-
Type: Improvement
-
Resolution: Done
-
Priority: Major - P3
-
Affects Version/s: None
-
Component/s: Replication
-
None
it would be bad to do writes for an hour and then discover the node transitioned to secondary 59 minutes ago.
of course one call getlasterror today, but this is solveable without that so we should do it.
one possible solution: close all connections on a transition from primary->secondary.
- related to
-
JAVA-275 with default write concern (no getLastError), if master changes in repl set, small edge case where writes will fail undetected
- Closed