-
Type: Bug
-
Resolution: Duplicate
-
Priority: Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
None
-
Cluster Scalability
-
ALL
As part of the ShardingTest teardown, SERVER-88191 is adding new logic to drain inflight reshardCollection requests based on the response of the currentOp command.
Test failures showed that the execution of such logic is causing a segfault while accessing a nullptr _metrics field: the problem seems to only affect the jstests/sharding/reshard_collection_retryability_with_uuid.js and the jstests/sharding/reshard_collection_retry_after_failover.js tests when executed under the sharding_auto_bootstrap suite.
While the teardown validation has been disabled for the failing tests as a workaround, the objective of this ticket is to understand the root cause of the problem.
- duplicates
-
SERVER-88651 Ensure resharding metrics object still exists when reporting currentOp info
- Closed
- is related to
-
SERVER-88191 CheckRoutingTableConsistency might check sharding's catalog in transient state
- Closed
- related to
-
SERVER-89063 Complete TODO listed in SERVER-88620
- Open