::abort_task.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
auto id = tasks::task_id{utils::UUID{req->param["task_id"]}};
co_await tasks::task_manager::invoke_on_task(ctx.tm, id, [] (tasks::task_manager::task_ptr task) -> future<> {
if (!task->is_abortable()) {
co_await coroutine::return_exception(std::runtime_error("Requested task cannot be aborted"));
}
co_await task->abort();
});
co_return json_void();
});
tm::wait_task.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
auto id = tasks::task_id{utils::UUID{req->param["task_id"]}};
auto task = co_await tasks::task_manager::invoke_on_task(ctx.tm, id, std::function([] (tasks::task_manager::task_ptr task) {
return task->done().then_wrapped([task] (auto f) {
task->unregister_task();
f.get();
return make_foreign(task);
});
}));
auto s = co_await retrieve_status(task);
co_return make_status(s);
});
tm::get_task_status_recursively.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
auto& _ctx = ctx;
auto id = tasks::task_id{utils::UUID{req->param["task_id"]}};
std::queue<tasks::task_manager::foreign_task_ptr> q;
utils::chunked_vector<full_task_status> res;
// Get requested task.
auto task = co_await tasks::task_manager::invoke_on_task(_ctx.tm, id, std::function([] (tasks::task_manager::task_ptr task) -> future<tasks::task_manager::foreign_task_ptr> {
auto state = task->get_status().state;
if (state == tasks::task_manager::task_state::done || state == tasks::task_manager::task_state::failed) {
task->unregister_task();
}
co_return task;
}));
// Push children's statuses in BFS order.
q.push(co_await task.copy()); // Task cannot be moved since we need it to be alive during whole loop execution.
while (!q.empty()) {
auto& current = q.front();
res.push_back(co_await retrieve_status(current));
for (size_t i = 0; i < current->get_children().size(); ++i) {
q.push(co_await current->get_children()[i].copy());
}
q.pop();
}
std::function<future<>(output_stream<char>&&)> f = [r = std::move(res)] (output_stream<char>&& os) -> future<> {
auto s = std::move(os);
auto res = std::move(r);
co_await s.write("[");
std::string delim = "";
for (auto& status: res) {
co_await s.write(std::exchange(delim, ", "));
co_await formatter::write(s, make_status(status));
}
co_await s.write("]");
co_await s.close();
};
co_return f;
});
tm::get_and_update_ttl.set(r, [&cfg] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
uint32_t ttl = cfg.task_ttl_seconds();
co_await cfg.task_ttl_seconds.set_value_on_all_shards(req->query_parameters["ttl"], utils::config_file::config_source::API);
co_return json::json_return_type(ttl);
});
}
}

View File

@@ -1,18 +0,0 @@
/*
* Copyright (C) 2022-present ScyllaDB
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
*/
#pragma once
#include "api.hh"
#include "db/config.hh"
namespace api {
void set_task_manager(http_context& ctx, httpd::routes& r, db::config& cfg);
}

View File

@@ -1,102 +0,0 @@
/*
* Copyright (C) 2022-present ScyllaDB
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
*/
#ifndef SCYLLA_BUILD_MODE_RELEASE
#include <seastar/core/coroutine.hh>
#include "task_manager_test.hh"
#include "api/api-doc/task_manager_test.json.hh"
#include "tasks/test_module.hh"
namespace api {
namespace tmt = httpd::task_manager_test_json;
using namespace json;
using namespace seastar::httpd;
void set_task_manager_test(http_context& ctx, routes& r) {
tmt::register_test_module.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
co_await ctx.tm.invoke_on_all([] (tasks::task_manager& tm) {
auto m = make_shared<tasks::test_module>(tm);
tm.register_module("test", m);
});
co_return json_void();
});
tmt::unregister_test_module.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
co_await ctx.tm.invoke_on_all([] (tasks::task_manager& tm) -> future<> {
auto module_name = "test";
auto module = tm.find_module(module_name);
co_await module->stop();
});
co_return json_void();
});
tmt::register_test_task.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
sharded<tasks::task_manager>& tms = ctx.tm;
auto it = req->query_parameters.find("task_id");
auto id = it != req->query_parameters.end() ? tasks::task_id{utils::UUID{it->second}} : tasks::task_id::create_null_id();
it = req->query_parameters.find("shard");
unsigned shard = it != req->query_parameters.end() ? boost::lexical_cast<unsigned>(it->second) : 0;
it = req->query_parameters.find("keyspace");
std::string keyspace = it != req->query_parameters.end() ? it->second : "";
it = req->query_parameters.find("table");
std::string table = it != req->query_parameters.end() ? it->second : "";
it = req->query_parameters.find("entity");
std::string entity = it != req->query_parameters.end() ? it->second : "";
it = req->query_parameters.find("parent_id");
tasks::task_info data;
if (it != req->query_parameters.end()) {
data.id = tasks::task_id{utils::UUID{it->second}};
auto parent_ptr = co_await tasks::task_manager::lookup_task_on_all_shards(ctx.tm, data.id);
data.shard = parent_ptr->get_status().shard;
}
auto module = tms.local().find_module("test");
id = co_await module->make_task<tasks::test_task_impl>(shard, id, keyspace, table, entity, data);
co_await tms.invoke_on(shard, [id] (tasks::task_manager& tm) {
auto it = tm.get_all_tasks().find(id);
if (it != tm.get_all_tasks().end()) {
it->second->start();
}
});
co_return id.to_sstring();
});
tmt::unregister_test_task.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
auto id = tasks::task_id{utils::UUID{req->query_parameters["task_id"]}};
co_await tasks::task_manager::invoke_on_task(ctx.tm, id, [] (tasks::task_manager::task_ptr task) -> future<> {
tasks::test_task test_task{task};
co_await test_task.unregister_task();
});
co_return json_void();
});
tmt::finish_test_task.set(r, [&ctx] (std::unique_ptr<http::request> req) -> future<json::json_return_type> {
auto id = tasks::task_id{utils::UUID{req->param["task_id"]}};
auto it = req->query_parameters.find("error");
bool fail = it != req->query_parameters.end();
std::string error = fail ? it->second : "";
co_await tasks::task_manager::invoke_on_task(ctx.tm, id, [fail, error = std::move(error)] (tasks::task_manager::task_ptr task) {
tasks::test_task test_task{task};
if (fail) {
test_task.finish_failed(std::make_exception_ptr(std::runtime_error(error)));
} else {
test_task.finish();
}
return make_ready_future<>();
});
co_return json_void();
});
}
}
#endif

View File

@@ -1,21 +0,0 @@
/*
* Copyright (C) 2022-present ScyllaDB
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
*/
#ifndef SCYLLA_BUILD_MODE_RELEASE
#pragma once
#include "api.hh"
namespace api {
void set_task_manager_test(http_context& ctx, httpd::routes& r);
}
#endif

View File

@@ -3,13 +3,26 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "atomic_cell.hh"
#include "atomic_cell_or_collection.hh"
#include "counters.hh"
#include "types/types.hh"
#include "types.hh"
atomic_cell atomic_cell::make_dead(api::timestamp_type timestamp, gc_clock::time_point deletion_time) {
return atomic_cell_type::make_dead(timestamp, deletion_time);
@@ -66,48 +79,36 @@ atomic_cell::atomic_cell(const abstract_type& type, atomic_cell_view other)
set_view(_data);
}
// Based on Cassandra's resolveRegular function:
// - https://github.com/apache/cassandra/blob/e4f31b73c21b04966269c5ac2d3bd2562e5f6c63/src/java/org/apache/cassandra/db/rows/Cells.java#L79-L119
//
// Note: the ordering algorithm for cell is the same as for rows,
// except that the cell value is used to break a tie in case all other attributes are equal.
// See compare_row_marker_for_merge.
// Based on:
// - org.apache.cassandra.db.AbstractCell#reconcile()
// - org.apache.cassandra.db.BufferExpiringCell#reconcile()
// - org.apache.cassandra.db.BufferDeletedCell#reconcile()
std::strong_ordering
compare_atomic_cell_for_merge(atomic_cell_view left, atomic_cell_view right) {
// Largest write timestamp wins.
if (left.timestamp() != right.timestamp()) {
return left.timestamp() <=> right.timestamp();
}
// Tombstones always win reconciliation with live cells of the same timestamp
if (left.is_live() != right.is_live()) {
return left.is_live() ? std::strong_ordering::less : std::strong_ordering::greater;
}
if (left.is_live()) {
// Prefer expiring cells (which will become tombstones at some future date) over live cells.
// See https://issues.apache.org/jira/browse/CASSANDRA-14592
auto c = compare_unsigned(left.value(), right.value()) <=> 0;
if (c != 0) {
return c;
}
if (left.is_live_and_has_ttl() != right.is_live_and_has_ttl()) {
// prefer expiring cells.
return left.is_live_and_has_ttl() ? std::strong_ordering::greater : std::strong_ordering::less;
}
// If both are expiring, choose the cell with the latest expiry or derived write time.
if (left.is_live_and_has_ttl()) {
// Prefer cell with latest expiry
if (left.expiry() != right.expiry()) {
return left.expiry() <=> right.expiry();
} else if (right.ttl() != left.ttl()) {
// The cell write time is derived by (expiry - ttl).
// Prefer the cell that was written later,
// so it survives longer after it expires, until purged,
// as it become purgeable gc_grace_seconds after it was written.
//
// Note that this is an extension to Cassandra's algorithm
// which stops at the expiration time, and if equal,
// move forward to compare the cell values.
} else {
// prefer the cell that was written later,
// so it survives longer after it expires, until purged.
return right.ttl() <=> left.ttl();
}
}
// The cell with the largest value wins, if all other attributes of the cells are identical.
// This is quite arbitrary, but still required to break the tie in a deterministic way.
return compare_unsigned(left.value(), right.value());
} else {
// Both are deleted
@@ -206,13 +207,14 @@ operator<<(std::ostream& os, const atomic_cell_view::printer& acvp) {
std::ostringstream cell_value_string_builder;
if (type.is_counter()) {
if (acv.is_counter_update()) {
fmt::print(cell_value_string_builder, "counter_update_value={}", acv.counter_update_value());
cell_value_string_builder << "counter_update_value=" << acv.counter_update_value();
} else {
cell_value_string_builder << "shards: ";
auto ccv = counter_cell_view(acv);
fmt::print(cell_value_string_builder, "shards: {}", fmt::join(ccv.shards(), ", "));
cell_value_string_builder << ::join(", ", ccv.shards());
}
} else {
fmt::print(cell_value_string_builder, "{}", type.to_string(to_bytes(acv.value())));
cell_value_string_builder << type.to_string(to_bytes(acv.value()));
}
fmt::print(os, "atomic_cell{{{},ts={:d},expiry={:d},ttl={:d}}}",
cell_value_string_builder.str(),

View File

@@ -3,14 +3,27 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "bytes.hh"
#include "timestamp.hh"
#include "mutation/tombstone.hh"
#include "tombstone.hh"
#include "gc_clock.hh"
#include "utils/managed_bytes.hh"
#include <seastar/net//byteorder.hh>

View File

@@ -3,18 +3,31 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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
// Not part of atomic_cell.hh to avoid cyclic dependency between types.hh and atomic_cell.hh
#include "types/types.hh"
#include "types.hh"
#include "types/collection.hh"
#include "atomic_cell.hh"
#include "atomic_cell_or_collection.hh"
#include "utils/hashing.hh"
#include "hashing.hh"
#include "counters.hh"
template<>

View File

@@ -3,14 +3,27 @@
*/
/*
* SPDX-License-Identifier: AGPL-3.0-or-later
* 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 "atomic_cell.hh"
#include "collection_mutation.hh"
#include "schema/schema.hh"
#include "schema.hh"
// A variant type that can hold either an atomic_cell, or a serialized collection.
// Which type is stored is determined by the schema.

View File

@@ -1,35 +0,0 @@
include(add_whole_archive)
add_library(scylla_auth STATIC)
target_sources(scylla_auth
PRIVATE
allow_all_authenticator.cc
allow_all_authorizer.cc
authenticated_user.cc
authenticator.cc
common.cc
default_authorizer.cc
password_authenticator.cc
passwords.cc
permission.cc
permissions_cache.cc
resource.cc
role_or_anonymous.cc
roles-metadata.cc
sasl_challenge.cc
service.cc
standard_role_manager.cc
transitional.cc)
target_include_directories(scylla_auth
PUBLIC
${CMAKE_SOURCE_DIR})
target_link_libraries(scylla_auth
PUBLIC
Seastar::seastar
xxHash::xxhash
PRIVATE
cql3
idl
wasmtime_bindings)
add_whole_archive(auth scylla_auth)

Some files were not shown because too many files have changed in this diff Show More