5.3 Changelog¶
On this page
5.3.2 Changelog¶
Sharding¶
- SERVER-62690 Shard is shutting down before finishing draining in test
- SERVER-64580 Step downs while sharding a collection in mixed binaries shards might crash the old binary
- SERVER-65821 Deadlock during setFCV when there are prepared transactions that have not persisted commit/abort decision
- SERVER-65924 Modify resharding_histogram_metrics.js to allow either “ops” or “totalCount” as the histogram field name
- SERVER-65930 DDL coordinators and rename participant initial checkpoint may incur in DuplicateKey error
- SERVER-66041 Chunk cloner must never consider too big a chunk with only one document
Internals¶
- SERVER-55173 Segmentation fault in WiredTigerSession::releaseCursor
- SERVER-56003 ninja + icecream with changing compilers doesn’t regen run-icecc.sh
- SERVER-60105 Inserts into timeseries collections are not observable from any opcounter
- SERVER-60485 Tasks which run on the MigrationUtilExecutor must not wait for shutdown to complete
- SERVER-61210 sample_timeseries.js can fail spuriously because of the trail phase
- SERVER-61460 Resmoke should merge config_svr options with mongod_options rather than overriding them
- SERVER-61853 suitesconfig.get_suite is occasionally unable to find test suite
- SERVER-62205 Include sanity check for max chunk size argument of [auto]splitVector
- SERVER-62513 RunDBCheckInBackground should retry on Interrupt errors
- SERVER-62946 Use failpoints to exercise audit log rotation failure on startup
- SERVER-63070 Minor bug in keystore.cpp
- SERVER-63254 Add Index usage metrics to serverStatus
- SERVER-63522 Reject user DDL operations when user write blocking is enabled
- SERVER-63531 commitQuorum incorrectly includes buildIndexes:false nodes and error message incorrectly says that only voting nodes are eligible
- SERVER-63850 Add the count command to API version 1
- SERVER-63910 Generalize this error message related to the critical section
- SERVER-64031 serverStatus should not take PBWM lock
- SERVER-64059 Tenant migration recipient can miss updating the config.transactions table for those donor transactions that are committed at a timestamp <= startApplyingDonorOpTime.
- SERVER-64065 connection_status test fails under selinux
- SERVER-64102 $project field that references time-series meta field can be referenced by second $project field
- SERVER-64118 Don’t run encrypt-related JS tests on SBE yielding variant
- SERVER-64183 Add requires-gcm tag to kmip_activate_rollover_rotate.js
- SERVER-64244 RunDBCheckInBackground should be resilient to interrupts
- SERVER-64250 add requires_replication tag to geo_array1.js
- SERVER-64322 reduce resmoke jobs for external_auth on windows
- SERVER-64403 Find queries with SORT_MERGE collation-encode the missing sort attribute
- SERVER-64405 Remove FCBIS passthroughs from non-enterprise variants
- SERVER-64440 Investigate enabling libunwind for s390x and ppc64le
- SERVER-64485 Use _id to determine the update type in extractUpdateType()
- SERVER-64509 Fix ShardRegistry threadpool shutdown order
- SERVER-64519 Critical section reason for create collection is multiversion incompatible
- SERVER-64552 Replace invariant with assertion for non-joinable rename collection participants
- SERVER-64617 [5.3] Allow new unique index data formats to exist
- SERVER-64664 ninja tool should not consider install files generated source
- SERVER-64725 Make ShardRegistry::periodicReloader interruptible
- SERVER-64727 Make all DDL coordinator documents parsing not strict
- SERVER-64757 Improve error messages when scons fails to generate-ninja on Windows
- SERVER-64780 Resharding change stream events are not filtered by namespace
- SERVER-64815 Prevent buildindexes_false_commit_quorum.js from running in multiversion tests
- SERVER-64949 Disable the CST benchmarks
- SERVER-64983 Release Client lock before rolling back WT transaction in TransactionParticipant::_resetTransactionState
- SERVER-64999 Disable, rehabilitate, and re-enable sanitizer builds under icecream
- SERVER-65000 Delete Python fixtures and hooks used exclusively in the replicator project
- SERVER-65024 Multiple documents with the same _id value make reIndex invariant
- SERVER-65032 Pin python package BaseResponse for ocsp suite
- SERVER-65137 detect namespace changes when refreshing Collection after yielding
- SERVER-65180 [5.3] remove v4, cxx20, and macOS arm64 builders
- SERVER-65200 Package task not run as part of task group
- SERVER-65270 Multi-planned SBE queries involving large documents do not respect the ‘sort’ option
- SERVER-65284 Create collection coordinator should always perform cleanup on subsequent execution
- SERVER-65285 Gracefully handle empty group-by key when spilling in HashAgg
- SERVER-65429 Balancer defragmentation policy should be cleared on step down
- SERVER-65539 jepsen_list-append failures do not upload test files
- SERVER-65672 Fix python package pinning difficulties on macOS ARM
- SERVER-65718 Fix mypy error
- SERVER-65797 Remove invalid index specs in memory before parsing for listIndexes
- SERVER-65852 Fix issue in ese suite kmip tests
- SERVER-65863 DocumentSourceGroup::optimize() can leave DocumentSourceGroup::_idExpression in an inconsistent state
- SERVER-65907 Delete the CST benchmark
- SERVER-65989 Backport pin cryptography fixes
- SERVER-66054 Internal transactions tests in jstests/multiversion are not running on evergreen
- SERVER-66560 Check feature flag before fetching child transaction history
- SERVER-66719 dbCheck FCV lock upgrade causes deadlock with setFCV
5.3.1 Changelog¶
Query¶
SERVER-64812 SBE interprets objects containing aggregation expressions incorrectly, leading to error response
Internals¶
- SERVER-56003 ninja + icecream with changing compilers doesn’t regen run-icecc.sh
- SERVER-60426 Spurious rebuilds of the intel decimal library when using SCons
- SERVER-63055 gcov and clang cause failures in ValidateCollections
- SERVER-63479 Correct $$SEARCH_META ban in sharded environments
- SERVER-63877 Determine if resharding should migrate internal sessions for non-retryable writes
- SERVER-63880 [Retryability] Make resharding handle applyOps oplog entries with WouldChangeOwningShard sentinel noop entry
- SERVER-64331 Determine if chunk migration should migrate internal sessions for non-retryable writes
- SERVER-64410 Investigate whether we need to sort on searchScore when storedSource is true in $search
- SERVER-64579 Shorten the names for retryable internal transactions jstests
- SERVER-64796 Clustered Index Bug
5.3.0 Changelog¶
Sharding¶
- SERVER-34322 Best effort ssv to recipient shard will never succeed for the first time
- SERVER-45149 replSetStepDown command in txn_two_phase_commit_failover.js should not timeout
- SERVER-57766 Omit shardingStatistics.resharding section when resharding hasn’t been run on the cluster
- SERVER-58036 TTL monitor can delete documents during chunk migration’s and resharding’s critical sections
- SERVER-58275 Deduplicate code to make a noop write to participants in ShardingDDLCoordinators
- SERVER-58622 DDL coordinator handle write concern error incorrectly when removing coordinator document
- SERVER-58701 Enable Feature flag for PM-2423
- SERVER-59053 VectorClock is not gossiped in sessions started by a direct-to-shard command
- SERVER-59832 Prevent writes to orphan documents
- SERVER-60109 Ensure vector clock is recovered on step-up
- SERVER-60624 txn_commit_optimizations_for_read_only_shards.js pauses replication on coordinator and can leave transaction stuck in prepare
- SERVER-61003 ReadConcernMajorityNotAvailableYet errors from ShardRegistry must be retried
- SERVER-61105 The sessions migration logic during moveChunk pollutes the logs
- SERVER-61249 refine_collection_shard_key_basic.js relies on a best-effort refresh that might not happen in case of failover
- SERVER-61444 Resharding uses of bumpCollectionVersionAndChangeMetadataInTxn are not idempotent
- SERVER-61478 Create fsm workload to check that no change stream events are generated due to writes on orphan documents
- SERVER-61533 Make the balancer defragmentation policy resilient to non-retryable errors
- SERVER-61584 Remove BalancerChunkMerger class
- SERVER-61628 Do not reschedule range deletion task when numDeleted < numDocsToRemovePerBatch
- SERVER-61637 Review range deleter batching policy
- SERVER-61735 Add a new timing log step in MigrationDestinationManager to track the time waiting for critical section release
- SERVER-61755 Migration recovery should handle refined shard key
- SERVER-61760 The new implementation of CollMod should rely on the AllowMigrations to stop/resume migrations
- SERVER-61767 Optimizing the new way of filtering writes over orphan documents
- SERVER-61840 create_index_background_partial_filter.js and create_index_background_wildcard.js should not run on _with_balancer suites
- SERVER-61847 Handle direct updates/deletes to shards that modify orphan docs
- SERVER-61880 Remove TODOs to create a single defragmentation thread client
- SERVER-61938 ConfigureCollectionAutoSplit should check maxChunkSize in MB
- SERVER-61940 Make balancer_defragmentation_merge_chunks test config stepdowns
- SERVER-61945 Resharding collection cloning may fail with NamespaceNotSharded when “nearest” read preference chooses secondary
- SERVER-61950 ReshardingOplogFetcher waits on network request completing without interruption, potentially preventing shard step-up from ever completing
- SERVER-61968 Fix the FCV-code that enables the long names support on the createCollection path
- SERVER-61969 Simplify exact_shard_key_target.js so that it does not depend on orphan writing
- SERVER-61976 [Resharding] Shards can error while refreshing their shard version following step-up, stalling the resharding operation
- SERVER-62036 Failures in preCacheMongosRoutingInfo() will cause mongos to fail startup
- SERVER-62065 Upgrade path from 3.6 to 4.0 can leave chunk entries without history on the shards
- SERVER-62072 _configsvrReshardCollection may return without having waited for unsetting “reshardingFields” to replicate to majority
- SERVER-62128 Do not assume FCV stability while filtering orphan documents in update/delete ops
- SERVER-62149 Reuse ShardKeyPattern object from collection metadata in the write path
- SERVER-62171 Add logs to output of runConcurrentMoveChunk in sharding_statistics_server_status.js
- SERVER-62178 Resharding can fail with NamespaceNotSharded if recipient primary fails over before creating temporary resharding collection
- SERVER-62207 ReshardCollection with small maxTimeMS can crash the shard due to incorrect BSON object lifetime
- SERVER-62233 Make SessionCatalogMigrationSource handleWriteHistory filter out oplogs outside of the chunkRange with OpType ‘n’
- SERVER-62245 MigrationRecovery must not assume that only one migration needs to be recovered
- SERVER-62281 Ensure drain of migrations pending recovery on FCV upgrade/downgrade
- SERVER-62282 Migration recovery should be retried until success
- SERVER-62296 MoveChunk should recover any unfinished migration before starting a new one
- SERVER-62315 Cache the sharding ownership filter in the update/delete stage
- SERVER-62332 RefineCollectionShardKeyCoordinator doesn’t disallow migrations while it’s executing
- SERVER-62343 Create Feature flag for PM-2664
- SERVER-62418 Drain recoverable ReshardCollectionCoordinators on FCV downgrade
Replication¶
- SERVER-54374 Race between signalOplogWaiters and StorageEngine::loadCatalog
- SERVER-56756 Primary cannot stepDown when experiencing disk failures
- SERVER-60053 More info logs during replication startup sequence
- SERVER-62151 Restarting the node could trigger rollback which closes the client connection and fail x509_cluster_auth_rollover.js
Query¶
- SERVER-56932 Make constant folding optimization throw a special kind of error
- SERVER-58750 Investigate whether InternalPlans::indexScan() executor always scans forward
- SERVER-59754 Incorrect logging of queryHash/planCacheKey for operations that share the same $lookup shape
- SERVER-61962 Rename db/exec/delete.h/.cpp to delete_stage.h/.cpp
- SERVER-62230 Forward port new exhaust cursor tests to the master branch
Write Operations¶
SERVER-14669 Updates/deletes on sharded collections shouldn’t affect orphan documents
Storage¶
- SERVER-30846 Run dbCheck as background workload in FSM tests
- SERVER-62350 replace usage of read_before_oldest with reset_snapshot
Operations¶
- SERVER-21070 Add option to gather collection stats
- SERVER-61491 Log stacktraces to file
- SERVER-61574 CSFLE: Create shared library on Windows and add to evergreen.
- SERVER-61595 CSFLE: investigate how gdb interacts with static library
Build and Packaging¶
- SERVER-50465 Package tests not run on ARM hosts
- SERVER-61480 Improve efficiency of change stream pre-image loading
- SERVER-61885 Add showExpandedEvents flag
- SERVER-61892 Replace documentKey paths cache with document key from the oplog
- SERVER-61894 Support the ‘showRawUpdateDescription’ option in change stream specs
- SERVER-61959 Backport MSVS 2022 support into vendored SCons
Internals¶
- SERVER-27209 BSONObj::getStringField() does not handle embedded null bytes correctly
- SERVER-42505 Periodically check if KMIP key is in the Active State
- SERVER-42971 ThreadClient should restore original thread name on destruction
- SERVER-49015 Automatically install db-contrib-tool
- SERVER-49230 Evaluate upgrade of ZStandard to 1.4.9+
- SERVER-50081 Support validation repair mode with duplicates on unique indexes
- SERVER-51218 “resmoke.py run –help” doesn’t show options’ default values
- SERVER-51456 Database Profiler outputs incorrect value of property “keysDeleted” for a remove operation when a write conflict occurs
- SERVER-51842 Fix duplicate localThresholdMs (localThreshold) server parameters
- SERVER-51851 Create Feature flag for Selective Backup and Restore
- SERVER-51983 Create feature flag for Bucket unpacking with sorting
- SERVER-52297 Enable feature flag for Filling missing values based on surrounding values
- SERVER-52373 Enable feature flag for KMIP Activate
- SERVER-52991 Complete TODO listed in SERVER-32198
- SERVER-53993 Attach client strand before releasing the opCtx in AsyncCommandExecution tests
- SERVER-54592 Enable Feature flag for PM-2189
- SERVER-54900 Blocking networking calls can delay sync-source resolution indefinitely
- SERVER-55030 Remove mutexes that serialize secondary and majority read operations
- SERVER-55043 canonicalizeBSONType() should be implemented as a lookup table
- SERVER-55055 Add cumulative metric for the total refreshed connections
- SERVER-55816 Investigate negative timeouts in connection pool
- SERVER-56146 Incorrect commitQuorum error message
- SERVER-56346 Update certificate definitions to match MacOS 10.15+ requirements
- SERVER-56387 Check optionalBool-typed IDL fields for backward compatibility
- SERVER-56578 Create Feature flag for PM-2290
- SERVER-56972 Architecture Guide updates for PM-2311
- SERVER-56974 Enable Feature flag for PM-2311
- SERVER-57122 Make RecoveryUnit decorable and use it for uncommited collections and collection catalog updates
- SERVER-57188 Remove mongos binVersion check in retryable_write_error_labels.js
- SERVER-57312 Pin transitive Python dependencies and use the pinned file for installation in Evergreen
- SERVER-57383 Include merge sort in sharded aggregation explain output
- SERVER-57411 Enable Feature flag for PM-2321
- SERVER-57416 Create Feature flag for PM-2323
- SERVER-57516 RWC defaults pollutes the logs with duplicate “Refreshed RWC defaults” messages
- SERVER-57546 Increase test-specific timeouts for roles lock acquisition
- SERVER-57814 Create Feature flag for PM-2333
- SERVER-57962 Create Feature flag for PM-2346
- SERVER-58035 Remove db.runCommandWithMetadata from mongo shell
- SERVER-58134 Map agg comparison predicates on time-series collections to bucket-level fields
- SERVER-58151 Architecture Guide updates for Remove Faulty Mongos From Cluster Topology
- SERVER-58153 Enable Feature flag for Remove Faulty Mongos From Cluster Topology
- SERVER-58257 Add better diagnostics to predictive_connpool.js
- SERVER-58310 ThreadPoolTaskExecutor is memory unsafe when task cancellation occurs around the same time an exhaust network response is received
- SERVER-58358 Datasize command does not validate the syntax and can cause major performance degradation
- SERVER-58410 Discontinue writing to appliedThrough/minValid as part of secondary batch application
- SERVER-58426 Add planning logic for $lookup
- SERVER-58436 Implement spilling HashAgg
- SERVER-58437 Implement pushdown logic for $lookup
- SERVER-58505 Add metric to serverStatus for number of load balanced connections
- SERVER-58506 Command to expose settability for Server Parameters
- SERVER-58518 Report progress of collection defragmentation through balancerCollectionStatus command
- SERVER-58530 Remove unused code related to nReturnedSoFar()
- SERVER-58672 Create featureFlagPerShardCursor
- SERVER-58692 Introduce cluster-wide pre-image expiration parameter
- SERVER-58694 Implement writing of pre-images for transactional update/replace/delete operations
- SERVER-58865 Remove obsolete restriction on capped collection objects
- SERVER-58875 Implement missing optimizations for $set* expressions
- SERVER-58920 Enable multiversion testing of rewrite_state_change_errors.js
- SERVER-59002 Create Feature flag for PM-2449
- SERVER-59163 Enable partial indexes on time-series collections
- SERVER-59220 Connectivity probes in ocsp_server_refresh.js should use fresh shells
- SERVER-59241 Allow mongosymb.py to stream from stdin.
- SERVER-59260 Create Feature flag for PM-2466
- SERVER-59290 Re-evaluate sync source after incrementing config version
- SERVER-59382 Enforce non-critical facets not entering ActiveFault state
- SERVER-59384 Should provide ability to perform periodic DNS health checks
- SERVER-59386 Should provide the ability to perform periodic LDAP health checks
- SERVER-59390 Should provide the ability to perform periodic health checks against config server
- SERVER-59391 Should fault if LDAP facets are enabled, but the server feature is not configured.
- SERVER-59395 Setup integration test for DNS failure scenario
- SERVER-59409 Race between reconfig replication and stepup can cause RSM to be stuck in reporting ReplicaSetNoPrimary
- SERVER-59521 Remove TODO linked to SERVER-47602
- SERVER-59657 Add ErrorExtraInfo for the WriteConflictException and use in transaction_participant.cpp updateSessionEntry()
- SERVER-59682 Recover SBE plans from the new plan cache
- SERVER-59689 Prevent setting FCV < 5.3 in case of ongoing defragmentation
- SERVER-59696 Include SBE plan cache as part of cache size metric
- SERVER-59700 Add programming support for tracepoints
- SERVER-59772 Enable setWindowFields in transactions or with readConcern snapshot
- SERVER-59779 Call asCluster() before replSetFreeze in ReplSetTest
- SERVER-59813 Allow IDL server parameters with “none” type for set_at
- SERVER-59839 ShardServerProcessInterface::getCollectionOptions returns format differently from non-sharded versions
- SERVER-59912 Initial no-op scaffolding of the Ldap health checker
- SERVER-59951 Make object form of the ‘_id’ group-by expression work to handle multiple group-by keys.
- SERVER-59958 Extend multi-collection locking to the AutoGetCollectionForRead/Command classes
- SERVER-59960 Ensure CompileCtx::root is set before accessing it
- SERVER-60059 Add a mongo packaged_task
- SERVER-60063 Log server discovery times
- SERVER-60079 Common health observer code to incorporate generic patterns from Ldap observer
- SERVER-60112 Audit server commands for accepted numeric types and accept int32 where necessary
- SERVER-60117 Try removing … exception handler in ASIOSession constructor
- SERVER-60123 Support using clusterKey {_id : 1} for shardKey in a sharded collection
- SERVER-60176 Delta-updates should only validate the diff for storage
- SERVER-60184 Create Feature flag for PM-2550
- SERVER-60197 Support $group pushdown to SBE when $match stage with $or exists in a pipeline
- SERVER-60210 Create a future git tag evergreen variant
- SERVER-60235 Extend&develop performance tests for change streams with pre-/post-images enabled
- SERVER-60240 Ensure pre-/post-images are available after a resharding operation
- SERVER-60241 Grant access to the pre-images collection for root user
- SERVER-60251 Add tests for change stream pre-images/post-images with retryable writes
- SERVER-60254 Implement $group pushdown when sub-planning is involved
- SERVER-60298 Explain can include ICU collation keys encoded as strings, which can be invalid UTF-8
- SERVER-60300 Support pushdown of sugared $group stages
- SERVER-60311 Add execution stats etc for pushed down $group
- SERVER-60316 FaultManager should start with periodic checks disabled
- SERVER-60331 External sorter can use more memory than the configured maximum
- SERVER-60349 Make IDL compatibility checker allow addition of fields with type ‘optionalBool’
- SERVER-60356 Performance test for File Copy Based Initial Sync
- SERVER-60390 Replace incorrect assertion in AtomicWord with is_always_lock_free from TODO C++17
- SERVER-60412 Host memory limit check does not honor cgroups v2
- SERVER-60421 raise the log level in topology_state_machine.cpp
- SERVER-60447 Complete TODO listed in SERVER-60202
- SERVER-60479 Implement split chunks routine Phase III
- SERVER-60500 Implement linear interpolation window function
- SERVER-60524 [Retryability] Make retryable internal transactions retryable across data placement changes
- SERVER-60533 Distinct preceded by sharded collection drop fails with “database not found”
- SERVER-60587 Implement FaultFacet
- SERVER-60596 Replace boost::optional::get() with boost::optional::value in DensifyValue to avoid undefined behavior
- SERVER-60598 Investigate optimizing repetitive sorts from $densify and $fill
- SERVER-60657 Create Feature flag for PM-2584
- SERVER-60672 Simpler pushdown when timeseries collection has no mixed-schema buckets
- SERVER-60694 Create DDL coordinator for collMod command
- SERVER-60728 Improved MDB crash recovery testing
- SERVER-60733 Test create collection after drop collection
- SERVER-60817 setDefaultRWConcern accepts invalid write concern options
- SERVER-60832 Code Coverage variant not generating coverage data
- SERVER-60846 Remove double severity in favor of discrete type
- SERVER-60848 Log which engine was used for processing a query
- SERVER-60876 Run testing under WT stress configuration
- SERVER-60891 Ensure Ninja compile tasks run on suggested builders
- SERVER-60911 Update FCV references for secondary indexes on time-series measurements when kLatest is 5.3
- SERVER-60917 [Retryability] Make transaction participants throw RetryableTransactionInProgress if a retry arrives while the transaction has not been committed or aborted
- SERVER-60939 Fix disabled Query Opt tests failing due to changing latest release to 5.2
- SERVER-60944 Change interface for updateWithSuppliedFacet and update
- SERVER-60959 Insert to a Time-Series getting error collection ErrorCodes::TimeseriesBucketCleared
- SERVER-60974 Multiversion suites are overwritting receiveChunkWaitForRangeDeleterTimeoutMS
- SERVER-61000 Coverity analysis defect 121043: Uninitialized pointer field
- SERVER-61005 rs.initiate() fails with “Invariant failure” under specific startup options
- SERVER-61009 Make createIndex a no-op on a cluster key if the collection exists
- SERVER-61011 Evaluate RecordId size limit for clustered collections
- SERVER-61020 Add negative test case for NIST GCM vectors
- SERVER-61068 Add evergreen task for scale 1 normalized TPC-H benchmark
- SERVER-61069 socket_disconnect_kills.js doesn’t check that operations fail with network error
- SERVER-61085 [Upgrade/downgrade] Do not persist txnRetryCounter if value is default
- SERVER-61088 [Upgrade/downgrade] Make transaction participants check if txnRetryCounter is supported during startTransaction and have mongos use it
- SERVER-61090 [Upgrade/downgrade] Make setFCV wait for all TransactionCoordinators of internal transactions to be cleaned up
- SERVER-61091 [Upgrade/downgrade] Test that there cannot be in-progress transactions in child sessions when FCV is about to be set to fully downgraded
- SERVER-61107 Serialize concurrent accesses to Socket in TCPResetAfterConnectionIsSilentlySwallowed
- SERVER-61130 Create backup cursor on donor primary
- SERVER-61131 Store backup cursor results and set state to “learned filenames”
- SERVER-61132 Keep backup cursor alive until file copy finishes
- SERVER-61133 Copy donor files to temp directory
- SERVER-61154 Wrong default minSnapshotHistoryWindowInSeconds in sharding architecture guide
- SERVER-61184 Compound 2dsphere indexes can generate a massive number of index keys per document
- SERVER-61185 Use prefix_search for unique index lookup
- SERVER-61203 Add record metadata and history in validate output when corruption is detected
- SERVER-61220 Integration test for Progress monitor
- SERVER-61241 Add a test to ensure plan cache size serverStatus metric is decreased when an entry is evicted
- SERVER-61260 Support collations on clustered collections
- SERVER-61277 Allow concurrent writes on capped clustered collections
- SERVER-61278 Guard clustered capped collections behind the enableTestCommands parameter
- SERVER-61285 Capped parameter for the create command is currently stable
- SERVER-61287 IDL compatibility checker mishandles enterprise module
- SERVER-61297 Complete TODO listed in SERVER-44644
- SERVER-61300 Investigate memory usage of exact-topN operators for in memory storage engine in jstests/aggregation/spill_to_disk.js
- SERVER-61315 Ldap health check executor should support aborted tasks
- SERVER-61354 Audit log creates empty file and rotates when one does not exist
- SERVER-61367 Incorrect tag for shard split feature flag test
- SERVER-61368 FaultManager test suite should use real thread pool
- SERVER-61383 Refactor expired pre-image purging job to perform range-deletion
- SERVER-61385 Migrate most callers of legacy ‘DBClientBase::query()’ API to modern ‘find()’ API
- SERVER-61391 clustered_collection_passthrough is not running with generated tasks
- SERVER-61420 Create MatchExpression visitor to set paramId on tree nodes
- SERVER-61425 Split donor monitors recipient for split acceptance
- SERVER-61426 Legacy mongo shell sends authenticate command with database “admin” for X.509
- SERVER-61442 Verify that $hinted writes on cluster key result in bounded collscans
- SERVER-61447 Investigate skipping argument expression evaluation in $group for $first/firstN once accumulator has necessary values
- SERVER-61463 Create javascript library for adding and removing entries from config.shards
- SERVER-61472 Coverity analysis defect 121149: Uninitialized scalar variable
- SERVER-61474 Ban directoryPerDb and directoryForIndexes with Merge
- SERVER-61488 Enable $linearFill without a flag guard
- SERVER-61501 Create sharding suite where collections are clustered by default
- SERVER-61505 Remove checks in initial sync code that check the sync source wireVersion for resumable initial sync feature compatibility
- SERVER-61529 Ldap tests should wait for connection reaper to be terminated
- SERVER-61551 IDL compatibility checker should check chained structs
- SERVER-61558 Rename configureCollectionAutosplitter command
- SERVER-61583 Add restart test for retryable internal transactions for findAndModify
- SERVER-61588 Incorrect header comment for CatalogCache::onStaleDatabaseVersion
- SERVER-61592 Querying with ms precision does not return expected results (TS collection)
- SERVER-61608 Unreachable code in topology_coordinator.cpp
- SERVER-61616 Restrict Command execution in SecurityToken mode
- SERVER-61617 Limit actionType/matchType combinations in SecurityToken mode
- SERVER-61630 Serialize a string sbe::value::RecordId to KeyString
- SERVER-61635 Validate that the defragmentation algorithm is fairly executed across collections
- SERVER-61654 Implement the Last point on timeseries optimization use case 1
- SERVER-61662 SCons configure checks should always run with verbosity
- SERVER-61663 Dynamically split concurrency_metrics and concurrency_replication_metrics into smaller tasks
- SERVER-61686 Adding a JS test file to backports_required_for_multiversion_tests.yml doesn’t disable all multiversion tests.
- SERVER-61696 Make change stream pre- and post- image tests multiversion testing compatible
- SERVER-61699 Failure to rotate audit log on startup does not cause node startup to fail
- SERVER-61703 Complete TODO listed in SERVER-53642
- SERVER-61709 Introduce implicitly replicated namespaces
- SERVER-61726 Synchronise the execution of the defragmentation VS autosplit on every collection.
- SERVER-61727 Implement the “move and merge chunks” phase of the Collection Chunks defragmentation algorithm
- SERVER-61732 Prevent threads spawned outside ThreadContextTest from updating test counters
- SERVER-61737 Make a deep copy of RuntimeEnvironment for SBE plans to cache
- SERVER-61741 Add evergreen task for scale 1 of the denormalized TPC-H benchmark
- SERVER-61744 Handle a new WiredTiger system log component hierarchy in LOGV2
- SERVER-61754 dbCheck should not hold a strong collection lock during batches
- SERVER-61757 Add a dbCheck command option to customize the batch size
- SERVER-61772 Implement tenant access blocking for the split service
- SERVER-61773 Enable WT stable timestamp rule
- SERVER-61795 Incorrect Usage of Javascript Comparison to Compare Arrays and Objects
- SERVER-61817 Complete TODO listed in SERVER-58759
- SERVER-61819 Fix bug in SBE around capped collections and (new) yielding
- SERVER-61821 Retire usage of OwnedPointerMap
- SERVER-61835 Fix how SBE plan cache deals with ShardFilterer
- SERVER-61836 The TimestampMonitor must ignore killOp interruptions and retry/continue
- SERVER-61839 $elemMatch w/ positional $ project SBE bug
- SERVER-61844 Allow mongos and mongoq to have different startup initialization
- SERVER-61846 Prevent redact from throwing BSONObjectTooLarge
- SERVER-61852 dbCheck should tryLock the collection with backoff
- SERVER-61854 Move collmod_convert_to_unique_violation_count.js to core
- SERVER-61861 dns_container_tester.js - wait for dnsmasq startup
- SERVER-61868 $max accumulator with collation might produce incorrect result for NaN
- SERVER-61870 Retry loading oplog entries on CappedPositionLost in retryable_internal_transaction_test.js
- SERVER-61872 Fix thread pool starvation in FaultManager
- SERVER-61873 LDAP health observer runtime params
- SERVER-61877 Remove catalog consistency verification from dbCheck
- SERVER-61879 Refreshes to recover migrations must never join ongoing refreshes
- SERVER-61882 Update FCV constants to recognize 5.3 as latest
- SERVER-61891 Add document key to the insert oplog entry
- SERVER-61900 Coverity analysis defect 121221: Parse warning
- SERVER-61901 Coverity analysis defect 121224: Uninitialized scalar variable
- SERVER-61902 Coverity analysis defect 121189: Not restoring ostream format
- SERVER-61904 Coverity analysis defect 121223: Wrapper object use after free
- SERVER-61911 Remove consistency check exception for ‘system.preimages’ collection
- SERVER-61913 Investigate tracking deletes for unique index conversion
- SERVER-61914 add fault facet details to FaultImpl::toBSON
- SERVER-61916 Ensure all no-op entries are inserted before the recipient reads opLog in the unit test OplogApplierResumesFromLastNoOpOplogEntry
- SERVER-61921 Link failure in noSSL mode in FaultManager
- SERVER-61923 Shell’s consistency checker can mask an error
- SERVER-61926 clean up geo_circle1.js
- SERVER-61929 Add additional SConstruct architecture checks
- SERVER-61930 Individual health observers should return an error if a timeout period elapses when doing a single health check
- SERVER-61931 Allow ClusterManager role to operate against system.buckets.* collections
- SERVER-61932 Returns all duplicates from concurrent writes during collMod converting to unique index
- SERVER-61934 Race between creating new opCtx and killing opCtx’s before switching out the storage engine
- SERVER-61939 Explore bounding clustered collection scans more tightly
- SERVER-61941 $sortArray comparator does not satisfy “Compare” requirement with -1 sort
- SERVER-61946 Remove requires_fcv_51 tag from internal sessions multiversion tests
- SERVER-61947 health check interval should be propery of the health observer
- SERVER-61954 Improve decoding performance for BSONColumn
- SERVER-61955 Expose dbCheck as a generally available command
- SERVER-61956 fix data race when accessing the state machine’s state
- SERVER-61958 Make a mocked health observer class that takes in failpoint data
- SERVER-61964 FCBIS needs to handle transient error when opening backup cursor
- SERVER-61965 Coverity analysis defect 121236: Unchecked return value
- SERVER-61974 Call quickExit() in mongoauditdecrypt before returning from main()
- SERVER-61977 Concurrent rollback and stepUp can cause a node to fetch from a timestamp before lastApplied once it has stepped down.
- SERVER-61979 extend _parallelShellRunCommand() to support additional result validation
- SERVER-61980 collMod does not consider unique option in no-op hidden parameter request processing
- SERVER-61983 Create TenantNamespace class
- SERVER-61984 Create TenantDatabase class
- SERVER-61986 Add TenantNamespace to Collection object
- SERVER-61987 Change DatabaseHolder’s map to be keyed by TenantDatabase
- SERVER-61988 Change CollectionCatalog maps that are keyed by db name to be keyed by TenantDatabase
- SERVER-61989 balancer_defragmentation_merge_chunks should recreate collection between tests
- SERVER-61993 BSONColumnBuilder::skip need to close interleaved mode if reference obj contain empty subobjects
- SERVER-61996 Improve performance of DocumentStorage::appendField
- SERVER-61997 Object with only empty sub object should not start subobj compression in BSONColumn
- SERVER-62000 Do not generate “unique” field in oplog if collMod “unique” parameter is a no-op
- SERVER-62003 Null-equality on ‘fullDocument’ is not rewritten correctly for change stream ‘delete’ events
- SERVER-62004 Add unittest coverage for storage engine API changes
- SERVER-62016 Write concern accepts embedded replica set tagset
- SERVER-62018 ReplSetConfig supports an optional splitConfig subdocument
- SERVER-62019 Remove flag guard from $linearFill
- SERVER-62020 [Retryability] Make retryable write statements executed with or without internal transactions retryable whether or not internal transactions are involved on retries
- SERVER-62021 Allow replacement of local replica set config
- SERVER-62022 Reduce dbCheck info logging in production, log start and stop
- SERVER-62023 Improve dbCheck observability
- SERVER-62024 Enable new yielding behavior for find/getMore operations by default
- SERVER-62031 Complete TODO listed in SERVER-58962
- SERVER-62037 Fix replica set definition in the system perf yaml file for linux-1-node-15gbwtcache
- SERVER-62038 Tag file copy based initial sync tests with requires_fcv_52
- SERVER-62039 Include TenantID in Javascript scope identifier
- SERVER-62041 Add a maximum batch execution time to dbCheck
- SERVER-62045 Add basic jstest coverage for $bucketAuto for the exact topN accumulators
- SERVER-62046 Remove ActionType::emptycapped from Builtin Roles
- SERVER-62047 [Retryability] Enforce txnNumber ordering in a session in the presence of retryable internal transactions
- SERVER-62049 Add debugging aids for some of the new SBE types
- SERVER-62050 Remove _opCtx from AutoGetDb and instead pass in an opCtx to AutoGetDb::ensureDbExists()
- SERVER-62051 Increase file handle limit in Windows CRT
- SERVER-62052 Usage of BSONColumn in the validate command can run out-of-memory
- SERVER-62053 Add retry for errors in AWS server-side conversation
- SERVER-62055 Enable tests in clustered_collection_passthrough that try to recreate the clustered index
- SERVER-62060 Skip stepdown tests for tenant_migration_disallowed_on_config_server jstest
- SERVER-62064 Serialize resharding with other DDL operations on stepup
- SERVER-62067 migrate ParsedCollModIndexRequest::indexHidden away from BSONElement
- SERVER-62068 migrate ParsedCollModIndexRequest::indexExpireAfterSeconds away from BSONElement
- SERVER-62069 remove ParsedCollModIndexRequest::indexObj
- SERVER-62070 clean up geo_nearwithin.js
- SERVER-62073 Make agg_graph_lookup more robust
- SERVER-62080 Make read and write concern configurable for internal transaction API
- SERVER-62081 Exhaustively test change stream rewrites against null-equality predicates
- SERVER-62084 Serializer for FaultFacetType is broken
- SERVER-62085 Use more bits for hashedMultikeyMetadataPaths in validation
- SERVER-62086 Update HashAggStage’s default maximum memory usage to the bigger value comparable to the DocumentSourceGroup’s
- SERVER-62091 Set small range deleter batch size in disable_resumable_range_deleter.js
- SERVER-62092 clean up and relocate geo_array1.js
- SERVER-62094 Incorrect memory usage calculation in $top/$bottom/$topN/$bottomN window functions due to Document field caching
- SERVER-62095 Update phase transitions for balancer defragmentation policy
- SERVER-62096 /proc/<id>/smaps is not available
- SERVER-62097 Generated tasks should inherit timeouts from build variants
- SERVER-62098 Guard healthCheckContexts with mutex in fault_manager.cpp
- SERVER-62101 Aggregation can run lock-free and expects a ViewCatalog access separate from the AutoGet*MaybeLockFree to always return a valid ViewCatalog – not guaranteed
- SERVER-62112 Re-add handleRIDRangeMinMax inside planner_access.cpp
- SERVER-62113 Ignore IndexBuildAborted errors when building indexes in find_flip_sbe_enabled.js
- SERVER-62126 Disable the rollback resumable index build test on the slow variant
- SERVER-62127 Exclude test scripts that record change stream pre-images from initial sync testing
- SERVER-62133 Fix multiversion issue in change_stream_pre_image_lookup_whole_db_whole_cluster.js
- SERVER-62134 Ensure Phase 3 is executed before ending defragmentation
- SERVER-62140 Reduce resmoke job factor for multi_stmt_txn_jscore_passthrough_with_migration on RHEL 8.0 inMemory
- SERVER-62146 [SBE] Fix use-after-free bug with $arrayElemAt, $first, and $last
- SERVER-62148 Move changeStreamPreAndPostImages collection option tests from timeseries to changeStream test file
- SERVER-62153 max_time_ms_repl_targeting.js should succeed if the test step succeeds at least once
- SERVER-62157 Explicitly link rpc_test to auth
- SERVER-62158 Fix or suppress novel UBSAN reports when building with the v4 clang
- SERVER-62160 Remove TODO listed in SERVER-62061
- SERVER-62161 Complete TODO listed in SERVER-60459
- SERVER-62167 Coverity analysis defect 121257: Uninitialized scalar field
- SERVER-62168 Use –oplogMinRetentionHours in tenant_migration_multi_writes.js to prevent the oplog from being truncated
- SERVER-62169 ViewCatalog::resolveView should handle when time-series bucket collection does not exist
- SERVER-62170 Increase initial sync logging level in FCBIS tests
- SERVER-62173 Applying dropIndexes oplog entry can get BackgroundOperationInProgressForNamespace if FCV not upgraded
- SERVER-62174 FaultManager to support dynamic configuration of health check intervals
- SERVER-62179 Roundtrip BSONColumn in the validate command after verifying no corruption is present
- SERVER-62180 Enable signal processing (or red/green) checks for Google Microbenchmark
- SERVER-62181 JStests including multiple parallel migrations with failpoints shouldn’t be run in the config server stepdown suites
- SERVER-62185 Ensure TenantMigrationRecipentService Fetcher instance survives future completion
- SERVER-62186 Update FCV reference for dropping an index while having a separate index build on the same collection
- SERVER-62187 Modify ‘originalSpec’ to be part of the unstable API for listIndexes
- SERVER-62191 Analog to burn_in_tests for running any user-specified test
- SERVER-62192 Handle feature flag disabled for implicit sharding accessed collection
- SERVER-62194 Add multiple collection locking capability to AutoGetCollection
- SERVER-62195 Refactor usage of non-standard Array.sort
- SERVER-62197 Remove extra state mutex lock.
- SERVER-62198 Fix Shutdown error with Progress Monitor
- SERVER-62200 Do not run cached_plan_trial_does_not_discard_work.js in parallel with fsyncLock tests
- SERVER-62202 Log ID 5936504 should contain observer type as string
- SERVER-62203 change the thread name “Health checks progress monitor” to “FaultManagerProgressMonitor”
- SERVER-62204 do not schedule health check if observer is not enabled
- SERVER-62208 Handle upgrade/downgrade scenario for collMod DDL coordinator
- SERVER-62210 Fix dbCheck progress meter handling of dropped and recreated collections
- SERVER-62212 Support writeConcern for dbCheck
- SERVER-62214 Use explicit read concern in InitialSyncer and CollectionCloner
- SERVER-62215 Improve the error message for logRotate command
- SERVER-62216 When abortTenantIndexBuilds failed to abort during a tenant migration, we should wait for the createIndex to finish before continuing the MTM
- SERVER-62217 inject_tenant_prefix.js override may not wait for newly elected primary after tenant migration commits
- SERVER-62225 FTSSpec::fixSpec() attempts to convert number outside the range of representable values of type ‘int’
- SERVER-62226 Disable dbcheck_no_history_on_secondary.js on EFT
- SERVER-62229 Fix invariant when applying index build entries while recoverFromOplogAsStandalone=true
- SERVER-62234 Add support for MSVS 2022 builds in C++17 and C++20 mode
- SERVER-62235 Add experimental MSVS builders to the master waterfall
- SERVER-62239 Put tenantId on opCtx if passed as $tenant
- SERVER-62240 Enable sharding for CRUD_clustered_collection.js
- SERVER-62241 Prohibit using arbitrary cluster key
- SERVER-62242 $indexOfArray does not work with duplicate values in array
- SERVER-62243 Wait for vector clock document majority-commit without timeout
- SERVER-62247 Avoid making copies of the vector in TargetedWriteBatch::getWrites()
- SERVER-62248 Add requires_fcv_51 tag to load_balanced_server_status_metrics.js
- SERVER-62251 Extend ServerParameter class to support cluster-wide server parameters
- SERVER-62255 Create base ClusterServerParameter class
- SERVER-62268 Handle balancer configuration refresh in balancer_defragmentation_merge_chunks.js
- SERVER-62277 Performance regression from dbstats due to occupied disk space calculation
- SERVER-62283 Temporary workaround of the problem in SERVER-61281
- SERVER-62285 validate cachedir add push failure debug messages
- SERVER-62286 Fix disabled Repl tests failing due to changing latest release to 5.3
- SERVER-62289 Fix disabled Security tests failing due to changing latest release to 5.3
- SERVER-62290 Fix disabled Query Optimization tests failing due to changing latest release to 5.3
- SERVER-62292 Add Task to Commit Queue to Ensure only Public SERVER Tickets are Referenced
- SERVER-62293 Race between recipientForgetMigration cmd and TenantMigrationRecipientService future chain restart on errors machinery.
- SERVER-62312 Enable feature flag in 4.4 backport and other housekeeping
- SERVER-62314 Fix unreliable testing intermediate state in file_copy_based_initial_syncer_test.cpp
- SERVER-62319 Fix hidden_index.js in multiversion suite
- SERVER-62321 Increase kActiveFaultDuration for fault manager test suite.
- SERVER-62322 Consistent validity treatment of empty objects (i.e., {})
- SERVER-62324 Failure to build mongo::unique_function uses with MSVS 2022 in C++20 mode
- SERVER-62333 MurmurHash3 getblock/putblock can overflow
- SERVER-62334 Regression following SERVER-46399
- SERVER-62336 Tolerate SnapshotTooOld errors in dbCheck testing
- SERVER-62340 Tenant Migration can lead to leakage of “TenantMigrationBlockerAsync” threads.
- SERVER-62347 Fix disabled clustered_collection.js test due to changing latest release to 5.3
- SERVER-62349 Disable featureFlagRequireTenantId for the all feature flags required evergreen tests
- SERVER-62351 Make plan executor creation aware of multiple collections
- SERVER-62352 Fix disabled Serverless tests failing due to changing latest release to 5.3
- SERVER-62360 Handle reconfiguring the WiredTiger LOGV2 component hierarchy
- SERVER-62361 Complete TODO listed in SERVER-52247
- SERVER-62364 Add correctness tests for arbitrary value in $fill
- SERVER-62365 Support a timeout for the shard split operation
- SERVER-62368 Range deleter must honor rangeDeleterBatchDelayMS
- SERVER-62371 Syntax error on rare code path in ldap_mongos_health_checking.js
- SERVER-62373 LDAP health check integration test should assert stats
- SERVER-62378 Remove improperly merged lines from unit test
- SERVER-62379 Fix deadlock between ReplicationCoordinator and BackgroundSync on stepUp
- SERVER-62380 Wait for majority commit point to be advanced on each node before starting rollback test in rollback_set_fcv.js
- SERVER-62381 CSFLE: remove shared library dependency to libsasl2
- SERVER-62382 Several Amazon Linux 2 tests on the 5.0, 5.1, 5.2 and master waterfalls are not running
- SERVER-62383 Use durable_timestamp instead of commit_timestamp for set_timestamp
- SERVER-62386 Coverity analysis defect 121298: Parse warning
- SERVER-62387 Coverity analysis defect 121302: Parse warning
- SERVER-62388 Coverity analysis defect 121318: Parse warning
- SERVER-62389 Allow waitAfterCommandFinishesExecution failpoint to match on command comment
- SERVER-62391 Add “tid” field to oplog entries
- SERVER-62392 Prevent timeseries lookup jstests from running in unsupported multiversion environments
- SERVER-62394 Follow up work for SERVER-61131
- SERVER-62397 Robustify exact_top_n_feature_flag.js
- SERVER-62399 Use of v3 tooclhain virtualenv is a python2 environment
- SERVER-62400 Add $_passthroughToShard parameter to AggregateCommandRequest
- SERVER-62404 Simplify mutex locking in fault_manager.cpp
- SERVER-62406 Create new ActionType for use of $tenant parameter
- SERVER-62419 recover_multiple_migrations_on_stepup.js fails when executed in config server stepdown suite
- SERVER-62422 Consolidate WriteConcernOptions parsing logic
- SERVER-62423 Fix replsetinitiate_works_with_keyfile_profile_verbose_options.js to work on ephemeralForTest
- SERVER-62424 _configsvrRemoveShard must wait for majority before returning
- SERVER-62426 Backup cursor indicates which files are required to copy for selective backups
- SERVER-62429 Add CollectionUUIDMismatch error code and corresponding ErrorExtraInfo
- SERVER-62430 Set resmoke_jobs_max=1 in Enterprise Windows DEBUG (Unoptimized) build variant