Actions
Feature #4835
closedMonitor: better handle aborted synchronizations
Status:
Resolved
Priority:
Normal
Assignee:
-
Category:
Correctness/Safety
Target version:
-
% Done:
0%
Source:
other
Tags:
Backport:
Reviewed:
Affected Versions:
Component(RADOS):
Pull request ID:
Description
See #4812. That should not be an assert (graceful shutdowns!), and in that specific case we don't actually want to exclude the MDS which aborted anyway. Either it should have done a nicer abort (telling the sync requester to restart), or maybe the requester should be okay with going back to the aborter? I'm not really sure. Either way, this should be cleaned up, and if we can do so in a way that reduces the size of the interface, so much the better.
Actions