-
Type: Bug
-
Resolution: Fixed
-
Priority: Major - P3
-
Affects Version/s: None
-
Component/s: Sharding
-
Fully Compatible
-
ALL
-
v4.2
-
Sharding 2019-02-25, Sharding 2019-06-03, Sharding 2019-06-17
The range deleter deletes orphan documents by constructing a PlanExecutor with a YIELD_MANUAL yield policy. Because of this, any write conflict errors encountered will be re-thrown, but there is no WriteConflictRetry loop in CollectionRangeDeleter::cleanUpNextRange, causing post-migration orphan cleanup and cleanupOrphaned commands to fail unnecessarily.
The net effect of this is that more orphans remain on the system.
- related to
-
SERVER-38972 mongos takes forever for multi shard queries
- Closed
-
SERVER-43518 Complete TODO listed in SERVER-39141
- Closed