Known Issues
5166 tracked MongoDB server issues from Jira with status, affected versions, and resolution details.
SERVER-138862Backlog
Co-change graph
Build weighted source-to-test links from Git history before the PR base. Include golden-output mappings.
SERVER-139189Open
Test initial sync and recovery components after restarting from clean / unclean shutdown
In SERVER-136961, we found a deterministic crash in FCBIS after the node restarts from an unclean shutdown. This reveals a testing gap, and we should have proper testing of each path that a node can go through after restarting from both clean and unclean shutdown.
h2. Acceptance Criteria
Targeted test that exercises every initial sync and startup recovery code path after restarting from clean or unclean shutdown. Deduplicate other tests if they already execute one of these code paths.
SERVER-139117Open
Concurrent setParameter calls can leave query-memory load shedding stopped or misconfigured
Items 2 and 3 of SERVER-138793. {{ServerParameter::set()}} runs validate, store and {{on_update}} with no lock across them ({{server_parameter_with_storage.h:291-303}}), and {{CmdSet}} takes no lock around it. The two load-shedding knobs rely on that sequence being atomic, and it is not.
*Start/stop lost update.* {{onUpdateQueryMemoryLoadSheddingEnablement}} acts on the {{newPercent}} of its own call, and {{startQueryMemoryRssMonitor}} checks {{sheddingActive()}} before taking {{holder.mutex}}.
SERVER-130332Needs Scheduling
Switch participant term transport from $replData.term to dedicated participantTerm field once 9.0 is last-LTS
h3. Context
SERVER-128128 carries the participant replication term in {{$replData.term}} rather than a dedicated {{participantTerm}} reply field. This was necessary because adding a new top-level generic reply field and backporting it to older branches causes {{IDLUnknownField}} failures in mixed-version clusters — older mongos binaries use {{strict: true}} IDL reply structs, which reject unknown fields as hard errors.
This ticket tracks switching to a dedicated {{participantTerm}} field o
SERVER-138261In Code Review
Remove featureFlagReshardingCloneNoRefresh and featureFlagReshardingNoRefreshApplyingAndBlockingWrites refresh paths
Fixed Versions
9.1.0-rc09.1.0-rc1026
SERVER-139311Open
Standalone mongod 8.0 loses writes acknowledged with {w:"majority", j:true} after SIGKILL (6.0 and 7.0 do not)
On a standalone {{mongod}} 8.0, a write acknowledged with {{\{w:"majority", j:true\}}} can be missing after the {{mongod}} process is killed with SIGKILL and restarted. Only the process is killed. The host keeps running, so the kernel page cache survives, and anything {{mongod}} handed to the kernel should survive too. The same probe loses nothing on 6.0.29 or 7.0.45. It also loses nothing on any of these versions when run as a three-member replica set.
h3. Repro (script below)
# Start a stand
Affected Versions
8.0.328.0.348.0.38
SERVER-139297Open
Prevent unparseable collection ID in config.image_collection
In SERVER-137912, we handle what to do if a secondary replicates an oplog entry with an invalid session ID in this collection. We should also add validation on the write/input/primary side to not allow this.
SERVER-136920In Progress
Make E2E validate repair procedure tests work after checkpoint validation is completed
SERVER-139100Open
Add a metric to track the number of incoming connections by listener interface
Our incoming connections counters don't currently differentiate by listener interface. We should add a new otel metric which measures incoming connections with an attribute for the port that the connection was accepted on. This could be useful for internal dashboards since we could get a better idea of where incoming connections are coming from and how that would affect things like connection rate limiting among others.
PYTHON-6281Ready for Work
Enable all ruff FURB rules
h3. *Context*
{color:#505f79}Describe the background behind the problem.{color}
Only a portion of ruff's {{FURB}} linting rules are enabled by default. Enabling all of them improves performance and code consistency.
h3. *Definition of done*
{color:#505f79}What must be done to consider the task complete?{color}
Enable the entire suite of ruff {{FURB}} linting rules and make the required changes to pass them.
h3. *Pitfalls*
{color:#505f79}What should the implementer watch out for? What
Components
Infrastructure
PYTHON-6280Ready for Work
Enable non-design PLR ruff rules
h3. *Context*
{color:#505f79}Describe the background behind the problem.{color}
We disable all of ruff's PLR rules, but only the design-related ones should be disabled. The rest (PLR0402, PLR1704, PLR1711, PLR1714, PLR1730, PLR1733, PLR5501, PLR0124, PLR0133, PLR0206, PLR1708, PLR1716, PLR1722, PLR1736, PLR2044) do not deal with code design but instead reduce unnecessary or overly verbose code and should be enabled.
h3. *Definition of done*
{color:#505f79}What must be done to consider the
Components
Infrastructure
SERVER-128590Needs Scheduling
$persistentStats permissions
New stage created $persitentStats that will be used to read persisted stats needs the appropriate permissions. Look into items 2 and 6 from the threat model doc in: [https://docs.google.com/document/d/1n0SPv6ur9iwzddvy__TzIJxK3pzLXr17Dh-xOB931Cg/edit?tab=t.0]
SERVER-124353Open
$persistedStats unsharded
Handle $persistedStats for unsharded collections through mongos.
SERVER-124352Needs Scheduling
$persistedStats — mongos
"$persistedStats" with allHosts support for cross-shard visibility through mongos.
SERVER-133902Needs Scheduling
$persistedStats - mongod: all mode
This ticket should handle implementing all mode as specified [here|https://docs.google.com/document/d/1cLItPQB02UXl3r57SkgJOLN0cksIkbGpPOuFWTMww2M/edit?tab=t.0#heading=h.75wvx26w9k6m].