5.1 Changelog

Sharding

• SERVER-45149 replSetStepDown command in txn_two_phase_commit_failover.js should not timeout
• SERVER-56127 Retryable update may execute more than once if chunk is migrated and shard key pattern uses nested fields
• SERVER-56227 Add user-facing command to set allowMigrations to false for a sharded collection
• SERVER-60682 TransactionCoordinator may block acquiring WiredTiger write ticket to persist its decision, prolonging transactions being in the prepared state
• SERVER-60860 ReshardingCollectionCloner uses primary read preference when nearest was intended
• SERVER-61003 ReadConcernMajorityNotAvailableYet errors from ShardRegistry must be retried
• SERVER-61268 Fixing the release of a recoverable critical section on secondaries
• SERVER-61416 Indefinitely retry errors in rename coordinator
• SERVER-61459 ShardingCatalogManager::assignKeyRangeToZone() reads stale version of CollectionType while running locally on config server primary
• SERVER-61637 Review range deleter batching policy
• SERVER-61689 ActiveMigrationsRegistry::lock method doesn't properly handle its exceptions
• SERVER-61816 cancel_coordinate_txn_commit_with_tickets_exhausted.js can hang forever due to race condition between transaction reaper and transaction coordinator
• 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

Query

• SERVER-57588 Inconsistent query results when an array position is indexed whose value is an array
• SERVER-61548 Using $densify with partition on the nested field produces extra document and a flat partition field

Aggregation

SERVER-59924 Error executing aggregate with $out with "available" read concern on sharded clusters

Storage

• SERVER-58409 Startup RecordId initialization is flawed with durable history and reconstructing prepared transactions
• SERVER-60956 collMod command allows converting index on capped collection to TTL, leading to invariant failure

Internals

• SERVER-49748 Initial sync should clone admin.system.version before any other collections
• SERVER-57037 Improve precision of operator counters
• SERVER-58687 Implement writing of pre-images for non-transactional delete operations
• SERVER-59781 multi_statement_transaction.js does not retry transaction on StaleConfig
• SERVER-60310 OCSP response validation should not consider statuses of irrelevant certificates
• SERVER-60513 Run burn_in_tags compile on -large distro
• SERVER-60517 Enforce eviction_dirty_target < eviction_dirty_trigger in the fuzzer
• SERVER-60569 Clean up the OpObserver unittest TestFundamentalOnDeleteOutputs
• SERVER-60685 TransactionCoordinator may interrupt locally executing update with non-Interruption error category, leading to server crash
• SERVER-60690 Refactor OplogUpdateEntryArgs/CollectionUpdateArgs to use RetryableWriteImageRecordingType enum for image recording settings.
• SERVER-60779 Extend CheckReplDBHash to work with TenantMigrationsFixture and ClusterToClusterFixture
• SERVER-60788 merge_causes_infinite_loop.js attempts to expose a problem that no longer exists
• SERVER-60809 Add ability to not idLookup after $search
• SERVER-60820 Hang analyzer fails to run in Evergreen
• SERVER-61012 Instantiation of TEMPLATE rule sometimes has a command in generated ninja
• SERVER-61119 LDAP pooled timeout test expects deterministic behavior from connection pool
• SERVER-61188 When storeImageInSideCollection=true, pre-image noop entries for collection with preImageRecordingEnabledForCollection=true are assigned wrong opTimes
• SERVER-61194 Prevent time-series bucket OID reuse with coarse granularity
• SERVER-61358 validate cachedir incorrectly re-raising InvalidChecksum
• SERVER-61427 Unique index builds can cause a loss of availability during commit due to checking many false duplicates
• SERVER-61479 Increase the number of retries to connect to a replica set following a stepdown
• SERVER-61489 Revert the executor_with_mock_net_stress_test test
• SERVER-61550 Modify auto_workload_path in perf.yml to be relative to cwd
• SERVER-61591 Robustify currentop_shell.js test
• SERVER-61602 timeseries_min_max.js assumes measurements are returned in insertion order
• SERVER-61650 Disconnect nodes before restarting replication in sync_source_selection_ignores_minvalid_after_rollback.js
• SERVER-61681 Wait for replSetGetStatus to update in replSetGetStatus_member_wall_times.js
• SERVER-61690 Adjust stored fields protocol for atlas search
• SERVER-61743 The -fno-builtin-memcmp flag shouldn't be applied except on x86_64 platforms
• SERVER-61764 disable SELinx tests on ARM
• SERVER-61791 pin pymongo
• SERVER-61839 $elemMatch w/ positional $ project SBE bug
• SERVER-61858 Wait for node to become primary before disconnecting in sync_source_selection_ignores_minvalid_after_rollback.js
• SERVER-61910 initiate_takes_stable_checkpoint.js doesn't account for closing connections in rollback
• SERVER-61931 Allow ClusterManager role to operate against system.buckets.* collections
• SERVER-62003 Null-equality on 'fullDocument' is not rewritten correctly for change stream 'delete' events
• SERVER-62037 Fix replica set definition in the system perf yaml file for linux-1-node-15gbwtcache
• SERVER-62113 Ignore IndexBuildAborted errors when building indexes in find_flip_sbe_enabled.js
• SERVER-62164 Remove several obsoleted build variants across all stable branches
• SERVER-62243 Wait for vector clock document majority-commit without timeout
• WT-8241 Skip value return for largest key

Platform Support

Starting in MongoDB 5.1.2 the following platforms are no longer supported.

Sharding

• SERVER-51329 Unexpected non-retryable error when shutting down a mongos server
• SERVER-55382 Ignore error from retryable write being converted to transaction and that transaction failing in random_moveChunk_update_shard_key.js
• SERVER-57686 We need test coverage that runs resharding in the face of elections
• SERVER-58343 Re-enable reshard_collection_failover_shutdown_basic.js
• SERVER-59719 shardsvr{Commit, Abort}ReshardCollection may return unrecoverable error on stepdown, leading to fassert() on config server
• SERVER-59806 Optimized path of shard collection fails to succeed with a large number of chunks
• SERVER-60730 shardsvrDropDatabase should always join existing coordinator
• SERVER-60751 move_chunk_critical_section_non_internal_client_abort.js does not consider config server stepdowns
• SERVER-60945 Increase the resharding critical section timeout value for resharding_large_number_of_initial_chunks.js
• SERVER-61027 Port permitMigrations changes
• SERVER-61186 Remove the ReshardingTest fixture's isMixedVersionCluster() method
• SERVER-61289 Make resharding_retryable_writes.js more robust to timing
• SERVER-61473 Resharding coordinator calls ReshardingMetrics::onCompletion() multiple times on transient errors, leading to config server crash
• SERVER-61482 Updates to config.reshardingOperations wait for PrimaryOnlyService to be rebuilt while holding oplog slot, stalling replication on config server indefinitely
• SERVER-61483 Resharding coordinator fails to recover abort decision on step-up, attempts to commit operation as success, leading to data inconsistency
• SERVER-61607 Accept DuplicateKey as a possible error in resharding_nonblocking_coordinator_rebuild.js
• SERVER-61633 Resharding's RecipientStateMachine doesn't join thread pool for ReshardingOplogFetcher, leading to server crash at shutdown

Replication

• SERVER-54909 In replSetGetStatus, report last durable and last applied operation wall times for all members
• SERVER-59721 Node may become unable to sync from other members after performing rollback to stable timestamp
• SERVER-60946 Race condition in replsetprio1.js when initiating replset with nodes having different priorities
• SERVER-60969 ReplClientInfo::getLastOp may not be updated before being used to wait for replication in tenant migration
• SERVER-61440 Race in tenant_migration_recipient_current_op.js

Aggregation

SERVER-59924 Error executing aggregate with $out with "available" read concern on sharded clusters

Storage

SERVER-58736 Avoid quadratic behavior in rollback with many collections

Internals

• SERVER-55535 Performance tests to exercise change streams optimizations
• SERVER-59297 Allow system threads to survive InterruptedDueToStorageChange
• SERVER-59871 startup_recovery_for_restore_restarts.js needs to make sure a checkpoint happens after failpoint
• SERVER-60393 increase replica_sets_jscore_passthrough time limit to 3hr.
• SERVER-60567 Fix SBE command's handling of collection UUIDs
• SERVER-60616 Coverity analysis defect 120880: Unsigned compared against 0
• SERVER-60632 Inconsistent error on mongos when renameCollection target already exists
• SERVER-60671 Remove dagger
• SERVER-60729 Explicitly step up first node in initiate_takes_stable_checkpoint.js
• SERVER-60756 Add additional logging for failed updates in multi_statement_transaction_atomicity_isolation.js
• SERVER-61021 Documents may be retrieved out of order in timeseries_delete.js
• SERVER-61039 Sharded time-series listIndexes should report the view's namespace
• SERVER-61164 Accept error code 48 (Inappropriate authentication) as a valid response for LDAP liveness check
• SERVER-61178 Use less concurrency in find_cmd_with_indexes_timeseries.js
• SERVER-61208 Balancer might be running when it shouldn't in transactions_stale_shard_version_errors.js
• SERVER-61238 Resmoke hook's thread logs not being shown
• SERVER-61269 Adding log to awaitdata_getmore_cmd.js
• SERVER-61270 Coverity analysis defect 121083: Macro compares unsigned to 0
• SERVER-61291 Fix forward compatibility with checkout_idl_files_from_past_releases.py
• SERVER-61295 Complete TODO listed in SERVER-35721
• SERVER-61296 Complete TODO listed in SERVER-26792
• SERVER-61304 Unable to use --use-system-{pcre,boost,...} with r5.1.0
• SERVER-61308 CappedPositionLost if capped collection wraps around during tenant migration
• SERVER-61309 Fix time-series bucket lock reacquisition logic
• SERVER-61353 MongoDB 5.1.0 displays development version start-up warning
• SERVER-61360 reduce Debian 9 variants jlink to prevent OOM cases
• SERVER-61405 Consider removing FCV check from $$SEARCH_META
• SERVER-61489 Revert the executor_with_mock_net_stress_test test
• SERVER-61496 [SBE] buildSortMerge() does not account for nested MERGE_SORT stages
• SERVER-61532 Opcounters to detect constrain violations need to be exposed.
• SERVER-61597 Use deterministic order for random_moveChunk_timeseries_inserts.js