Actions
Fix #7564
closedsynchronize MDS and client times in a way that makes pjd happy even under clock skew
Status:
Duplicate
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:
0%
Source:
Q/A
Tags:
Backport:
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(FS):
MDS
Labels (FS):
Pull request ID:
Crash signature (v1):
Crash signature (v2):
Description
See #854. We have ops happen on both the client and the MDS, and so sometimes one time wins and sometimes the other does. This is usually okay, but eg can make Hadoop unhappy when the clocks change unexpectedly from what's read after an op completes on the client.
Actions