Actions
Feature #8609
openImprove ceph pg repair
Status:
New
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:
0%
Source:
Community (user)
Tags:
Backport:
Reviewed:
Affected Versions:
Component(RADOS):
Pull request ID:
Description
At the moment, ceph uses the primary copy to repair an inconsistent pg. This is not always a good choice, if there has been a previous disk-error, write error or something else discovered by a deep-scrub.
It would be better, if
1) ceph could be told, which copy to replicate from as an option to "ceph pg repair"
2) ceph would automatically exclude a copy, which fails the tests performed by a deep-scrub
Files
Updated by Aaron T almost 10 years ago
+1 I think this is really important and would reduce the amount of time I spend managing my cluster.
Updated by cheng li about 7 years ago
- File pg_repair.txt pg_repair.txt added
Has anyone fixed this bug?
My test results show that pg repair is smart enough.
ceph can find the right replica, and copy it to nodes which have bad replica.
See the attachment for test details.
Updated by Patrick Donnelly about 5 years ago
- Project changed from Ceph to RADOS
- Category deleted (
ceph cli)
Updated by Laura Flores over 1 year ago
- Translation missing: en.field_tag_list set to low-hanging-fruit
Updated by Laura Flores 7 months ago
- Translation missing: en.field_tag_list deleted (
low-hanging-fruit)
Actions