-
Type: Task
-
Resolution: Won't Fix
-
Priority: Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
None
-
Replication
We should determine if there is a way to tracks metrics around replication coordinator mutex contention without taking too much of a performance hit. These metrics would have been useful for diagnosing recent cluster slowness issues more quickly, as the high mutex contention impacted read and write latency but it wasn't immediately obvious.
There is a possibility that this would be too costly, but we should investigate in this ticket.
- related to
-
SERVER-90745 Improve observability for mutex contention
- Closed