`system.client_routes` is a system table that sets the target address and ports for each `host_id`, for one or more connection (e.g., Private Link) represented by `connection_id`. Cloud will write the table via REST, and drivers will read it via CQL to override values obtained from `system.local` and `system.peers`.
This patch series contains:
- Introduction of `CLIENT_ROUTES` feature flag.
- Implementation of raft-based `system.client_routes` table
- Implementation of `v2/client-routes` POST/DELETE/GET endpoints
- Implementation of new `CLIENT_ROUTES_CHANGE` event that is sent to drivers when `system.client_routes` is changed
- New tests that verifies the aforementioned features
Ref: scylladb/scylla-enterprise#5699
For now, no automatic backport. However, the changes are planned to be release on `2025.4` either as a backport or a private build.
Closesscylladb/scylladb#27323
* https://github.com/scylladb/scylladb:
docs: describe CLIENT_ROUTES_CHANGE extension
test: add test for CLIENT_ROUTES event
service: transport: add CLIENT_ROUTES_CHANGE event
test: add cluster tests for client routes
test: add API tests for client_routes endpoints
test: add `timeout` parameter to `delete` in RESTClient
test: allow json_body in send
api: implement client_routes endpoints
api: add client_routes.json
service: main: add client_routes_service
db: add system.client_routes table
gms: add CLIENT_ROUTES feature
Fixes#26744
If a segment to replay is broken such that the main header is not zero, but still broken, we throw header_checksum_error. This was not handled in replayer, which grouped this into the "user error/fundamental problem" category.
However, assuming we allow for "real" disk corruption, this should really be treated same as data corruption, i.e. reported data loss, not failure to start up.
The `test_one_big_mutation_corrupted_on_startup` test accidentally sometimes provoked this issue, by doing random file wrecking, which on rare occasions provoked this, and thus failed test due to scylla not starting up, instead of losing data as expected.
Closesscylladb/scylladb#27556
* github.com:scylladb/scylladb:
test::cluster::dtest::tools::files: Remove file
commitlog_replay: Handle fully corrupt files same as partial corruption.
test::pylib::suite::base: Split options.name test specifier only once
Fix multiple cases where the captured `std::exception_ptr` has been re-thrown via simple `throw eptr;`, which results in losing the original exception type and details.
Resolved at various places found by clang-tidy:
1. db::schema_applier
When applying schema changes, the previous implementation attempted to handle exceptions by catching and rethrowing them, but did so incorrectly: using `throw ex` with a `std::exception_ptr` loses the original exception type and details.
However, in this case, explicit exception handling is unnecessary. The only reason for catching was to ensure `ap.destroy()` is called before propagating the exception. This can be more cleanly and safely achieved using Seastar's `.finally()` continuation, which guarantees cleanup regardless of success or failure.
2. directories
The `std::exception_ptr()` has been captured for logging and then again re-thrown incorrectly via `throw ex;`. We could use `std::rethrow_exception()` here instead, but it seems to be simpler to just use regular `throw;` to rethrow the original exception, and only use the `std::current_exception()` for logging (which is a pattern used in other places as well).
3. storage_service
Here the exception has been re-thrown incorrectly in a coroutine. There it is best to use the `co_await coroutine::return_exception_ptr` to propagate exception more efficiently in a coroutine-friendly manner.
Fixes: SCYLLADB-94
Refs: scylladb/scylladb#27501
No backport: This fixes an error logging issue, that isn't a production problem by itself (only found in test), therefore not backporting to older branches.
Closesscylladb/scylladb#27613
* https://github.com/scylladb/scylladb:
db: schema_applier: improve exception-safe cleanup
directories: fix exception rethrowing
storage_service: use coroutine-friendly exception propagation in join_node_response_handler
Introduce `system.client_routes`, a system table that sets the target
address and ports for each `host_id`, for one or more connections
(e.g., Private Link) represented by `connection_id`. Cloud will write
the table via REST, and drivers will read it via CQL to override
values obtained from `system.local` and `system.peers`.
The table is Raft-managed to provide consistent replication across
nodes.
Schema overview: each row is identified by `(connection_id, host_id)`
and describes where clients should connect: `address` and one or more of
`port`, `tls_port`, `alternator_port`, `alternator_https_port`.
`host_id` is a UUID (just as in ScyllaDB) but `connection_id` can be
any string to accept formats of all cloud providers. `address` is
also a regular string because it can represent either an IP address or
a domain. Ports are optional in the sense that at least one of
the four must be provided.
Ref: scylladb/scylla-enterprise#5699
The batchlog table contains an entry for each logged batch that is processed by the local node as coordinator. These entries are typically very short lived, they are inserted when the batch is processed and deleted immediately after the batch is successfully applied.
When a table has `tombstone_gc = {'mode': 'repair'}` enabled, every repair has to flush all hints and batchlogs, so that we can be certain that there is no live data in any of these, older than the last repair. Since batches can contain member queries from any number of tables, the whole batchlog has to be flushed, even if repair-mode tombstone-gc is enabled for a single table.
Flushing the batchlog table happens by doing a batchlog replay. This involves reading the entire content of this table, and attempting to replay+delete any live entries (that are old enough to be replayed). Under normal operating circumstances, 99%+ of the content of the batchlog table is partition tombstones. Because of this, scanning the content of this table has to process thousands to millions of tombstones. This was observed to require up to 20 minutes to finish, causing repairs to slow down to a crawl, as the batchlog-flush has to be repeated at the end of the repair of each token-range.
When trying to address this problem, the first idea was that we should expedite the garbage-collection of these accumulated tombstones. This experiment failed, see https://github.com/scylladb/scylladb/pull/23752. The commitlog proved to be an impossible to bypass barrier, preventing quick garbage-collection of tombstones. So long as a single commit-log segment is alive, holding content from the batchlog table, all tombstones written after are blocked from GC.
The second approach, represented by this PR, is to not rely in tombstone GC to reduce the tombstone amount. Instead restructure the table such that a single higher-order tombstone can be used to shadow and allow for the eviction of the myriads of individual batchlog entry tombstones. This is realized by reorganizing the batchlog table such that individual batches are rows, not partitions.
This new schema is introduced by the new `system.batchlog_v2` table, introduced by this PR:
CREATE TABLE system.batchlog_v2 (
version int,
stage int,
shard int,
written_at timestamp,
id uuid,
data blob,
PRIMARY KEY ((version, stage, shard), written_at, id));
The new schema organization has the following goals:
1) Make post-replay batchlog cleanup possible with a simple range-tombstone. This allows dropping the individual dead batchlog entries, as they are shadowed by a higher level tombstone. This enables dropping tombstones without tombstone GC.
2) To make the above possible, introduce the stage key component: batchlog entries that fail the first replay attempt, are moved to the failed_replay stage, so the initial stage can be cleaned up safely.
3) Spread out the data among Scylla shards, via the batchlog shard column.
4) Make batchlog entries ordered by the batchlog create time (id). This allows for selecting batchlogs to replay, without post-filtering of batchlogs that are too young to be replayed.
Fixes: https://github.com/scylladb/scylladb/issues/23358
This is an improvement, normally not a backport-candidate. We might override this and backport to allow wider use of `tombstone_gc: {'mode': 'repair'}`.
Closesscylladb/scylladb#26671
* github.com:scylladb/scylladb:
db/config: change batchlog_replay_cleanup_after_replays default to 1
test/boost/batchlog_manager_test: add test for batchlog cleanup
replica/mutation_dump: always set position weight for clustering positions
service/storage_proxy: s/batch_replay_throw/storage_proxy_fail_replay_batch/
test/lib: introduce error_injection.hh
utils/error_injection: add debug log to disable() and disable_all()
test/lib/cql_test_env: forward config to batchlog
test/lib/cql_test_env: add batch type to execute_batch()
test/lib/cql_assertions: add with_size(predicate) overload
test/lib/cql_assertions: add source location to fail messages
test/lib/cql_assertions: columns_assertions: add assert_for_columns_of_each_row()
test/lib/cql_assertions: rows_assertions::assert_for_columns_of_row(): add index bound check
test/lib/cql_assertions: columns_assertions: add T* with_typed_column() overload
db/batchlog_manager: config: s/write_timeout/reply_timeot/
db,service: switch to system.batchlog_v2
db/system_keyspace: introduce system.batchlog_v2
service,db: extract generation of batchlog delete mutation
service,db: extract get_batchlog_mutation_for() from storage-proxy
db/batchlog_manager: only consider propagation delay with tombstone-gc=repair
db/batchlog_manager: don't drop entire batch if one mutations' table was dropped
data_dictionary: table: add get_truncation_time()
db/batchlog_manager: batch(): replace map_reduce() with simple loop
db/batchlog_manager: finish coroutinizing replay_all_failed_batches
db/batchlog_manager: improve replayAllFailedBatches logs
When applying schema changes, the previous implementation attempted to
handle exceptions by catching and rethrowing them, but did so
incorrectly: using `throw ex` with a `std::exception_ptr` loses the
original exception type and details. The correct approach is to use
`std::rethrow_exception()`.
However, in this case, explicit exception handling is unnecessary. The
only reason for catching was to ensure `ap.destroy()` is called before
propagating the exception. This can be more cleanly and safely achieved
using Seastar's `.finally()` continuation, which guarantees cleanup
regardless of success or failure.
This change removes the manual try/catch/rethrow and uses `.finally()`
to ensure proper cleanup, letting exceptions propagate naturally and
preserving their type and information.
Fixes: SCYLLADB-94
Refs: scylladb/scylladb#27501
This reverts commit 8192f45e84.
The merge exposed a bug where truncate (via drop) fails and causes Raft
errors, leading to schema inconsistencies across nodes. This results in
test_table_drop_with_auto_snapshot failures with 'Keyspace test does not exist'
errors.
The specific problematic change was in commit 19b6207f which modified
truncate_table_on_all_shards to set use_sstable_identifier = true. This
causes exceptions during truncate that are not properly handled, leading
to Raft applier fiber stopping and nodes losing schema synchronization.
This reverts commit faad0167d7. It causes
a regression in
test_two_tablets_concurrent_repair_and_migration_repair_writer_level
in debug mode (with ~5%-10% probability).
Fixes#27510.
Closesscylladb/scylladb#27560
Fixes#26744
If a segment to replay is broken such that the main header is not zero,
but still broken, we throw header_checksum_error. This was not handled in
replayer, which grouped this into the "user error/fundamental problem"
category. However, assuming we allow for "real" disk corruption, this should
really be treated same as data corruption, i.e. reported data loss, not
failure to start up.
The `test_one_big_mutation_corrupted_on_startup` test accidentally sometimes
provoked this issue, by doing random file wrecking, which on rare occasions
provoked this, and thus failed test due to scylla not starting up, instead
of loosing data as expected.
Changed test to consistently cause this exact error instead.
With the introduction of rack-lists and the reliance of materialized views on them, the `get_view_natural_endpoint` function can be greatly simplified. When using tablets, instead of doing any index-matching, we can now pair base tables with views only in the same rack.
In this series we remove no longer needed code and reorganize the needed code for better clarity.
After the changes, the `get_view_natural_endpoint` function goes down from 245 lines to 85 lines, while the whole pairing-related text goes down from 346 lines to 239 lines.
Fixes https://github.com/scylladb/scylladb/issues/26313Closesscylladb/scylladb#27383
* github.com:scylladb/scylladb:
mv: replace the simple/complex rack-aware pairing with exact rack matching
mv: split out vnode pairing code from get_view_natural_endpoint
mv: unify self-pairing and rack-aware pairing into one bool
mv: remove the workaround for left nodes when sending view updates
This patch adds tablet repair progress report support so that the user
could use the /task_manager/task_status API to query the progress.
In order to support this, a new system table is introduced to record the
user request related info, i.e, start of the request and end of the
request.
The progress is accurate when tablet split or merge happens in the
middle of the request, since the tokens of the tablet are recorded when
the request is started and when repair of each tablet is finished. The
original tablet repair is considered as finished when the finished
ranges cover the original tablet token ranges.
After this patch, the /task_manager/task_status API will report correct
progress_total and progress_completed.
Fixes#22564Fixes#26896Closesscylladb/scylladb#26924
This change adds a new option to the REST api and correspondingly, to scylla nodetool: use_sstable_identifier.
When set, we use the sstable identifier, if available, to name each sstable in the snapshots directory
and the manifest.json file, rather than using the sstable generation.
This can be used by the user (e.g. Scylla Manager) for global deduplication with tablets, where an sstable
may be migrated across shards or across nodes, and in this case, its generation may change, but its
sstable identifier remains sstable.
Currently, Scylla manager uses the sstable generation to detect sstables that are already backed up to
object storage and exist in previous backed up snapshots.
Historically, the sstable generation was guaranteed to be unique only per table per node,
so the dedup code currently checks for deduplication in the node scope.
However, with tablet migration, sstables are renamed when migrated to a different shard,
i.e. their generation changes, and they may be renamed when migrated to another node,
but even if they are not, the dedup logic still assumes uniqueness only within a node.
To address both cases, we keep the sstable_id stable throughout the sstable life cycle (since 3a12ad96c7).
Given the globally unique sstable identifier, scylla manager can now detect duplicate sstables
in a wider scope. This can be cluster-wide, but we practically need only rack-wide deduplication
or dc-wide, as tablets are migrated across racks only in rare occasions (like when converting from a
numerical replication factor to a rack list containing a subset of the available racks in a datacenter).
Fixes#27181
* New feature, no backport required
Closesscylladb/scylladb#27184
* github.com:scylladb/scylladb:
database: truncate_table_on_all_shards: set use_sstable_identifier to true
nodetool: snapshot: add --use-sstable-identifier option
api: storage_service: take_snapshot: add use_sstable_identifier option
test: database_test: add snapshot_use_sstable_identifier_works
test: database_test: snapshot_works: add validate_manifest
sstable: write_scylla_metadata: add random_sstable_identifier error injection
table: snapshot_on_all_shards: take snapshot_options
sstable: add get_format getter
sstable: snapshot: add use_sstable_identifier option
db: snapshot_ctl: snapshot_options: add use_sstable_identifier options
db: snapshot_ctl: move skip_flush to struct snapshot_options
Large reserves in allocating_section can cause stalls. We already log
reserve increase, but we don't know which table it belongs to:
lsa - LSA allocation failure, increasing reserve in section 0x600009f94590 to 128 segments;
Allocating sections used for updating row cache on memtable flush are
notoriously problematic. Each table has its own row_cache, so its own
allocating_section(s). If we attached table name to those sections, we
could identify which table is causing problems. In some issues we
suspected system.raft, but we can't be sure.
This patch allows naming allocating_sections for the purpose of
identifying them in such log messages. I use abstract_formatter for
this purpose to avoid the cost of formatting strings on the hot path
(e.g. index_reader). And also to avoid duplicating strings which are
already stored elsewhere.
Fixes#25799Closesscylladb/scylladb#27470
Range tombstones are represented as entry attributes, which applies to
the interval between entries. So if a range tombstone covers many
rows, to apply it we have to update all covered entries. In some
workloads that could be many entries, even the whole cache. Before
the patch, we did this update without preemption, which can cause
reactor stalls in such workloads.
This scenario is already covered by mvcc_tests,
e.g. test_apply_to_incomplete_respects_continuity. And I verified that
the new preemption point is hit in the test.
perf-row-cache-update results show no significant stalls anymore (max
2ms scheduling delay, instead of previous 1.5 s):
Generated 1124195 rows
Memtable fill took 4179.457520 [ms], {count: 8295, 99%: 0.654949 [ms], max: 32.817176 [ms]}
Draining...
took 0.000616 [ms]
cache: 2506/2948 [MB], memtable: 781/1024 [MB], alloc/comp: 1051/662 [MB] (amp: 0.630)
update: 2874.157471 [ms], preemption: {count: 26650, 99%: 1.131752 [ms], max: 2.068762 [ms]}, cache: 3027/3973 [MB], alloc/comp: 3951/2424 [MB] (amp: 0.614), pr/me/dr 1124195/0/0
Fixes#23479Fixes#2578Closesscylladb/scylladb#27469
* github.com:scylladb/scylladb:
cache, mvcc: Preempt cache update when applying range tombstone from memtable
partition_snapshot_row_cursor: Clarify non-obvious semantic difference of range_tombstone()
perf-row-cache-update: Add scenario with large tombstone covering many rows
We switched to using v3 schema tables (in system_schema keyspace) in
2017, in 9eb91bc30b.
So no system should have the old schema any more.
No need to run legacy_schema_migrator on boot.
Closesscylladb/scylladb#27420
The PRUNE MATERALIZED VIEW statement is performed as follows:
1. Perform a range scan of the view table from the view replicas based
on the ranges specified in the statement.
2. While reading the paged scan above, for each view row perform a read
from all base replicas at the corresponding primary key. If a discrepancy
is detected, delete the row in the view table.
When reading multiple rows, this is very slow because for each view row
we need to performe a single row query on multiple replicas.
In this patch we add an option to speed this up by performing many of the
single base row reads concurrently, at the concurrency specified in the
USING CONCURRENCY clause.
Aside from the unit test, I checked manually on a 3-node cluster with 10M rows, using vnodes. There were actually no ghost rows in the test, but we still had to iterate over all view rows and read the corresponding base rows. And actual ghost rows, if there are any, should be a tiny fraction of all rows. I compared concurrencies 1,2,10,100 and the results were:
* Pruning with concurrency 1 took total 1416 seconds
* Pruning with concurrency 2 took total 731 seconds
* Pruning with concurrency 10 took total 234 seconds
* Pruning with concurrency 100 took total 171 seconds
So after a concurrency of 10 or so we're hitting diminishing returns (at least in this setup). At that point we may be no longer bottlenecked by the reads, but by CPU on the shard that's handling the PRUNE
Fixes https://github.com/scylladb/scylladb/issues/27070Closesscylladb/scylladb#27097
* github.com:scylladb/scylladb:
mv: allow setting concurrency in PRUNE MATERIALIZED VIEW
cql: add CONCURRENCY to the USING clause
To be used for naming sstables in the snapshot by their
sstable identifiers rather than their generation, to
facilitate global deduplication of sstables in backup.
Signed-off-by: Benny Halevy <bhalevy@scylladb.com>
This patch adds separate group for vector search parameters in the
documentation and fixes small typos and formatting.
Fixes: SCYLLADB-77.
Closesscylladb/scylladb#27385
Even though that `view_building_coordinator::work_on_view_building` has
an `if` at the very beginning which checks whether the currently
processed base table is set, it only prints a message and continues
executing the rest of the function regardless of the result of the
check. However, some of the logic in the function assumes that the
currently processed base table field is set and tries to access the
value of the field. This can lead to the view building coordinator
accessing a disengaged optional, which is undefined behavior.
Fix the function by adding the clearly missing `co_await` to the check.
A regression test is added which checks that the view building state
observer - a different fiber which used to print a weird message due to
erroneus view building coordinator behavior - does not print a warning.
Fixes: scylladb/scylladb#27363Closesscylladb/scylladb#27373
Now that batchlog cleanup is cheap, on account of memtable flush on the
system.batchlog table garbage-collecting tombstones (previous patch), we
can afford to do cleanup on each replay, keeping the memtable size small
and more importantly -- the amount of tombstones in the memtable small.
Although the value of this item is indeed derived from the write timeout
config, the name doesn't reflect what it is used for. Change it to
reflect it better.
New batchlogs are written to the batchlog_v2 table and replay also uses
the v2 table.
The content of system.batchlog is attempted to be migrated to
system.batchlog_v2 after each start of the batchlog_manager service.
The migration is retried on each replay if it fails. This is reduntant
but simple.
Batchlog cleanup now doesn't involve flushing memtables, the only
remaining user of replica/database.hh is gone, so the include is
dropped.
Rearranges the system.batchlog schema as follows:
CREATE TABLE system.batchlog_v2 (
version int,
stage int,
shard int,
written_at timestamp,
id uuid,
data blob,
PRIMARY KEY ((version, stage, shard), written_at, id));
With the following goals:
1) Make post-replay batchlog cleanup possible with a simple
range-tombstone. This allows dropping the individual dead batchlog
entries, as they are shadowed by a higher level tombstone. This
enables dropping tombstones without tombstone GC.
2) To make the above possible, introduce the stage key component:
batchlog entries that fail the first replay attempt, are moved to the
failed_replay stage, so the initial stage can be cleaned up safely.
3) Spread out the data among Scylla shards, via the batchlog shard
column.
4) Make batchlog entries ordered by the batchlog create time (id). This
allows for selecting batchlogs to replay, without post-filtering of
batchlogs that are too young to be replayed.
Don't build batchlog delete mutations in storage-proxy code. Move this
code into db/batchlog_manager.cc, exposed via db/batchlog.hh.
This serves multiple goals:
1) Concentrates low-level batchlog related logic in
db/batchlog_manager.cc
2) Reduce current and future code duplication.
3) Make future changes to this logic easier.
Don't build batchlog mutations in storage-proxy code. Move this code
into db/batchlog_manager.cc, exposed via db/batchlog.hh.
This serves multiple goals:
1) Concentrates low-level batchlog related logic in
db/batchlog_manager.cc
2) Reduce current and future code duplication.
2) Make future changes to this logic easier.
Just skip the mutation(s) whose tables were dropped instead.
Use the newly introduced data_dictionary::table::get_truncation_time()
to avoid looking up real table object.
The map_reduce achieves no concurrency, both map and reduce are
synchronous. It only achieves two redundant lookups for the table and
hard-to-read code. Convert it into a simple loop. Preserve the
stall-protection by adding a maybe_yield() to the loop.
When the initial version of rack-aware pairing was introduced, materialized
views with tablets were still experimental. Since then, we decided that
we'll only allow materialized views in clusters where the base table and
the view are replicated on the same racks, with one replica of each tablet
on each rack.
This allows us to remove almost all logic from our base-view pairing. The
only check for the paired view replica is now whether it's in the same
rack as the base replica sending the update.
In this patch we replace the simple and complex rack-aware pairing with
the simple check above.
Because of this, we have to remove a test case from network_topology_strategy_test
which was testing complex pairing. The tested topology is not supported
for views with tablets (or is unlikely to be supported, as it's a random test),
so there's no use keeping the test.
The test case for simple rack aware pairing was kept, but now we only test
the case where each rack has one replica, not multiple.
Additionally, we split finding of an unpaired replica to a separate function
and partially rewrite it without reusing the helper stuctures that were
present when calculating the simple and complex rack-aware pairing.
We only look for an unpaired replica if we couldn't find a paired replica
ourselves or if the number of view replicas didn't match the base replicas.
If an unpaired replica appears while these conditions pass, we won't send
an extra update, but that would be a new bug altogether, because we only
expect the unpaired replica to appear during RF changes, so when these
conditions aren't fulfilled.
Fixes https://github.com/scylladb/scylladb/issues/26313
Add cleanup flag value to start message and drop cpu, it is redundant as
Scylla already adds the shard number to the logs.
Add all_replayed to finish message.
To avoid repeatedly checking whether we're using tablets and having
to use unnecesarily flexible code fitting both cases, we split out
the base-view pairing code for the case of vnodes to another function.
The get_view_natural_endpoint will now have only common steps,
a call to that function, and steps specific to tablets.
We always use "legacy self pairing" when not using tablets, and
the "rack aware pairing" has been enabled in every version where
views with tablets isn't experimental. So in practice, instead
of checking these variables we can just look at whether the
table uses tablets.
At one point, the get_view_natural_endpoint was using IP for the
view update (and hint) destinations, but the hint code was using
host_id for the destinations. When a node left, we could no longer
have a mapping for a IP to host_id and when trying to store a hint
for this IP, we'd crash.
We worked around this issue by dropping the view update completely
if the target is in the "left" state.
Since then, we also moved to host_id's in the view update code, so
there's no longer any translation needed when storing the hints.
Additionally, we now drain hints not when entering the "left" state,
but when the node actually stops owning tokens.
Because of that, the workaround is not needed anymore, so we remove
it in this commit.
The existing test_mv_tablets_empty_ip case verifies that indeed, we
do not crash in the original problematic scenario.
This patch increases the compatibility with DynamoDB Streams by integrating the DynamoDB's event type rules (described in https://github.com/scylladb/scylladb/issues/6918) into Alternator. The main changes are:
- introduce a new flag `alternator_streams_strict_compatibility`, meant as a guard of performance-intensive operations that increase the compatibility with DynamoDB Streams. If enabled, Alternator always performs a RBW before a data-modifying operation, and propagates its result to CDC. Then, the old item is compared to the new one, to determine the mutation type (INSERT vs MODIFY). This option is a no-op for tables with disabled Alternator Streams,
- reduce splitting of simple Alternator mutations,
- correctly distinguish event types described in #6918, except for item deletes. Deleting a missing item with DeleteItem, BatchWriteItem, or a missing field with UpdateItem still emit REMOVEs.
To summarize, the emitted events of the data manipulation operations should be as follows:
- DeleteItem/BatchWriteItem.DeleteItem of existing item: REMOVE (OK)
- DeleteItem of nonexistent item: nothing (OK)
- BatchWriteItem.DeleteItem of nonexistent item: nothing (OK)
- PutItem/UpdateItem/BatchWriteItem.PutItem of existing and not equal item: MODIFY (OK)
- PutItem/UpdateItem/BatchWriteItem.PutItem of existing and equal item: nothing (OK)
- PutItem/UpdateItem/BatchWriteItem.PutItem of nonexistent item: INSERT (OK)
No backport is necessary.
Refs https://github.com/scylladb/scylladb/pull/26149
Refs https://github.com/scylladb/scylladb/pull/26396
Refs https://github.com/scylladb/scylladb/issues/26382
Fixes https://github.com/scylladb/scylladb/issues/6918
Closes scylladb/scylladb#26121
* github.com:scylladb/scylladb:
test/alternator: Enable the tests failing because of #6918
alternator, cdc: Don't emit events for no-op removes
alternator, cdc: Don't emit an event for equal items
alternator/streams, cdc: Differentiate item replace and item update in CDC
alternator: Change the return type of rmw_operation_return
config: Add alternator_streams_strict_compatibility flag
cdc: Don't split a row marker away from row cells
When we come across a segment truncation, this information may
be helpful to determine when the error occurred exactly and
hint at what code path might've led to it.
Closesscylladb/scylladb#27207
Fixes#24346
When reading, we check for each entry and each chunk, if advancing there
will hit EOF of the segment. However, IFF the last chunk being read has
the last entry _exactly_ matching the chunk size, and the chunk ending
at _exactly_ segment size (preset size, typically 32Mb), we did not check
the position, and instead complained about not being able to read.
This has literally _never_ happened in actual commitlog (that was replayed
at least), but has apparently happened more and more in hints replay.
Fix is simple, just check the file position against size when advancing
said position, i.e. when reading (skipping already does).
v2:
* Added unit test
Closesscylladb/scylladb#27236
The PRUNE MATERALIZED VIEW statement is performed as follows:
1. Perform a range scan of the view table from the view replicas based
on the ranges specified in the statement.
2. While reading the paged scan above, for each view row perform a read
from all base replicas at the corresponding primary key. If a discrepancy
is detected, delete the row in the view table.
When reading multiple rows, this is very slow because for each view row
we need to performe a single row query on multiple replicas.
In this patch we add an option to speed this up by performing many of the
single base row reads concurrently, at the concurrency specified in the
USING CONCURRENCY clause.
Fixes https://github.com/scylladb/scylladb/issues/27070
Previously, the view building coordinator relied on setting each task's state to STARTED and then explicitly removing these state entries once tasks finished, before scheduling new ones. This approach induced a significant number of group0 commits, particularly in large clusters with many nodes and tablets, negatively impacting performance and scalability.
With the update, the coordinator and worker logic has been restructured to operate without maintaining per-task states. Instead, tasks are simply tracked with an aborted boolean flag, which is still essential for certain tablet operations. This change removes much of the coordination complexity, simplifies the view building code, and reduces operational overhead.
In addition, the coordinator now batches reports of finished tasks before making commits. Rather than committing task completions individually, it aggregates them and reports in groups, significantly minimizing the frequency of group0 commits. This new approach is expected to improve efficiency and scalability during materialized view construction, especially in large deployments.
Fixes https://github.com/scylladb/scylladb/issues/26311
This patch needs to be backported to 2025.4.
Closesscylladb/scylladb#26897
* github.com:scylladb/scylladb:
docs/dev/view-building-coordinator: update the docs after recent changes
db/view/view_building: send coordinator's term in the RPC
db/view/view_building_state: replace task's state with `aborted` flag
db/view/view_building_coordinator: batch finished tasks reporting
db/view/view_building_worker: change internal implementation
db/view/view_building_coordinator: change `work_on_tasks` RPC return type
This PR adds support for limiting the maximum shares allocated to a
compaction scheduling class by the compaction controller. It introduces
a new configuration parameter, compaction_max_shares, which, when set
to a non zero value, will cap the shares allocated to compaction jobs.
This PR also exposes the shares computed by the compaction controller
via metrics, for observability purposes.
Fixes https://github.com/scylladb/scylladb/issues/9431
Enhancement. No need to backport.
NOTE: Replaces PR https://github.com/scylladb/scylladb/pull/26696
Ran a test in which the backlog raised the need for max shares (normalized backlog above normalization_factor), and played with different values for new option compaction_max_shares to see it works (500, 1000, 2000, 250, 50)
Closesscylladb/scylladb#27024
* github.com:scylladb/scylladb:
db/config: introduce new config parameter `compaction_max_shares`
compaction_manager:config: introduce max_shares
compaction_controller: add configurable maximum shares
compaction_controller: introduce `set_max_shares()`
To avoid case when an old coordinator (which hasn't been stopped yet)
dictates what should be done, add raft term to the `work_on_view_building_tasks`
RPC.
The worker needs to check if the term matches the current term from raft
server, and deny the request when the term is bad.
After previous commits, we can drop entire task's state and replace it
with single boolean flag, which determines if a task was aborted.
Once a task was aborted, it cannot get resurrected to a normal state.
In previous implementation to execute view building tasks, the
coordinator needed to firstly set their states to `STARTED`
and then it needed to remove them before it could start the next ones.
This logic required a lot of group0 commits, especially in large
clusters with higher number of nodes and big tablet count.
After previous commit to the view building worker, the coordinator
can start view building tasks without setting the `STARTED` state
and deleting finished tasks.
This patch adjusts the coordinator to save finished tasks locally,
so it can continue to execute next ones and the finished tasks are
periodically removed from the group0 by `finished_task_gc_fiber()`.
Add support for the new configuration parameter `compaction_max_shares`,
and update the compaction manager to pass it down to the compaction
controller when it changes. The shares allocated to compaction jobs will
be limited by this new parameter.
Fixes#9431
Signed-off-by: Lakshmi Narayanan Sreethar <lakshmi.sreethar@scylladb.com>
When migrating tablet size during the end_migration tablet transition stage, we need the pending and leaving replica hosts. The leaving and pending replicas are gathered in objects of type std::optional<tablet_replica> and are not checked if they contain a value before dereferencing which could cause an exception in the topology coordinator.
This patch adds a check for leaving and pending replicas, and only performs the tablet size migration if neither are empty.
This bug was introduced in 10f07fb95a
This change also adds the ability to create a tablet size in load_stats during end_migration stage of a tablet rebuild. We compute the new tablet size from by averaging the tablet sizes of the existing replicas.
This change also adds the virtual table tablet_sizes which contains tablet sizes of all the replicas of all the tablets in the cluster.
A version containing this bug has not yet been released, so a backport is not needed.
Closesscylladb/scylladb#27118
* github.com:scylladb/scylladb:
test: add tests for tablet size migration during end_migration
virtual_table: add tablet_sizes virtual table
load_stats: update tablet sizes after migration or rebuild
This commit doesn't change the logic behind the view building worker but
it changes how the worker is executing view building tasks.
Previously, the worker had a state only on shard0 and it was reacting to
changes in group0 state. When it noticed some tasks were moved to
`STARTED` state, the worker was creating a batch for it on the shard0
state.
The RPC call was used only to start the batch and to get its result.
Now, the main logic of batch management was moved to the RPC call
handler.
The worker has a local state on each shard and the state
contains:
- unique ptr to the batch
- set of completed tasks
- information for which views the base table was flushed
So currently, each batch lives on a shard where it has its work to do
exclusively. This eliminates a need to do a synchronization between
shard0 and work shard, which was a painful point in previous
implementation.
The worker still reacts to changes in group0 view building state, but
currently it's only used to observe whether any view building tasks was
aborted by setting `ABORTED` state.
To prepare for further changes to drop the view building task state,
the worker ignores `IDLE` and `STARTED` states completely.
During the initial implementation of the view builing coordinator,
we decided that if a view building task fails locally on the worker
(example reason: view update's target replica is not available),
the worker will retry this work instead of reporting a failure to the
coordinator.
However, we left return type of the RPC, which was telling if a task was
finished successfully or aborted.
But the worker doesn't need to report that a task was aborted, because
it's the coordinator, who decides to abort a task.
So, this commit changes the return type to list of UUIDs of completed
tasks.
Previously length of the returned vector needed to be the same as length
of the vector sent in the request.
No we can drop this restriction and the RPC handler return list of UUIDs
of completed tasks (subset of vector sent in the request).
This change is required to drop `STARTED` state in next commits.
Since Scylla 2025.4 wasn't released yet and we're going to merge this
patch before releasing, no RPC versioning or cluster feature is needed.
