-
Type:
Task
-
Resolution: Unresolved
-
Priority:
Major - P3
-
None
-
Affects Version/s: None
-
Component/s: None
-
Replication
-
None
-
3
-
None
-
None
-
None
-
None
-
None
-
None
It is technically possible for an overeager user to run killOp against user threads, which would result in opCtx's getting unexpectedly interrupted. In some cases (like against the OplogWriterImpl thread) this can lead to a server crash if it is not handled.
This requires a user to have elevated privileges and kill an operation they have no reason to kill, but do we want to take action to prevent a server crash in instances like this?