Bug #50747
nautilus: osd: backfill_unfound state reset to clean after osd restart
% Done:
0%
Source:
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Component(RADOS):
Pull request ID:
Crash signature (v1):
Crash signature (v2):
Description
On nautilus we have been observing an issue when an EC pg is in active+backfill_unfound+degraded state (which happens when unfound objects are detected during backfill), and a non-primary osd for this pg is restarted, the state is changed to 'active+clean' after the restart and information about unfound objects is lost.
This issue does not observe for a post-nautilus osd. See https://tracker.ceph.com/issues/50351 for more details and steps to reproduce.
History
#1 Updated by Mykola Golub almost 3 years ago
- Status changed from In Progress to Fix Under Review
- Pull request ID set to 41293