Details
-
Bug
-
Resolution: Duplicate
-
Major - P3
-
None
-
3.6.0-rc3
-
None
-
ALL
Description
Currently, each write that's part of a batch performs two separate checks for shard version. Once at the beginning of the batch and a second time when the write has been written to the oplog.
Performing these checks causes a hot mutex to be acquired in order to retrieve the sharding state for the collection, so the first one should ideally be removed.
Attachments
Issue Links
- duplicates
-
SERVER-25061 Make CollectionShardingState a decoration on Collection
-
- Closed
-
- is depended on by
-
SERVER-31694 17% throughput regression in insert workload
-
- Closed
-