Docs HomeMongoDB Manual

6.3 Changelog

On this page本页内容

Changes are listed here when a patch release is available.

6.3.2 Changelog

Sharding

  • SERVER-72146 Make chunk migrations metrics more accessible from Atlas
  • SERVER-74380 Router role retry loop should allow namespaces different from the one it was created for
  • SERVER-75481 Consider ideal data size when balancing
  • SERVER-76004 Remove incorrect sharding tassert in getOwnershipFilter function
  • SERVER-76178 Fix moveRange authorization requirements
  • SERVER-76394 Allow the setAllowMigrations refresh to be sent to all shards in a backportable way
  • SERVER-76536 Increase receiveChunkWaitForRangeDeleterTimeoutMS in concurrency suites

Query

SERVER-72224 Clustered collections with non-"simple" default collation miss documents when querying on _id using "simple" collation

Internals

  • SERVER-67105 $in queries do not use clustered index
  • SERVER-68670 Cache imbalanced collections for balancing
  • SERVER-69952 Filtering on dates outside of the 32 bit epoch range returns wrong results
  • SERVER-71009 Make sure the select chunks operation doesn't take too long
  • SERVER-71985 Automatically retry time series insert on DuplicateKey error
  • SERVER-72520 Decommission Microbenchmark performance tests that are running on CBI instances
  • SERVER-72686 Add support for $collStats agg stage on timeseries collection
  • SERVER-73734 Support evergreen nightly project in burn_in_tests.py
  • SERVER-73830 Retrieving alternate buckets for inserts not handling buckets pending direct writes
  • SERVER-73855 Update recipientForgetMigration error handling logic to properly handle CallbackCanceled
  • SERVER-73887 Allow untimestamped writes for the entirety of initial sync retry
  • SERVER-74122 add readtest.c to the asan denylist
  • SERVER-74368 Reduce log verbosity for virtual collection queries
  • SERVER-74469 Make the ShardingCatalogManager set the 'validAfter' field of a chunk when it is migrated
  • SERVER-74694 cleanupOrphaned command may fail with "Range deleter service not up error" on step up
  • SERVER-74871 Clean up Jepsen tests
  • SERVER-75082 Make balancer migrations throttling configurable
  • SERVER-75230 Package test fails on Ubuntu 16.04
  • SERVER-75269 Re-introduce Classic Engine build variant on the 6.3 branch
  • SERVER-75479 Use system python on macos for ssl tests
  • SERVER-75566 Disable powercycle on windws
  • SERVER-75601 Make serverStatus apiversion field more robust
  • SERVER-75626 Special handling for macosx venv
  • SERVER-75990 Tenant Migrations are not resilient to recipient failover
  • SERVER-76038 Remove FLE2 tests from sharded_retryable_writes_downgrade suite
  • SERVER-76102 handleRIDRangeScan() sets hasCompatibleCollation wrong in recursive cases
  • SERVER-76172 Fix BalancerChunkSelection unit test on slow build variants
  • SERVER-76274 Move microbenchmarks to their own task distro
  • SERVER-76321 Buffer overrun while deserializing compound group key spilled to disk in SBE hash agg implementation
  • SERVER-76331 Bump mongo-task-generator version to 0.7.5
  • SERVER-76457 Support storing BuildID to debug symbols mapping for sanitizer builds
  • SERVER-76498 Update push task to run on rhel8.7-small
  • SERVER-76519 MIgrate crypt_push task to Garasign
  • SERVER-76543 The mongod-dump-sessions gdb pretty printer does not work
  • SERVER-76549 Remove symbolization from macos variants
  • SERVER-76550 Balancer is unable to drain shards with big chunks
  • SERVER-76666 [v6.3, v7.0] Make buildscripts_test run on v6.3 branch
  • SERVER-76690 Special handling for PYTHONPATH when activate venv
  • SERVER-77018 Deadlock between dbStats and 2 index builds
  • SERVER-77380 Using --serverless flag disables oplog truncation
  • SERVER-77383 ".msi.sha256" files have incorrect shasum
  • SERVER-77527 mongodbtoolchain/v3 stdlib unsafe against self-moves
  • SERVER-78080 Use fully random bytes for 8 low-order bytes of OID value for time series bucket _id
  • SERVER-78122 [6.3] Address bucket OID collisions with buckets pending compression

6.3.1 Changelog

Sharding

  • SERVER-73229 Logical sessions cache refresh ignores write errors from updating session document, leading to cursors being killed early
  • SERVER-73877 Wait for shard to have a primary before doing a migration in prepare_transaction_then_migrate.js

Write Operations

SERVER-75517 illegal format of _id possible via upsert

Internals

  • SERVER-66927 Remove reference to silentFail from mongo repo
  • SERVER-71636 Explain executionStats does not work for $lookup on sharded collection
  • SERVER-73163 shard_split_shutdown_while_blocking_reads should handle exception in readThread
  • SERVER-73383 Defragmentation may start over right after finishing
  • SERVER-73482 $natural hint overrides sort on clustered collections
  • SERVER-73936 Clean up timeseries_direct_update.js and add logs
  • SERVER-73973 [test-only bug] Skip orphans checking in server_status_with_time_out_cursors.js
  • SERVER-74131 Multiplanning together with subplanning causes server crash in mapReduce queries
  • SERVER-74303 Add a sizeStorer call to the onRollback handler for dataSize and numRecords changes
  • SERVER-74684 Size storer not being flushed periodically
  • SERVER-74720 The default 'shardingStatistics' serverStatus section takes locks
  • SERVER-74895 Expect command errors in validation when querying $collStats, and improve the error message
  • SERVER-75066 Check for parent directory before writing to SCons cache
  • SERVER-75090 ftdc_connection_reuse.js -- test fails when poolConnUsageTimeMillis does not increase
  • SERVER-75147 Remove FLE2 tests from replset continuous stepdowns suite
  • SERVER-75239 [v6.3] Remove accidentally added tests from backports_required_for_multiversion_tests.yml
  • SERVER-75261 "listCollections" command fails with BSONObjectTooLarge error
  • SERVER-75277 TTL deleter attributes resource metrics across multiple databases
  • SERVER-75299 Reduce severity of BalancerStatsRegistry logs
  • SERVER-75366 Additional debug log lines in range deleter service
  • SERVER-75431 Get rid or fix best effort check for primary db on rename path in sharded clusters
  • SERVER-75501 Don't run Sys-perf fCV variants on stable branches
  • SERVER-75519 Split task sharded_collections_jscore_passthrough due to timeout
  • SERVER-75561 Report and log more detailed information when validate encounters multikey inconsistencies
  • SERVER-75639 Remove dedicated executor for moveChunk/moveRange commands
  • SERVER-75652 Migrate from notary client to garasign for push task
  • SERVER-75745 Only log record and index metadata associated with missing or extra index keys
  • SERVER-75795 Update aws_e2e_ec2.js for evergreen changes to AWS test identity
  • SERVER-76009 Fix stepdown behavior in read_ticket_exhaustion_with_stepdown test