Known Issues
5166 tracked MongoDB server issues from Jira with status, affected versions, and resolution details.
SERVER-139403Needs Scheduling
Complete TODO listed in SERVER-136468
There is a TODO in code owned by your team which references a resolved ticket.
The TODO appears in:
* buildscripts/modules/atlas/matrix_suites/overrides/no_passthrough_disagg_override_replsets.yml:130
You can also search for all references to the resolved ticket here:
https://github.com/10gen/mongo/search?q=SERVER-136468&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The next steps for this ticket are to either remove the outdated TODO or follow the steps in the TODO if it is correct.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
GODRIVER-4266Needs Triage
mongo-go-driver - PR #2721: Bump go.opentelemetry.io/otel/sdk from 1.45.0 to 1.47.0 in /internal/test/prose
[dependabot[bot]|https://github.com/apps/dependabot] has created [PR #2721: Bump go.opentelemetry.io/otel/sdk from 1.45.0 to 1.47.0 in /internal/test/prose|https://github.com/mongodb/mongo-go-driver/pull/2721] in [mongo-go-driver|https://github.com/mongodb/mongo-go-driver]
GODRIVER-4265Needs Triage
mongo-go-driver - PR #2720: Bump go.opentelemetry.io/otel/sdk from 1.45.0 to 1.47.0 in /internal/test/compilecheck
[dependabot[bot]|https://github.com/apps/dependabot] has created [PR #2720: Bump go.opentelemetry.io/otel/sdk from 1.45.0 to 1.47.0 in /internal/test/compilecheck|https://github.com/mongodb/mongo-go-driver/pull/2720] in [mongo-go-driver|https://github.com/mongodb/mongo-go-driver]
SERVER-74926ClosedWon't Do
Add interleave support to BSONColumn::containsCompressed
Current implementation only supports scalars. Implement object and array support.
SERVER-139285ClosedFixed
Handle majority snapshot loss in side write drains before voting for commit readiness
Majority snapshot loss during side write draining should not immediately abort an index build or prevent it from processing a commit decision.
Before voting for commit readiness, we should retry draining with a bounded wait for snapshot recovery.
Components
Index Builds
Fixed Versions
9.1.0-rc09.1.0-rc1026
SERVER-65804ClosedFixed
Enhance BSONColumn to allow checking contained element types without decompressing
Occasionally we just need to know information about the types of elements in a compressed BSONColumn (e.g. does the column contain an array?). We should be able to provide this information in some cases without having to decompress and iterate over the elements of the column. This should provide a clear performance benefit.
Fixed Versions
7.0.0-rc0
SERVER-131816In Code Review
Parse and apply container writes with multiple int-keys
Following on the work in SERVER-131746 and SERVER-131747, we should also support "ci"/"cu"/"cd" oplog entries with multiple integer keys, in order to have reasonable parity between the data types supported. This may not be used by primary-driven index builds, but the oplog format should be general in its support for batched operations.
Fixed Versions
9.1.0-rc09.1.0-rc1026
SERVER-134134In Code Review
Emit histogram bucket sizes for networking reactor thread run/wait times in serverStatus/otel
Previously, SERVER-119617 added a moving average for networking reactor run/wait times. For more granular data, we should emit each histogram bucket size into serverStatus so that it can be ingested for dashboarding. We may also be able to use otel histograms, but we'll need to verify that it can be ingested to grafana.
GODRIVER-3600Backlog
Join all errors that happen during retries
h3. *Context*
Currently, there's a [bunch of logic|https://github.com/mongodb/mongo-go-driver/blob/6598290d5e92edc6de9428c4bf109decdd33ba50/x/mongo/driver/operation.go#L579-L597] that tries to pick which error to return after a failed retry. That logic frequently discards important error information and makes it difficult or impossible to troubleshoot problems without a bunch of additional event monitoring or logging enabled. It would be much more useful to collect all error information for all
SERVER-138732Backlog
Triage tests in jstests/noPassthrough that fail in disaggregated storage
SERVER-136468 included all of jstests/noPassthrough in the no_passthrough_disagg_override_replsets suite. However, many failing tests were put in an untriaged exclusion list.
This ticket should classify those failures by root cause, and open follow up tickets to fix those tests (or definitively exclude them if they test ASC exclusive functionality).
SERVER-139400In Progress
Deliver the invalidate event after a converted failure
The problem: for change streams opted into showErrorsAsEvents, a user stage failing while processing the invalidate event swallowed the invalidate. The client never saw it and never learned the stream had ended — it kept waiting on a silently dead stream (zombie).
The fix: the failure still converts to an error event (errors keep flowing), and the invalidate event is re-delivered right after — so the client sees the stream end and closes cleanly.
PYTHON-6187In Code Review
a cancelled async command gets no CommandFailedEvent
*Summary:* No CommandFailedEvent (and no "Command failed" log message) when an async operation is cancelled
*Description*
When an asyncio task is cancelled while a command waits for its reply — {{{}task.cancel(){}}}, {{asyncio.timeout()}} / {{{}asyncio.wait_for(){}}}, a failing TaskGroup sibling, or an ASGI server cancelling a disconnected client's request — PyMongo publishes the {{CommandStartedEvent}} but never a correlating {{CommandSucceededEvent}} or {{{}CommandFailedEvent{}}}. The st
SERVER-131747In Code Review
Parse and apply container updates with multiple kvs
Similar to SERVER-131746, we should extend support to do the same thing for container updates - so that multiple key/value pairs can be updated in a "cu" operation at once. While this is not used by primary-driven index builds, it is relevant to keep parity with the API.
PYTHON-6274In Code Review
Enable ruff rule PLR2004
h3. *Context*
{color:#505f79}Describe the background behind the problem.{color}
Ruff's [PLR2004|httphttps://docs.astral.sh/ruff/rules/too-many-branches/] linting rule forbids the use of magic numbers. Our coding standards specify not to use magic numbers.
h3. *Definition of done*
{color:#505f79}What must be done to consider the task complete?{color}
Enable the PLR2004 linting rule and make the code changes required to pass it.
h3. *Pitfalls*
{color:#505f79}What should the implementer
Components
Infrastructure
SERVER-137827Open
Add antithesis system testing for socket handoff with Client, Monguard, and Server in hostile network environment
PYTHON-6173ClosedDuplicate
Release Workflow post-publish step failure
h3. *Context*
{color:#505f79}Describe the background behind the problem.{color}
{color:#c1c7d0}post-publish step fails to set the new version: [https://github.com/mongodb/mongo-python-driver/actions/runs/38623825974/job/116067722000]{color}
h3. *Definition of done*
{color:#505f79}What must be done to consider the task complete?{color}
{color:#c1c7d0}fix the failure so that new versions are set automatically again{color}
h3. *Pitfalls*
{color:#505f79}What should the implementer w
NODE-7887ClosedFixed
OIDC ENVIRONMENT confusion bypasses ALLOWED_HOSTS and leaks human-callback bearer tokens
h3. Use Case
As a... Node Driver developer whose application supplies OIDC_HUMAN_CALLBACK and ALLOWED_HOSTS programmatically but consumes the conn string from a less-trusted source
I want... the driver to enforce ALLOWED_HOSTS no matter which auth mechanism properties the conn string adds
So that... the user's OIDC token is not sent to a host outside ALLOWED_HOSTS
h3. User Experience
* After the fix, MongoClient constructor should throw if both ENVIRONMENT and OIDC_CALLBACK OR OIDC_HUMAN
Components
Authentication
Fixed Versions
7.9.0
SERVER-139250Needs Scheduling
Add ThreadPool pause/resume and start workers for queued tasks
h3. Current behavior
* If every task is already queued, raising {{maxThreads}} does not start another worker.
* There is no way to stop the pool from running tasks already in its queue. {{join()}} runs whatever is left ({{{}_drainPendingTasks{}}}).
h3. Requested behavior
* A pause method that stops workers from pulling tasks from the pending queue. A task already running finishes, and tasks submitted while paused stay queued. A resume method that continues them in FIFO order. Lowering
SERVER-135985ClosedFixed
Implement aggregation pipeline logical operators
Similar to SERVER-134934, but we should now implement narrowType for the aggregation expression versions.
Fixed Versions
9.1.0-rc09.1.0-rc1026
SERVER-135299Open
Node should not vote during STARTUP2
A data bearing node usually votes yes only if the candidate is ahead of the voter. During Initial Sync, a node cannot know if the candidate is ahead or not, so it should not be able to vote. This would mimic the behavior of a “newlyAdded” node to the replset (node added using replSetReconfig).
We first need to verify this change would not have unintended consequences with TLA+ (should be done as part of SERVER-136912). This should be behind a feature flag.
AI taken from WRITING-39616 and W