-
Type:
Bug
-
Resolution: Fixed
-
Priority:
Major - P3
-
Affects Version/s: None
-
Component/s: None
-
None
-
Replication
-
Fully Compatible
-
ALL
-
Repl 2026-03-16, Repl 2026-03-30, Repl 2026-04-13, Repl 2026-04-27, Repl 2026-05-11
-
None
-
None
-
None
-
None
-
None
-
None
-
None
For majority reads, the operationTime uses replCoord->getCurrentCommittedSnapshotOpTime(). That time may advance on a different thread, leading to operationTime being reported as larger than it was at the time of the read. Would the recoveryUnit's getPointInTimeReadTimestamp() be more correct?
- related to
-
SERVER-126539 Add TLA+ spec for snapshot-time vs operationTime ordering (companion to SERVER-120304)
-
- Needs Verification
-