Details
-
Task
-
Resolution: Done
-
Major - P3
-
None
-
None
Description
I think there may be a case similar to SERVER-5999 where a kill op request to a remove operation that has not yet performed a write can trigger an interrupt exception inside a cursor advance() in ClientCursor::aboutToDelete(). This could leave a MultiCursor in an inconsistent state, as in SERVER-5999. Worth investigating and fixing if necessary.
Attachments
Issue Links
- is related to
-
SERVER-5999 page fault exception in ClientCursor::aboutToDelete() can leave another Cursor / ClientCursor in an inconsistent state
-
- Closed
-