Skip to content

Commit b410301

Browse files
committed
DOCS-13486: 4.2.4 upcoming
1 parent b076fd8 commit b410301

File tree

3 files changed

+223
-0
lines changed

3 files changed

+223
-0
lines changed
Lines changed: 201 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,201 @@
1+
.. _4.2.4-changelog:
2+
3+
4.2.4 Changelog
4+
---------------
5+
6+
Sharding
7+
~~~~~~~~
8+
9+
- :issue:`SERVER-42617` Race in CloneDocumentsCatchesInsertErrors can causes it to return an unexpected error
10+
- :issue:`SERVER-44103` clear_jumbo.js should wait for balancer to be on mode: full for more than one round
11+
- :issue:`SERVER-44130` Flip order of assert.lte arguments in logical_time_metadata.js
12+
- :issue:`SERVER-44839` Frequent schema changes in mongos ftdc metrics limits retention period
13+
- :issue:`SERVER-44915` Extend $indexStats output to include full index options and shard name
14+
- :issue:`SERVER-45273` Remove the mongos version check in allow_partial_results.js and return_partial_shards_down.js
15+
- :issue:`SERVER-46001` Move checkShardingIndex outside shardCollection's read critical section on the primary shard
16+
- :issue:`SERVER-46121` mongos crashes with invariant error after changing taskExecutorPoolSize
17+
18+
Replication
19+
~~~~~~~~~~~
20+
21+
- :issue:`SERVER-34768` Rollback can fail if run against a lagged node that catches up
22+
- :issue:`SERVER-35050` Don't abort collection clone due to negative document count
23+
- :issue:`SERVER-38028` Participant with prepared transaction on session should block request for higher txn number on session rather than failing the new request
24+
- :issue:`SERVER-39112` Primary drain mode can be unnecessarily slow
25+
- :issue:`SERVER-43867` Work around unrecoverability of rollback via refetch in tests
26+
- :issue:`SERVER-44260` Transaction can conflict with previous transaction on the session if the all committed point is held back
27+
- :issue:`SERVER-45010` Clean shutdown after rollbackViaRefetch with eMRC=false can cause us to incorrectly overwrite unstable checkpoints
28+
- :issue:`SERVER-45178` Rollback via refetch can cause rollback to happen successfully w/o updating the rollback id.
29+
- :issue:`SERVER-45421` Fix transactions_block_ddl.js to use write concern "majority" for commands run as part of setup phase.
30+
- :issue:`SERVER-45492` Tag rollback_dup_ids.js with 'requires_persistence' and 'requires_journaling'
31+
- :issue:`SERVER-45493` temporarily disable failing assertion in election_candidate_and_participant_metrics.js
32+
- :issue:`SERVER-45612` Remove the mapReduce + prepare testing in 4.2 concurrency_simultaneous_replication suite
33+
- :issue:`SERVER-45839` recoverFromOplogUpTo should not invariant if there are no oplog entries to apply between the given range
34+
- :issue:`SERVER-45840` Blacklist tests that run emptycapped from replica_sets_kill_secondaries_jscore_passthrough
35+
- :issue:`SERVER-45842` Remove assertion that checks whether the last applied oplog entry has the same timestamp as the requested recover timestamp in recoverFromOplogUpTo
36+
- :issue:`SERVER-45906` Initial stable checkpoint not triggered properly when enableMajorityReadConcern=false
37+
- :issue:`SERVER-46050` Use getLastAppliedOpTime rather than getHeartbeatAppliedOpTime for checking primary's position
38+
- :issue:`SERVER-46188` Blacklist write_conflicts_with_non_txns.js from sharded transaction passthrough suites due to maxTimeMS incompatibility
39+
- :issue:`SERVER-46218` Race between removal and shutdown in arbiter
40+
41+
Query
42+
~~~~~
43+
44+
- :issue:`SERVER-32903` Ambiguous field name error should be ignored during initial sync
45+
- :issue:`SERVER-45279` GranularityRounder can get stuck in loop when rounding infinity
46+
- :issue:`SERVER-45363` Issue with mongodb text indexes and weights when using wildcard specifier
47+
- :issue:`SERVER-45927` Alias for Atlas FTS
48+
49+
Write Operations
50+
~~~~~~~~~~~~~~~~
51+
52+
:issue:`SERVER-45611` Lazily enforce that persisted collection validators are well formed
53+
54+
Aggregation
55+
~~~~~~~~~~~
56+
57+
- :issue:`SERVER-40603` Perform more thorough review/audit of new mongohouse DocumentSources
58+
- :issue:`SERVER-44942` $out will hit invariant if temp collection is dropped before indexes are copied from source collection
59+
- :issue:`SERVER-45418` DocumentSourceCursor batching memory accounting does not account for empty documents, leads to unbounded memory use for count-like aggregates
60+
61+
Catalog
62+
~~~~~~~
63+
64+
:issue:`SERVER-45137` Increasing memory allocation in Top::record with high rate of collection creates and drops
65+
66+
Storage
67+
~~~~~~~
68+
69+
- :issue:`SERVER-41968` IndexBuildTest.getIndexBuildOpId() not selective enough when IndexBuildsCoordinator is enabled
70+
- :issue:`SERVER-42830` Applying a rename operation can result in multiple WT transactions
71+
- :issue:`SERVER-43794` Change MongoDB backup cursor API to include offset/length ranges
72+
- :issue:`SERVER-44370` Have openBackupCursor accept inputs for incremental backup requests
73+
- :issue:`SERVER-44406` Add $backupCursor API to forcefully disable incremental backups
74+
- :issue:`SERVER-44407` Connect the incremental backup cursor pipes with WT output.
75+
- :issue:`SERVER-44410` Change backup cursor results to include a filesize
76+
- :issue:`SERVER-44426` Have queryable backup mode forward FS write calls to the backend API
77+
- :issue:`SERVER-44433` Have queryable backup mode forward file opening/creation calls to the backend API
78+
- :issue:`SERVER-44438` Have queryableBackupMode open WT in read/write mode
79+
- :issue:`SERVER-44442` Allow queryableBackupMode to perform replication recovery
80+
- :issue:`SERVER-45006` LockerImpl::wasGlobalLockTaken() always returns true
81+
- :issue:`SERVER-45007` PBWM acquisition in GlobalLock constructor ignores deadline
82+
- :issue:`SERVER-45288` switch Idempotency tests to use WiredTiger instead of ephemeralForTest
83+
- :issue:`SERVER-45289` Conditional jump or move depends on uninitialised value(s) in validate_adaptor.cpp
84+
- :issue:`SERVER-45374` Unique index that gets rebuilt as part of startup recovery/repair is using data format version as 6 or 8 and NOT 11 or 12.
85+
- :issue:`SERVER-45481` Change the backup API to return the blocks to copy for incremental backup
86+
- :issue:`SERVER-45581` Change the incremental backup API to allow block size granularity to be specified
87+
- :issue:`SERVER-45660` Remove blockSize from queryable backup
88+
- :issue:`SERVER-45663` Implement file renaming for queryableBackupMode and fix file size tracking
89+
- :issue:`SERVER-45821` Re-enable disabled incremental backup cursor tests and add additional tests
90+
- :issue:`SERVER-46010` Incremental backups should remain viable across replication rollback on the backup source
91+
- :issue:`SERVER-46366` Append the backup cursors user arguments to the metadata document
92+
- :issue:`SERVER-46375` Include the offset, length pair for files as part of the first full backup for incremental
93+
- :issue:`SERVER-46488` queryableBackupMode should recover from the journal files on startup using the recovery timestamp
94+
- :issue:`SERVER-46565` Extend ESE backup tests to use $extendBackupCursor
95+
96+
Operations
97+
~~~~~~~~~~
98+
99+
:issue:`SERVER-45662` Flow Control currentOp timeAcquiringMicros stat is no longer updated
100+
101+
Build and Packaging
102+
~~~~~~~~~~~~~~~~~~~
103+
104+
- :issue:`SERVER-45713` Run rhel7 push and publish tasks on large rhel70 distro
105+
- :issue:`SERVER-45732` Filter flags from net-snmp-config invocation more aggressively
106+
- :issue:`SERVER-45879` No Builds Produced for s390x Ubuntu 18.04 for 4.2.3
107+
- :issue:`SERVER-45922` Build failure in overflow_arithmetic.h: constexpr function never produces a constant expression
108+
109+
Tools
110+
~~~~~
111+
112+
:issue:`TOOLS-2430` mongorestore: in dotted index keys, replace "hashed" with "1"
113+
114+
Internals
115+
~~~~~~~~~
116+
117+
- :issue:`SERVER-37148` Remove mr_shard_version.js and mr_during_migrate.js in favor of MR FSM test
118+
- :issue:`SERVER-41872` PlanEnumerator AndAssignment::choices ordering not stable and is relevant to set of plans generated
119+
- :issue:`SERVER-42930` ConnectionPool controller updates must batch across hosts
120+
- :issue:`SERVER-43308` Invariant failure in mqlrun: doc.value().isOwned()
121+
- :issue:`SERVER-43339` PBWM locks acquired part of GlobalLock::GlobalLock() should be interruptible. Else, it won't respect MaxTimeMS.
122+
- :issue:`SERVER-43470` Complete TODO listed in SERVER-35156
123+
- :issue:`SERVER-43726` Make ssl_alert_reporting.js tolerate LEGACY crypto policy on RHEL8
124+
- :issue:`SERVER-43763` Figure out if global LDAP synchronization can be disabled when libldap is built with OpenSSL.
125+
- :issue:`SERVER-43971` MQLRUN crashes if pipeline is specified in file
126+
- :issue:`SERVER-44325` Add polyfill for certificate chain verification
127+
- :issue:`SERVER-44435` Allow x509 authorization to be selectively enabled based on the CA
128+
- :issue:`SERVER-44620` Fix sslProviderSupports helpers on Debian 10
129+
- :issue:`SERVER-44853` MSI Should skip install MongoD service dialog when mongod is not selected to be installed
130+
- :issue:`SERVER-44904` Startup recovery should not delete corrupt documents while rebuilding unfinished indexes
131+
- :issue:`SERVER-45261` create more state in NiTL ctor
132+
- :issue:`SERVER-45486` Add information to debug index key count mismatch issue in 'hashed_index_bad_keys_cleanup.js'
133+
- :issue:`SERVER-45544` burn_in_tests for certain tests can time out regardless of what changed
134+
- :issue:`SERVER-45546` Do not create HostPools for passive members
135+
- :issue:`SERVER-45644` Reevaluate timeouts used by burn_in_test
136+
- :issue:`SERVER-45726` Allow empty firstBatch with TaskExecutorCursor
137+
- :issue:`SERVER-45748` burn_in_tags_bypass_compile is not looking at the correct task
138+
- :issue:`SERVER-45764` Generate resmoke tasks need to take setup time into account when setting timeouts
139+
- :issue:`SERVER-45766` Remove "requires" from the server yaml
140+
- :issue:`SERVER-45825` Disable tool_replset for ephemeralForTest storage engine
141+
- :issue:`SERVER-45867` Use latest version of signal processing
142+
- :issue:`SERVER-46003` Send "search" internally from mongod to mongot instead of "searchBeta"
143+
- :issue:`SERVER-46082` Use bin/analysis.py for performance projects
144+
- :issue:`SERVER-46174` Free peer certificate in SSL_get0_verified_chain polyfill
145+
- :issue:`SERVER-46197` Make build flag to disable diagnostic latches
146+
- :issue:`TOOLS-2461` Backport mongorestore hashed index fix to 4.2
147+
- :issue:`WT-4886` Avoid depending on particular error message text for test_bug018
148+
- :issue:`WT-4968` Remove all_committed timestamp
149+
- :issue:`WT-4999` Migrate Jenkins “wiredtiger-test-format-stress-zseries” job to Evergreen
150+
- :issue:`WT-5003` Migrate Jenkins “wiredtiger-test-race-condition-stress-sanitizer” job to Evergreen
151+
- :issue:`WT-5024` Migrate Jenkins “wiredtiger-compatibility” job to Evergreen
152+
- :issue:`WT-5041` Distribute modifications properly in test_calc_modify.py
153+
- :issue:`WT-5081` Add support for ASan builds in the Python test suite
154+
- :issue:`WT-5119` Birthmark records can be read as normal updates if reads race with checkpoints
155+
- :issue:`WT-5159` Make wiredtiger work with SWIG 4.0.0 beyond
156+
- :issue:`WT-5165` Add cursor copy debug mode
157+
- :issue:`WT-5192` Don't allow checkpoints to evict without a snapshot
158+
- :issue:`WT-5199` Adding correctness tasks into rhel80 Evergreen build variant
159+
- :issue:`WT-5206` Return the correct checkpoint-modified list of blocks
160+
- :issue:`WT-5235` Craft a workload to show lookaside contention
161+
- :issue:`WT-5255` Turn off external programs if direct I/O is configured for test/format
162+
- :issue:`WT-5261` coverage-report task timed out in Evergreen
163+
- :issue:`WT-5334` Add testing coverage for static wt build
164+
- :issue:`WT-5366` read-committed and read-uncommitted transactions can stall eviction
165+
- :issue:`WT-5371` Fix test_stat08.py assertion failure
166+
- :issue:`WT-5372` Skip known errors for long-running format stress sanitizer tasks
167+
- :issue:`WT-5376` WT_UPDATE.type field can race with visibility checks when returning key/value pairs
168+
- :issue:`WT-5377` variable-length column store insert locking in format test program can stall eviction
169+
- :issue:`WT-5379` Missing varargs cleanup
170+
- :issue:`WT-5380` Stat clear is not clearing the proper lookaside cursor stats
171+
- :issue:`WT-5383` Update WiredTiger source code to include 2020 copyright notices
172+
- :issue:`WT-5387` Prepared transaction resolution can stall eviction on active pages
173+
- :issue:`WT-5393` Prepared transaction rollback and API error handling fixes
174+
- :issue:`WT-5395` Fix a bug in the read lock implementation when there is a lot of contention
175+
- :issue:`WT-5398` Adding more correctness tasks into rhel80 Evergreen build variant
176+
- :issue:`WT-5405` Make format LSM test a separate Evergreen task
177+
- :issue:`WT-5410` WiredTiger formatted timestamp string buffer size is too small
178+
- :issue:`WT-5437` Salvage's excessive consumption of cache memory causing eviction to stall
179+
- :issue:`WT-5440` __wt_txn_clear_read_timestamp() has an unnecessary serialization point
180+
- :issue:`WT-5442` Fix failure in test_shared_cache01.py
181+
- :issue:`WT-5443` Disable PPC format tests in Evergreen
182+
- :issue:`WT-5444` Re-enable PPC format tests in Evergreen
183+
- :issue:`WT-5445` Handle relative "home" directories in format.sh
184+
- :issue:`WT-5449` Increase contention in history store stress workload
185+
- :issue:`WT-5450` Support timeout setting for long-running operations in test/format
186+
- :issue:`WT-5458` Fix Evergreen timeout failures in linux-directio test
187+
- :issue:`WT-5460` Buffer alignment failure captured by linux-directio test
188+
- :issue:`WT-5468` Improve documentation for "wt load"
189+
- :issue:`WT-5480` Don't take threads resolving prepared transactions to assist with eviction
190+
- :issue:`WT-5488` Dump the failing CONFIG for Evergreen test/format tasks
191+
- :issue:`WT-5521` Cache stuck during format initial load, configured with library checkpoints
192+
- :issue:`WT-5534` Incremental backup needs to accept older metadata
193+
- :issue:`WT-5536` Add zstd as a builtin compressor for WiredTiger Python package
194+
- :issue:`WT-5537` Use correct WT_ITEM fields per memory sanitizer
195+
- :issue:`WT-5538` Change format to ignore Evergreen timestamps in configuration files
196+
- :issue:`WT-5553` Return error if primary cursor is not an incremental source
197+
- :issue:`WT-5554` Account for logpath when retrieving file size in incremental backup
198+
- :issue:`WT-5564` Buffer not extended properly when incremental bitstring grows
199+
- :issue:`WT-5584` Change format to ignore Evergreen timestamps in configuration files
200+
- :issue:`WT-5587` Limit how many checkpoints are dropped by a subsequent checkpoint
201+

source/release-notes/4.2-changelog.txt

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -10,6 +10,8 @@
1010
:depth: 1
1111
:class: singlecol
1212

13+
.. include:: /includes/changelogs/releases/4.2.4.rst
14+
1315
.. include:: /includes/changelogs/releases/4.2.3.rst
1416

1517
.. include:: /includes/changelogs/releases/4.2.2.rst

source/release-notes/4.2.txt

Lines changed: 20 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -21,6 +21,26 @@ MongoDB 4.2 Released Aug 13, 2019
2121
Minor Releases
2222
--------------
2323

24+
.. _4.2.4-release-notes:
25+
26+
4.2.4-Upcoming
27+
~~~~~~~~~~~~~~
28+
29+
Issues fixed:
30+
31+
- :issue:`SERVER-44915`: Extend $indexStats output to include full index options and shard name
32+
- :issue:`SERVER-46121`: mongos crashes with invariant error after changing taskExecutorPoolSize
33+
- :issue:`SERVER-45137`: Increasing memory allocation in Top::record with high rate of collection creates and drops
34+
- :issue:`SERVER-44904`: Startup recovery should not delete corrupt documents while rebuilding unfinished indexes
35+
- :issue:`SERVER-44260`: Transaction can conflict with previous transaction on the session if the all committed point is held back
36+
- :issue:`SERVER-35050`: Don't abort collection clone due to negative document count
37+
- :issue:`SERVER-39112`: Primary drain mode can be unnecessarily slow
38+
39+
- `All JIRA issues closed in 4.2.4
40+
<https://jira.mongodb.org/issues/?jql=project%20in%20(SERVER%2CTOOLS%2CWT)%20AND%20resolution%3D%27Fixed%27%20and%20fixversion%3D%274.2.4%27>`_
41+
42+
- :ref:`4.2.4-changelog`
43+
2444
.. _4.2.3-release-notes:
2545

2646
4.2.3 - Jan 27, 2020

0 commit comments

Comments
 (0)