Known Issues
5279 tracked MongoDB server issues from Jira with status, affected versions, and resolution details.
SERVER-138920ClosedFixed
Complete TODO listed in SERVER-136251
There is a TODO in code owned by your team which references a resolved ticket.
The TODO appears in:
* buildscripts/modules/atlas/suites/disagg_format_snapshot.yml:81
You can also search for all references to the resolved ticket here:
https://github.com/10gen/mongo/search?q=SERVER-136251&type=Code
The next steps for this ticket are to either remove the outdated TODO or follow the steps in the TODO if it is correct. If the latter, please update the summary and description of this ticket to repr
Fixed Versions
9.1.0-rc1025
PYTHON-5964ClosedDone
Adopt ruff 0.17 default rule set in addition to existing rule sets
h3. Context
{color:#505f79}[Ruff 0.16|https://astral.sh/blog/ruff-v0.16.0] enables 413 rules by default, up from 59 in previous versions. We should enable those rules where appropriate while preserving our existing rule set, especially including the "S" security rule set. Full details and verification: https://gist.github.com/blink1073/05b51a7dc689593cd3baa1a67ba083b1{color}
h3. Definition of done
{color:#505f79}
- Ruff pinned to 0.16.0; config updated so the new defaults are adopted wit
Components
Infrastructure
SERVER-139113Backlog
Exclude non-interruptible bulk frees during query execution from overdue interrupt-check accounting
h2. Problem
[SERVER-137477|https://jira.mongodb.org/browse/SERVER-137477] excludes SBE root {{close()}} and aggregation-stage {{dispose()}} from overdue interrupt-check accounting. Large frees also occur during execution. An interrupt check inside a container's bulk destruction cannot avoid completing that destruction, so attributing only that span to missed interrupt checks conflates required cleanup with interruptible execution. Do not exempt the surrounding spill, merge, or phase-transition
SERVER-99002Backlog
$collStats collection scan counter does not account for repeated scans done by SBE NLJ plans
SBE supports nested loop join (NLJ) plans for executing $lookup queries. Such plans will perform a collection scan of the inner collection for every document read from the $lookup's outer side. Following my example from the attached repro script, imagine a $lookup from an "outer" collection to an "inner" collection on the "key" field, where each collection contains the following documents:
There are 4 documents on the outer side. Therefore, an NLJ plan would scan the inner collection 4 ti
SERVER-138105ClosedFixed
Create the RecordIds groupId consistency marker collection
Add {{local.replset.recordIdsGroupId}}, the per-node consistency marker for set-wide Replicated RecordIds. A node honors oplog rids {{iff}} this collection holds a {{groupId}}.
h2. Details
- Single-document collection: {{\{_id: "recordIdsGroupId", groupId: UUID, since: \{ts: Timestamp, t: NumberLong\}\}}}, where since is the {{foundingOpTime}} (term and timestamp) of the fencing entry that minted the {{groupId}}.
- All document writes are timestamped and share the fencing entry's commit ti
Fixed Versions
9.1.0-rc09.1.0-rc1025
SERVER-138037In Code Review
Add a compact hash agg $push accumulator
h2. Overview
Add a {{$push}} implementation of the compact hash agg accumulator, so that {{CompactHashAggStage}} can run {{$push}} without the per-row array encoding that {{HashAggStage}} uses.
h2. Scope of Work
* {{CompactPushHashAggAccumulator}} with a {{CompactPushState}} of the pushed values and their approximate size.
* Enforce the same per-array size cap as the existing VM {{$push}}, through a shared {{ByteCode::addToArrayCapped()}} helper. Missing values are neither stored nor charged.
*
Fixed Versions
9.1.0-rc09.1.0-rc1025
SERVER-51495Backlog
Reenable reshard_collection_basic.js in sharding_continuous_config_stepdown suite
Reenable this test for the sharding_continuous_config_stepdown once milestone 3 is complete and it is resilient in the face of config stepdowns.
Components
Sharding
SERVER-139109Needs Scheduling
Query knobs for MCV field stats
Modify the query knob to include MCV stats:
- Rename {{internalQueryEnablePersistentNDVStats}} to {{internalQueryEnablePersistentFieldStats}}. The knob should enable/ disable both persisted NDV and MCV together.
The MCV should also be gated by {{featureFlagPersistentStats}} .
- Add {{internalMCVCount}} knob to specify the counter param for the MCV algorithm.
SERVER-122751Open
Nodes Have Leaking Heartbeat Threads on an asymmetrical network partition
{*}Summary{*}:
We observed an edge case in the heartbeat protocol for mongo, where if there are secondaries in a replica set which are in a network topology where we have an asymmetrical partition, meaning node A can reach node B, but node B cannot reach node A, we see a leaking of heartbeat threads.
This leak corresponds to 1 thread per heartbeat interval and over time can start contending on the shared [_mutex|https://github.com/mongodb/mongo/blob/master/src/mongo/db/repl/replication_co
Affected Versions
4.4.97.0.118.0.19
SERVER-119972ClosedFixed
The heuristic to restore extended range timestamps can fail to restore correctly in some cases
During startup, a heuristic determines if a collection requires extended range suppport by checking a key bit in the `_id` field. In some cases, this heuristic will fail to accurately determine the need for this flag.
Fixed Versions
9.0.0-rc0
SERVER-138108Open
Expose RecordIds groupId membership in hello and replSetGetStatus
Expose each node's effective {{groupId}} and {{foundingOpTime}} in {{hello}} and {{replSetGetStatus}} responses. The reported value is the effective membership: on ASC the persisted {{(groupId, foundingOpTime)}} and on DSC the provider-derived {{groupId}} sentinel, which is always present.
h2. Acceptance criteria
- {{hello}} and {{replSetGetStatus}} include the node's ({{groupId}}, {{foundingOpTime}}) when present.
- jstest asserting presence/absence across founding, clears, and initial
SERVER-139107Needs Scheduling
Design and test monguard-openssl to monguard-s2n upgrade path
Monguard-OpenSSL and monguard-s2n currently have some mutually exclusive configuration options (e.g., {{{}opensslCipherConfig{}}}, {{{}opensslCipherSuiteConfig{}}}, {{{}enableHandoff{}}}). As a result, hot restart won't work when upgrading to a monguard-s2n process from monguard-openssl because the latter will not recognize an OpenSSL-specific configuration file.
We should remove any TLS backend-specific guards on the configuration file so that both variants of monguard are able to start up w
SERVER-139088ClosedFixed
Make mochalite tests work with the golden test framework
{{golden_overrides.js}} writes everything passed to {{print}} to the golden file, so mochalite's test report ({{{}✔ ...{}}}, {{{}Test Report Summary{}}}) ends up in it and golden tests can't use mocha style.
Fix: route mochalite's {{{}stdout{}}}/{{{}stderr{}}} through {{globalThis.withoutGoldenCapture}} when it is defined:
Fixed Versions
9.1.0-rc09.1.0-rc1025
SERVER-139059ClosedFixed
Disable replicated size and count for incompatible multiversion suites
Fixed Versions
9.1.0-rc09.1.0-rc1025
PYTHON-5959ClosedFixed
Drop prefixPreview, suffixPreview, and substringPreview APIs
This ticket was split from {{DRIVERS-3548}}, please see that ticket for a detailed description.
Fixed Versions
4.20.0
SERVER-137666In Progress
Create sharded passthrough test for broader endOfTransaction coverage
Create a sharded cluster passthrough test for replica set change streams that automatically enables transactions and that sets the {{showEndOfTransaction: "required"}} change stream option if not already set.
This needs a new override method in {{jstests/libs/override_methods}} for setting the flag, auto-enabling transactions, and filtering out the {{endOfTransaction}} change events from any {{aggregate}} and {{getMore}} responses received in the command override.
Without filtering these event
SERVER-137873Needs Scheduling
Extend analyze command to create MCV statistics
Implement the extension of the analyze command according to the new syntax.
SERVER-139104Backlog
Add checks for LocalReshardingRegistry runtime state in donor validations
SERVER-12869Backlog
Index null values and missing values differently
A document with a missing value are stored *as if* they were
explicitly given the value 'null'. Given:
X and Y look i
Components
Index MaintenanceQuerying
Affected Versions
2.6.0-rc0
JAVA-6365Needs Triage
Add prose tests that a top-level error halts client bulkWrite execution (ordered and unordered)
This ticket was split from {{DRIVERS-3699}}, please see that ticket for a detailed description.
Components
CRUD & Command-Level Operations