-
Type:
Improvement
-
Resolution: Unresolved
-
Priority:
Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
None
-
Replication
-
Repl 2024-12-09, Repl 2024-12-23, Repl 2025-01-06, Repl 2025-01-20, Repl 2025-02-03, Repl 2025-03-03, Repl 2025-03-17, Repl 2025-03-31, Repl 2025-05-12, Repl 2025-05-26, Repl 2025-06-09, Repl 2025-07-07
-
None
-
3
-
None
-
None
-
None
-
None
-
None
-
None
-
None
We should add a log message at the end of stepdown summarizing the total time it took from a node finding out it should step down until it actually has killed all writes and starts replicating operations, with details about how long subsections took. This both can catch deadlocks and uninterruptible operations, and measures how long retryable write latency increases during elections.