Known Issues
5279 tracked MongoDB server issues from Jira with status, affected versions, and resolution details.
SERVER-138855ClosedFixed
Disable change streams fsm tests until duplicate events tolerated
disabling the tests affected by BF-47107 until https://jira.mongodb.org/browse/SERVER-138724 lands
Fixed Versions
9.1.0-rc09.1.0-rc1025
SERVER-137542ClosedFixed
Target each remote cursor's own namespace in ARM killCursors
h2. Summary
Send ARM {{killCursors}} to each remote cursor's own namespace ({{RemoteCursorData::cursorNss}}) rather than the ARM's {{_params.getNss()}}, matching the {{getMore}} requests. This fixes a latent correctness issue for views and {{$lookup}}/{{$unionWith}} sub-pipelines, where the two namespaces can differ.
h2. Context
Part of the {{SERVER-134598}} ARM retry refactor, delivered as a stack of 8 PRs. This is PR 8 of 8, counting from the bottom of the stack.
h2. Details
The implementing
Fixed Versions
9.1.0-rc09.1.0-rc1025
SERVER-138521ClosedFixed
Bump pinned jstestfuzz rev for deterministic fuzzer tasks to pick up SERVER-137767
h2. Overview
Bump the jstestfuzz rev pinned by the _deterministic_ fuzzer tasks from {{ffd9738aca7a2d11bbd334857112f9b81fc45268}} to {{4c568da85d23a3c579ecaff7790a2890460f763c}}, so they pick up the fix for non\-release branches.
h2. Background
The deterministic fuzzer tasks \({{aggregation\_multiversion\_fuzzer\_deterministic}}, {{aggregation\_timeseries\_fuzzer\_deterministic}}, {{aggregation\_optimization\_fuzzer\_deterministic}}, {{update\_fuzzer\_replication\_deterministic}}\) clone {{10ge
Fixed Versions
9.1.0-rc09.1.0-rc1025
SERVER-138985Needs Scheduling
Update Homebrew Formulae for New Major Patch Release (8.0.38)
Update Homebrew Formulae to support new Major Patch Version 8.0.38
SERVER-138984Needs Scheduling
Update Homebrew Formulae for New Major Patch Release (9.0.8)
Update Homebrew Formulae to support new Major Patch Version 9.0.8
SERVER-138983Needs Scheduling
Update Homebrew Formulae for New Major Patch Release (7.0.48)
Update Homebrew Formulae to support new Major Patch Version 7.0.48
PYTHON-6123ClosedDone
Transactions unified test error-labels-blockConnection is flaky due to configuring its failpoint through a client with timeout options
This ticket was split from {{DRIVERS-3659}}, please see that ticket for a detailed description.
Components
Unified Test Format
GODRIVER-4167Needs Triage
Transactions unified test error-labels-blockConnection is flaky due to configuring its failpoint through a client with timeout options
This ticket was split from {{DRIVERS-3659}}, please see that ticket for a detailed description.
Components
Unified Test Format
JAVA-6323Needs Triage
Transactions unified test error-labels-blockConnection is flaky due to configuring its failpoint through a client with timeout options
This ticket was split from {{DRIVERS-3659}}, please see that ticket for a detailed description.
Components
Unified Test Format
SERVER-137297In Code Review
Support interrupts and deadlines in extension commands
Add the command execution context from the design: its {{api.h}} struct and vtable, plus the {{execCtx}} parameter on {{run}}. Implement it in the host on top of {{OperationContext}}, and add the SDK handle. Don't opt in to stepdown interrupts; the built-in search index commands don't either. Add a polling mode to the echo extension.
We should also investigate whether it is better to expose a single host member to handle this, and possibly unify with the agg stage API.
SERVER-136915ClosedFixed
Allow built-in commands to be replaced by extension commands at startup
h2. Overview
Let a built-in command opt in, at registration, to a predicate that says "an extension replaces me". When the predicate is true at startup, the command is not constructed or registered, so an extension-provided command of the same name can be registered instead. No existing command opts in as part of this ticket, so there is no behavior change.
Part of SPM-4830 (Support Commands in the Extensions API). The search index management commands will opt in later, when extension or built-
Fixed Versions
9.1.0-rc09.1.0-rc1025
SERVER-137083ClosedFixed
Handle large requests better in the PALI rate limiter
The PALI rate limiter currently lets requests larger than the base bucket size bypass rate limiting entirely (due to folly token bucket implementation) and this can bypass our configured rates
While investigating of [HELP-101050|https://jira.mongodb.org/browse/HELP-101050] we saw multiple large page writes and append timeouts. Bypassing PALI rate limiting ended up in us producing extra traffic sent to LS and resulted in us being uxcessively rate limited by LS rate limiter
Requests larger t
Fixed Versions
9.1.0-rc0
SERVER-137756Investigating
create a lsn-based segment-aware frontier tracker for use by the mat tracker
Add a helper to {{MaterializedTracker}} ({{disagg_storage/pali/materialized_tracker.h}}) that tracks a single cell's materialized frontier per log segment, keeping each segment's materialized lsn and whether the segment is complete.
The cell's frontier is the lsn in the least incomplete segment, or else the lsn in the greatest complete segment, or else 0.
API:
* {{bool update(segment, lsn, isComplete)}}
* {{uint64_t get() const}}
Wiring it into the tracker's consensus calculation is out of sc
SERVER-138150ClosedDone
Complete TODO listed in SERVER-131939
There is a TODO in code owned by your team which references a resolved ticket.
The TODO appears in:
* src/mongo/db/storage/wiredtiger/wiredtiger_session.h:144
You can also search for all references to the resolved ticket here:
https://github.com/10gen/mongo/search?q=SERVER-131939&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 represent t
SERVER-138498ClosedDuplicate
Complete TODO listed in SERVER-130286
There is a TODO in code owned by your team which references a resolved ticket.
The TODO appears in:
* src/mongo/db/modules/atlas/src/disagg_storage/pali/log_server_connection.cpp:264
You can also search for all references to the resolved ticket here:
https://github.com/10gen/mongo/search?q=SERVER-130286&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 th
SERVER-119212ClosedFixed
Remove 'exportCollection' and 'importCollection' commands
This was an internal command for migrating serverless clusters. There are no serverless clusters anymore and this can be removed.
Fixed Versions
9.1.0-rc09.1.0-rc1025
SERVER-36606ClosedFixed
Remove size limits on BSON audit events
The user limits on BSONObj sizes shouldn't apply in the audit code paths.
Components
Security
Fixed Versions
3.6.114.0.74.1.4
SERVER-92032Needs Scheduling
Chunk and/or Truncate very large audit payloads
It's possible for a bulkWrite or similar operation to overflow the BSON limits of audit writes. Detect this and respond by either:
* Truncating document writes when the limit is exceeded
* Splitting the document payloads across multiple audit entries.
This may mean creating a setParameter to control this behavior.
SERVER-41196ClosedFixed
Mongos invariant failure crash with change streams startAfter
Mongos crashes with this invariant failure:
Attached is [^mongos-python-1720-crash.log] which includes the full mongos log.
The reproduction is fairly simple:
# Start a change stream on collection "x", drop "x", and obtain the invalidate change stream document's resume token
# Start a new change stream passing the previous resume token to startAfter.
# Run a getMore on the new change stream -> crash.
In code:
{code:js}
MongoDB Enterprise mongos> var cs = db.test.watch([{'$match
Affected Versions
4.1.11
Fixed Versions
4.2.0-rc34.3.1