Fixes#25683
Once a table drop is complete, there should be no reason to retain
truncation records for it, as any replay should skip mutations
anyway (no CF), and iff we somehow resurrect a dropped table,
this replay-resurrected data is the least problem anyway.
Adds a prune phase to the startup drop_truncation_rp_records run,
which ignores updating, and instead deletes records for non-existant
tables (which should patch any existing servers with lingering data
as well).
Also does an explicit delete of records on actual table DROP, to
ensure we don't grow this table more than needed even in long
uptime nodes.
Small unit test included.
Closesscylladb/scylladb#25699
(cherry picked from commit bc20861afb)
Closesscylladb/scylladb#25811
When the configuration option `rf_rack_valid_keyspaces` is enabled and there
is an RF-rack-invalid keyspace, starting a node fails. However, when the
configuration option is disabled, but there still is a keyspace that violates
the condition, we'd like Scylla to print a warning informing the user about
the fact. That's what happens in this commit.
We provide a validation test.
(cherry picked from commit 837d267cbf)
TRUNCATE TABLE performs a memtable flush and then discards the sstables
of the table being truncated. It collects the highest replay position
for both of these. When the highest replay position of the discarded
sstables is higher than the highest replay position of the flushed
memtable, that means that we have had writes during truncate which have
been flushed to disk independently of the truncate process. We check for
this and trigger an on_internal_error() which throws an exception,
informing the user that writing data concurrently with TRUNCATE TABLE is
not advised.
The problem with this is that truncate is also called from DROP KEYSPACE
and DROP TABLE. These are raft operations and exceptions thrown by them
are caught by the (...) exception handler in the raft applier fiber,
which then exits leaving the node without the ability to execute
subsequent raft commands.
This commit changes the on_internal_error() into a warning log entry. It
also outputs to keyspace/table names, the truncated_at timepoint, the
offending replay positions which caused the check to fail.
Fixes: #25173Fixes: #25013
(cherry picked from commit 268ec72dc9)
Prepare for next patch, the will use this shared_token_metadata
to make mutable_token_metadata_ptr:s
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
(cherry picked from commit 493a2303da)
Truncate doesn't really go well with concurrent writes. The fix (#23560) exposed
a preexisting fragility which I missed.
1) truncate gets RP mark X, truncated_at = second T
2) new sstable written during snapshot or later, also at second T (difference of MS)
3) discard_sstables() get RP Y > saved RP X, since creation time of sstable
with RP Y is equal to truncated_at = second T.
So the problem is that truncate is using a clock of second granularity for
filtering out sstables written later, and after we got low mark and truncate time,
it can happen that a sstable is flushed later within the same second, but at a
different millisecond.
By switching to a millisecond clock (db_clock), we allow sstables written later
within the same second from being filtered out. It's not perfect but
extremely unlikely a new write lands and get flushed in the same
millisecond we recorded truncated_at timepoint. In practice, truncate
will not be used concurrently to writes, so this should be enough for
our tests performing such concurrent actions.
We're moving away from gc_clock which is our cheap lowres_clock, but
time is only retrieved when creating sstable objects, which frequency of
creation is low enough for not having significant consequences, and also
db_clock should be cheap enough since it's usually syscall-less.
Fixes#23771.
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Closesscylladb/scylladb#24426
(cherry picked from commit 2d716f3ffe)
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Closesscylladb/scylladb#24875
Similar to how large_data_handler is handled, propagate through
sstables::sstables_manager and store its owner: replica::database.
Tests and tools are also patched. Mostly mechanical changes, updating
constructors and patching callers.
(cherry picked from commit ebd9420687)
In the previous commits we made sure that the base info is not dependent
on the base schema version, and the info dependent on the base schema
version is calculated when it's needed. In this patch we remove the
unnecessary re-setting of the base_info.
The set_base_info method isn't removed completely, because it also has
a secondary function - zeroing the view_info fields other than base_info.
Because of this, in this patch we rename it accordingly and limit its
use to the updates caused by a base schema change.
(cherry picked from commit d7bd86591e)
Each view update is correlated to a write that generates it (aside from view
building which is throttled separately). These writes are limited by a throttling
mechanism, which effectively works by performing the writes with CL=ALL if
ongoing writes exceed some memory usage limit
When writes generate view updates, they usually also need to perform a read. This read
goes through a read concurrency semaphore where it can get delayed or killed. The
semaphore allows up to 100 concurrent reads and puts all remaining reads in a queue.
If the number of queued reads exceeds a specific limit, the view update will fail on
the replica, causing inconsistencies.
This limit is not necessary. When a read gets queued on the semaphore, the write that's
causing the view update is paused, so the write takes part in the regular write throttling.
If too many writes get stuck on view update reads, they will get throttled, so their
number is limited and the number of queued reads is also limited to the same amount.
In this patch we remove the specified queue length limit for the view update read concurrency
semaphore. Instead of this limit, the queue will be now limited indirectly, by the base write
throttling mechanism. This may allow the queue grow longer than with the previous limit, but
it shouldn't ever cause issues - we only perform up to 100 actual reads at once, and the
remaining ones that get queued use a tiny amount of memory, less than the writes that generated
them and which are getting limited directly.
Fixes https://github.com/scylladb/scylladb/issues/23319Closesscylladb/scylladb#24112
(cherry picked from commit 5920647617)
Closesscylladb/scylladb#24168
When running those operations after a tablet replica is migrated away from
a shard, an assert can fail resulting in a crash.
Status quo (around the assert in truncate procedure):
1) Highest RP seen by table is saved in low_mark, and the current time in
low_mark_at.
2) Then compaction is disabled in order to not mix data written before truncate,
and data written later.
3) Then memtable is flushed in order for the data written before truncate to be
available in sstables and then removed.
4) Now, current time is saved in truncated_at, which is supposedly the time of
truncate to decide which sstables to remove.
Note: truncated_at is likely above low_mark_at due to steps 2 and 3.
The interesting part of the assert is:
(truncated_at <= low_mark_at ? rp <= low_mark : low_mark <= rp)
Note: RP in the assert above is the highest RP among all sstables generated
before truncated_at. RP is retrieved by table::discard_sstables().
If truncated_at > low_mark_at, maybe newer data was written during steps 2 and
3, and memtable's RP becomes greater than low_mark, resulting in a SSTable with
RP > low_mark.
So assert's 2nd condition is there to defend against the scenario above.
truncated_at and low_mark_at uses millisecond granularity, so even if
truncated_at == low_mark_at, data could have been written in steps 2 and 3
(during same MS window), failing the assert. This is fragile.
Reproducer:
To reproduce the problem, truncated_at must be > low_mark_at, which can easily
happen with both drop table and truncate due to steps 2 and 3.
If a shard has 2 or more tablets, the table's highest RP refer to just one
tablet in that shard.
If the tablet with the highest RP is migrated away, then the sstables in that
shard will have lower RP than the recorded highest RP (it's a table wide state,
which makes sense since CL is shared among tablets).
So when either drop table or truncate runs, low_mark will be potentially bigger
than highest RP retrieved from sstables.
Proposed solution:
The current assert is hacked to not fail if writes sneak in, during steps 2 and
3, but it's still fragile and seems not to serve its real purpose, since it's
allowing for RP > low_mark.
We should be able to say that low_mark >= RP, as a way of asserting we're not
leaving data targeted by truncate behind (or that we're not removing the wrong
data).
But the problem is that we're saving low_mark in step 1, before preparation
steps (2 and 3). When truncated_at is recorded in step 4, it's a way of saying
all data written so far is targeted for removal. But as of today, low_mark
refers to all data written up to step 1. So low_mark is now only one set
before issuing flush, and also accounts for all potentially flushed data.
Fixes#18059.
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Closesscylladb/scylladb#23560
(cherry picked from commit 0f59deffaa)
(cherry picked from commit 7554d4bbe09967f9b7a55575b5dfdde4f6616862)
Signed-off-by: Raphael S. Carvalho <raphaelsc@scylladb.com>
Closesscylladb/scylladb#23649
The row cache can garbage-collect tombstones in two places:
1) When populating the cache - the underlying reader pipeline has a `compacting_reader` in it;
2) During reads - reads now compact data including garbage collection;
In both cases, garbage collection has to do overlap checks against memtables, to avoid collecting tombstones which cover data in the memtables.
This PR includes fixes for (2), which were not handled at all currently.
(1) was already supposed to be fixed, see https://github.com/scylladb/scylladb/issues/20916. But the test added in this PR showed that the test is incomplete: https://github.com/scylladb/scylladb/issues/23291. A fix for this issue is also included.
Fixes: https://github.com/scylladb/scylladb/issues/23291
Fixes: https://github.com/scylladb/scylladb/issues/23252
The fix will need backport to all live release.
- (cherry picked from commit c2518cdf1a)
- (cherry picked from commit 6b5b563ef7)
- (cherry picked from commit 7e600a0747)
- (cherry picked from commit d126ea09ba)
- (cherry picked from commit cb76cafb60)
- (cherry picked from commit df09b3f970)
- (cherry picked from commit e5afd9b5fb)
- (cherry picked from commit 34b18d7ef4)
- (cherry picked from commit f7938e3f8b)
- (cherry picked from commit 6c1f6427b3)
- (cherry picked from commit 0d39091df2)
Parent PR: #23255Closesscylladb/scylladb#23673
* github.com:scylladb/scylladb:
test/boost/row_cache_test: add memtable overlap check tests
replica/table: add error injection to memtable post-flush phase
utils/error_injection: add a way to set parameters from error injection points
test/cluster: add test_data_resurrection_in_memtable.py
test/pylib/utils: wait_for_cql_and_get_hosts(): sort hosts
replica/mutation_dump: don't assume cells are live
replica/database: do_apply() add error injection point
replica: improve memtable overlap checks for the cache
replica/memtable: add is_merging_to_cache()
db/row_cache: add overlap-check for cache tombstone garbage collection
mutation/mutation_compactor: copy key passed-in to consume_new_partition()
So writes (to user tables) can be failed on a replica, via error
injection. Should simplify tests which want to create differences in
what writes different replicas receive.
(cherry picked from commit cb76cafb60)
The current memtable overlap check that is used by the cache
-- table::get_max_purgeable_fn_for_cache_underlying_reader() -- only
checks the active memtable, so memtables which are either being flushed
or are already flushed and also have active reads against them do not
participate in the overlap check.
This can result in temporary data resurrection, where a cache read can
garbage-collect a tombstone which still covers data in a flushing or
flushed memtable, which still have active read against it.
To prevent this, extend the overlap check to also consider all of the
memtable list. Furthermore, memtable_list::erase() now places the removed
(flushed) memtable in an intrusive list. These entries are alive only as
long as there are readers still keeping an `lw_shared_ptr<memtable>`
alive. This list is now also consulted on overlap checks.
(cherry picked from commit d126ea09ba)
Normally, when a node is shutting down, `gate_closed_exception` and `rpc::closed_error`
in `send_to_live_endpoints` should be ignored. However, if these exceptions are wrapped
in a `nested_exception`, an error message is printed, causing tests to fail.
This commit adds handling for nested exceptions in this case to prevent unnecessary
error messages.
Fixes scylladb/scylladb#23325
Fixes scylladb/scylladb#23305
Fixes scylladb/scylladb#21815
Backport: looks like this is quite a frequent issue, therefore backport to 2025.1.
- (cherry picked from commit 6abfed9817)
- (cherry picked from commit b1e89246d4)
- (cherry picked from commit 0d9d0fe60e)
- (cherry picked from commit d448f3de77)
Parent PR: #23336Closesscylladb/scylladb#23470
* github.com:scylladb/scylladb:
database: Pass schema_ptr as const ref in `wrap_commitlog_add_error`
database: Unify exception handling in `do_apply` and `apply_with_commitlog`
storage_proxy: Ignore wrapped `gate_closed_exception` and `rpc::closed_error` when node shuts down.
exceptions: Add `try_catch_nested` to universally handle nested exceptions of the same type.
The lambda which dumps the diagnostics for each semaphore, is static.
Considering that said lambda captures a local (writeln) by reference, this
is wrong on two levels:
* The writeln captured on the shard which happens to initialize this
static, will be used on all shards.
* The writeln captured on the first dump, will be used on later dumps,
possibly triggering a segfault.
Drop the `static` to make the lambda local and resolve this problem.
Fixes: scylladb/scylladb#22756Closesscylladb/scylladb#22776
(cherry picked from commit 820f196a49)
Closesscylladb/scylladb#22938
Move exception wrapping logic from `do_apply` and `apply_with_commitlog`
to `wrap_commitlog_add_error` to ensure consistent error handling.
(cherry picked from commit 0d9d0fe60e)
When a node is started with the option `rf_rack_valid_keyspaces`
enabled, the initialization will fail if there is an RF-rack-invalid
keyspace. We want to force the user to adjust their existing
keyspaces when upgrading to 2025.* so that the invariant that
every keyspace is RF-rack-valid is always satisfied.
Fixesscylladb/scylladb#23300
(cherry picked from commit 0e04a6f3eb)
When a replica get a write request it performs get_schema_for_write,
which waits until the schema is synced. However, database::add_column_family
marks a schema as synced before the table is added. Hence, the write may
see the schema as synced, but hit no_such_column_family as the table
hasn't been added yet.
Mark schema as synced after the table is added to database::_tables_metadata.
Fixes: #22347.
Closesscylladb/scylladb#22348
(cherry picked from commit 328818a50f)
Closesscylladb/scylladb#22604
`tablet_storage_group_manager::all_storage_groups_split()` calls `set_split_mode()` for each of its storage groups to create split ready compaction groups. It does this by iterating through storage groups using `std::ranges::all_of()` which is not guaranteed to iterate through the entire range, and will stop iterating on the first occurrence of the predicate (`set_split_mode()`) returning false. `set_split_mode()` creates the split compaction groups and returns false if the storage group's main compaction group or merging groups are not empty. This means that in cases where the tablet storage group manager has non-empty storage groups, we could have a situation where split compaction groups are not created for all storage groups.
The missing split compaction groups are later created in `tablet_storage_group_manager::split_all_storage_groups()` which also calls `set_split_mode()`, and that is the reason why split completes successfully. The problem is that
`tablet_storage_group_manager::all_storage_groups_split()` runs under a group0 guard, but
`tablet_storage_group_manager::split_all_storage_groups()` does not. This can cause problems with operations which should exclude with compaction group creation. i.e. DROP TABLE/DROP KEYSPACE
Fixes#22431
This is a bugfix and should be back ported to versions with tablets: 6.1 6.2 and 2025.1
- (cherry picked from commit 24e8d2a55c)
- (cherry picked from commit 8bff7786a8)
Parent PR: #22330Closesscylladb/scylladb#22560
* github.com:scylladb/scylladb:
test: add reproducer and test for fix to split ready CG creation
table: run set_split_mode() on all storage groups during all_storage_groups_split()
with_permit() creates a permit, with a self-reference, to avoid
attaching a continuation to the permit's run function. This
self-reference is used to keep the permit alive, until the execution
loop processes it. This self reference has to be carefully cleared on
error-paths, otherwise the permit will become a zombie, effectively
leaking memory.
Instead of trying to handle all loose ends, get rid of this
self-reference altogether: ask caller to provide a place to save the
permit, where it will survive until the end of the call. This makes the
call-site a little bit less nice, but it gets rid of a whole class of
possible bugs.
Fixes: #22588Closesscylladb/scylladb#22624
(cherry picked from commit f2d5819645)
Closesscylladb/scylladb#22704
This adds a reproducer for #22431
In cases where a tablet storage group manager had more than one storage
group, it was possible to create compaction groups outside the group0
guard, which could create problems with operations which should exclude
with compaction group creation.
(cherry picked from commit 8bff7786a8)
Currently, when we load a frozen schema into the registry, we lose
the base info if the schema was of a view. Because of that, in various
places we need to set the base info again, and in some codepaths we
may miss it completely, which may make us unable to process some
requests (for example, when executing reverse queries on views).
Even after setting the base info, we may still lose it if the schema
entry gets deactivated due to all `schema_ptr`s temporarily dying.
To fix this, this patch adds the base schema to the registry, alongside
the view schema. We store just the frozen base schema, so that we can
transfer it across shards. With the base schema, we can now set the base
info when returning the schema from the registry. As a result, we can now
assume that all view schemas returned by the registry have base_info set.
In this series we also make sure that the view schemas in the registry are
kept up-to-date in regards to base schema changes.
Fixes https://github.com/scylladb/scylladb/issues/21354
This issue is a bug, so adding backport labels 6.1 and 6.2
Closesscylladb/scylladb#21862
* github.com:scylladb/scylladb:
test: add test for schema registry maintaining base info for views
schema_registry: avoid setting base info when getting the schema from registry
schema_registry: update cached base schemas when updating a view
schema_registry: cache base schemas for views
db: set base info before adding schema to registry
Replace the reader concurrency semaphores for user reads and view
updates with the newly introduced reader concurrency semaphore group,
which assigns a semaphore for each service level.
Each group is statically assigned to some pool of memory on startup and
dynamically distribute this memory between the semaphores, relative to
the number of shares of the corresponding scheduling group.
The intent of having a separate reader concurrency semaphore for each
scheduling group is to prevent priority inversion issues due to reads
with different priorities waiting on the same semaphore, as well as make
memory allocation more fair between service levels due to the adjusted
number of shares.
In the following patches, we'll assure that view schemas returned by the
schema registry always have base info set. To prepare for that, make sure
that the base info is always set before inserting it into schema registry,
TRUNCATE TABLE saves the current commit log replay positions in case there is a crash so that replay knows where to begin replaying the mutations. These are collected and saved per shard into `system.truncated`. In case a shard received no mutations, its replay position will be an empty, default constructed object of type `db::replay_position` with its members set to 0. Truncate will incorrectly interpret these empty replay positions as if they were coming from shard 0, and save them as such, potentially overwriting an actual valid replay position coming from the actual shard 0. In the case of a crash, this will cause the commit log on shard 0 to be replayed from the beginning, and result with data resurrection.
Fixes#21719Closesscylladb/scylladb#21722
* github.com:scylladb/scylladb:
test: add test for truncate saving replay positions
database: correctly save replay position for truncate
since fedora 38 is EOL. and fedora 39 comes with fmt v10.0.0, also,
we've switched to the build image based on fedora 40, which ships
fmt-devel v10.2.1, there is no need to support fmt < 10.
in this change, we drop the support fmt < 10.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#21847
This commit fixes a problem with way truncate saves commit log
replay positions. On shards without mutations, truncate would save the
replay position into system.truncated with shard number 0 regardless of
the actual shard number that the replay position was saved for.
Once e.g. `ALTER KEYSPACE` is performed, all in-memory objects should be updated accordingly, but this is not entirely true for keyspace metadata object. The reason for that is that keyspace metadata are stored in 2 system tables: `system_schema.keyspaces` and `system_schema.scylla_keyspaces`. Up until now the in-memory keyspace metadata object has been updated only with entries from the first table, and missed updates when entries from the 2nd table changed. These entries were e.g. initial tablets or storage options.
This change fixes this oversight by considering both tables when checking if keyspace metadata need to be updated. From the implementation point of view, the change is simple: we're considering `system_schema.scylla_keyspaces` also in `merge_keyspaces()` and if old and new schemas have any differences, we include that when altering ks.
Fixes#20768
Backport: no need, I don't think the issue is severe, atm it seems like it can only influence the tablets number, which should not bring the cluster down nor result in returning bad data, it can mostly influence the speed of the db.
Closesscylladb/scylladb#20852
`coroutine::parallel_for_each` accepts both a range and a pair of
iterators. let's use the former when appropriate. it is simpler this way.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#21684
Stop taking snapshots of MVs and allow taking snapshot of individual tables, now one can take a snapshot of any base table, any view or index. Also add tests to cover new cases both boost test (using cc code) and pytest (using the API)
Also, update documentation to reflect the change
fixes: #21339fixes: #20760Closesscylladb/scylladb#21433
Modernize the codebase by replacing Boost range adaptors with C++23 standard library views,
reducing external dependencies and leveraging modern C++ language features.
Key Changes:
- Replace `boost::adaptors::filtered` with `std::views::filter`
- Remove `#include <boost/range/adaptor/filtered.hpp>`
- Utilize standard library range views
Motivation:
- Reduce project's external dependency footprint
- Leverage standard library's range and view capabilities
- Improve long-term code maintainability
- Align with modern C++ best practices
Implementation Challenges and Considerations:
1. Range Conversion and Move Semantics
- `std::ranges::to` adaptor requires rvalue references
- Necessitated updates to variable and parameter constness
- Example: `cql3/restrictions/statement_restrictions.cc` modified to remove `const`
from `common` to enable efficient range conversion
2. Range Iteration and Mutation
- Range views may mutate internal state during iteration
- Cannot pass ranges by const reference in some scenarios
- Solution: Pass ranges by rvalue reference to explicitly indicate
state invalidation
Limitations:
- One instance of `boost::adaptors::filtered` temporarily preserved
due to lack of a C++23 alternative for `boost::join()`
- A comprehensive replacement will be addressed in a follow-up change
This change is part of our ongoing effort to modernize the codebase,
reducing external dependencies and adopting modern C++ practices.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#21648
now that we are allowed to use C++23. we now have the luxury of using
`std::ranges::find_if`.
in this change, we:
- replace `boost::find_if` with `std::ranges::find_if`
- remove all `#include <boost/range/algorithm/find_if.hpp>`
to reduce the dependency to boost for better maintainability, and
leverage standard library features for better long-term support.
this change is part of our ongoing effort to modernize our codebase
and reduce external dependencies where possible.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Add a buffer hint to the multishard reader. This is an internal hint, used by the multishard reader to provide a hint to the shard reader, on how much data exactly is needed by the multishard reader from the respective shard. This hint allows eliminating extraneous cross-shard round-trips and possible shard reader evict-recreate cycles. Building on this, repair sets its own row buffer size as the max buffer size on the multishard reader, ensuring that the row buffer is filled with the minimum amount of cross-shard round trips and minimal reader recreation.
To further eliminate unnecessary evictions, this PR also disables the multishard reader's read-ahead which is a mechanism that was designed to reduce latency for user-reads but it can be too aggressive for repair, causing unnecessary extra congestion on the already struggling streaming semaphores.
Refs: https://github.com/scylladb/scylladb/issues/18269
Fixes: https://github.com/scylladb/scylladb/issues/21113
The performance impact was measured with an SCT test, which creates a cluster of 3 nodes with 16 shards, then adds a 4th one with 12 shards.
Currently, it is the bootstrap time which is the worse in the case of mixed shard clusters, see below for the improvement measured during bootstrap:
| | master | buffer-hint | metric |
| ------------ | ------------- | ------------- | --------------------------------------------------- |
| evictions | 0.9M | 93.0K | scylla_database_paused_reads_permit_based_evictions |
| read (bytes) | 9.0T | 3.9T | scylla_reactor_aio_bytes_read |
| read (ops) | 88.0M | 33.5M | scylla_reactor_aio_reads |
| time | 56min | 20min | N/A |
This is a performance improvement, no backport required.
Closesscylladb/scylladb#20815
* github.com:scylladb/scylladb:
test/boost/mutation_reader_test: add test for multishard reader buffer hint
repair/row_level: disable read-ahead
db/config: introduce repair_multishard_reader_enable_read_ahead
readers/multishard: implement the read_ahead flag
replica/database: make_multishard_streaming_reader(): expose the read_ahead parameter
readers/multishard: add read_ahead parameter
repair/row_level: set max buffer size on multishard reader
replica/database: make_multishard_streaming_reader(): expose buffer_hint parameter
db/config: introduce enable_repair_multishard_reader_buffer_hint
readers/multishard: multishard_reader: pass hint to shard_reader
readers/multishard: shard_reader_v2::fill_reader_buffer(): respect the hint
readers/multishard: propagate fill_buffer_hint to shard_reader:fill_reader_buffer()
readers/multishard: shard_reader: extract buffer-fill into its own method
Continuing the previous patch, expose the just added read_ahead
parameter of make_multishard_combining>_reader_v2().
Set to read_ahead::yes by all callers, keeping the current default.
Expose the buffer hint functionality added by the previous commits, to
callers of make_multishard_streaming_reader(). All callers disable it
currently, it will be used in the next patch.
now that we are allowed to use C++23. we now have the luxury of using
`std::ranges::any_of`.
in this change, we replace `boost::algorithm::any_of` with
`std::ranges::any_of`
to reduce the dependency to boost for better maintainability, and
leverage standard library features for better long-term support.
this change is part of our ongoing effort to modernize our codebase
and reduce external dependencies where possible.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
now that we are allowed to use C++23. we now have the luxury of using
`std::views::values`.
in this change, we:
- replace `boost::adaptors::map_values` with `std::views::values`
- update affected code to work with `std::views::values`
- the places where we use `boost::join()` are not changed, because
we cannot use `std::views::concat` yet. this helper is only
available in C++26.
to reduce the dependency to boost for better maintainability, and
leverage standard library features for better long-term support.
this change is part of our ongoing effort to modernize our codebase
and reduce external dependencies where possible.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Closesscylladb/scylladb#21265
When writing to some tables with materialized views, we need to read from the base table first to perform a delete of the old view row. When doing so, the memory used for the read is tracked by the user read concurrency semaphore. When we have a large number of such reads, we may use up all of the semaphore units, causing the following reads to be queued. When we have some user reads coming at the same time, these reads can have very high latency due to the write workload on the base table. We want to avoid this, so that the write workload doesn't have a high impact on the latency of the read workload.
This is fixed in this patch by adding a separate read concurrency semaphore just for view update read-before-writes. With the new semaphore, even if there are many view update read-before-writes, they will be queued on a different semaphore than the user reads, and they won't impact their latency.
The second issue fixed by this patch is the concurrency of the view updates that is currently unlimited. Because of that view updates may take up so much memory that they we may run out of memory.
This is fixed by using the read admission on the view update concurrency semaphore.
This limits the number of concurrent view update reads to
max_count_concurrent_view_update_reads, all other incoming view update reads are
queued using just a small chunk of memory. Without this, the reads would also get
queued after exceeding view_update_reader_concurrency_semaphore_serialize_limit_multiplier, but they would take much more memory while staying in the queue.
The new semaphore has half the capacity of the regular user read concurrency semahpore and is currently used only for user writes - is't used independently of the scheduling group on which we base the read semaphore selection, but we use a different code path for streaming (not database::do_apply) and we shouldn't have view updates in system writes or during compaction.
This patch also adds a test to confirm that the view update workload doesn't impact the read latency, as well as a test which confirms that we do not run out of memory even under heavy view udpate workload.
The issue of view updates causing increased latencies most often occurs in the following scenario:
* we have a medium to high write workload to a table with a materialized view which requires reading from the base table before sending the update to delete the old rows
* we have any read workload
* one replica is slower or is handling more writes due to an imbalance of data distribution
* we write with a cl<ALL, the mentioned replica is replying to write requests slower while new ones keep being sent to it.
* each write performs a read first taking resources from the user read concurrency semaphore, so when enough writes accumulate the reads using the semaphore start getting queued
* the queue is shared by regular reads and view update reads. When there's enough view update reads in the queue, regular reads start getting increased latencies
An sct test (perf-regression-latency-mv-read-concurrency) was prepared to somewhat resemble this scenario:
* the tables were prepared satisfying the conditions above
* we use a medium write workload and a very low read workload
* the imbalance is achieved by writing to just a few (10) partitions - some replicas (and shards) can have twice or more used partitions than others. We also keep writing to a limited (though high) number of rows, to cause overwrites which require reading before sending the view update
* to minimize the test case, we use a cluster of 3 nodes and rf=2, we write with cl=ONE to have background replica writes and read with cl=ALL to wait for the slower replica to respond.
In the test above:
* without the fix, the latency of reads increases over 50s
* with the fix, the latency of reads stays below 20ms
Fixes https://github.com/scylladb/scylladb/issues/8873
Fixes https://github.com/scylladb/scylladb/issues/15805
The patch is not that small and it isn't fixing a regression, so no backports
Closesscylladb/scylladb#20887
* github.com:scylladb/scylladb:
test: add test for high view update concurrency causing bad_allocs
test: add test for high view update concurrency degrading read latency
mv: add a dedicated read concurrency semaphore for view update read before writes
The all_datadirs keeps paths to directories where local sstables can be. In fact, Scylla doesn't put sstables there, but can try to find them on boot and when checking snapshots. The 0th element of this vector, called datadir, had recently been removed by #20675, now it's time to drop all_datadirs as well. The needed paths can be obtained from table's storage options (see #20542) and db::config::data_file_directories option.
Closesscylladb/scylladb#21212
* github.com:scylladb/scylladb:
sstables: Open-code format_table_directory_name() moved recently
replica,sstables: Move format_table_directory_name()
table: Remove all_datadirs
sstables: Generate table::all_datadirs from db::config and storage_options
replica: Prepare vector of fs::path-s with table dirs
table: Check storage options in get_snapshot_details()
the log.hh under the root of the tree was created keep the backward
compatibility when seastar was extracted into a separate library.
so log.hh should belong to `utils` directory, as it is based solely
on seastar, and can be used all subsystems.
in this change, we move log.hh into utils/log.hh to that it is more
modularized. and this also improves the readability, when one see
`#include "utils/log.hh"`, it is obvious that this source file
needs the logging system, instead of its own log facility -- please
note, we do have two other `log.hh` in the tree.
Signed-off-by: Kefu Chai <kefu.chai@scylladb.com>
Now this helper is not needed in replica code, as all manipulations of
tables' sstables now sit in the sstables/storage.cc.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
It's write-only now, all the places than wanted to know where table's
storage is (well -- "are", there can be several directories) already use
storage_options.
This finishes the work started by 9fe64b5d70.
Signed-off-by: Pavel Emelyanov <xemul@scylladb.com>
When writing to some tables with materialized views, we need to read from the base
table first to perform a delete of the old view row. When doing so, the memory used
for the read is tracked by the user read concurrency semaphore. When we have a large
number of such reads, we may use up all of the semaphore units, causing the following
reads to be queued. When we have some user reads coming at the same time, these reads
can have very high latency due to the write workload on the base table. We want to avoid
this, so that the write workload doesn't have a high impact on the latency of the
read workload.
This is fixed in this patch by adding a separate read concurrency semaphore just for
view update read-before-writes. With the new semaphore, even if there are many view
update read-before-writes, they will be queued on a different semaphore than the user
reads, and they won't impact their latency.
The second issue fixed by this patch is the concurrency of the view updates that is
currently unlimited. Because of that view updates may take up so much memory that
they we may run out of memory.
This is fixed by using the read admission on the view update concurrency semaphore.
This limits the number of concurrent view update reads to
max_count_concurrent_view_update_reads, all other incoming view update reads are
queued using just a small chunk of memory. Without this, the reads would also get
queued after exceeding view_update_reader_concurrency_semaphore_serialize_limit_multiplier,
but they would take much more memory while staying in the queue.
The new semaphore has half the capacity of the regular user read concurrency semahpore
and is currently used only for user writes - is't used independently of the scheduling
group on which we base the read semaphore selection, but we use a different code path
for streaming (not database::do_apply) and we shouldn't have view updates in system
writes or during compaction.
Fixes https://github.com/scylladb/scylladb/issues/8873
Fixes https://github.com/scylladb/scylladb/issues/15805
Move the repair_map definition to the tombstone_gc file where it is
mostly being used.
Refactor and add the accessors and setters for the group0 tombstone GC
time.
