Actions
Support #48530
closedceph pg status in incomplete.
Status:
Closed
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:
0%
Component(RADOS):
Pull request ID:
Description
We have a ceph cluster,
A server was down on December 3, and the server was restored on December 7. The pgid was 7.62b that entered the peering state at 9:15, but it entered the incomplete state after 9:33.
As a result ,some io requested was blocked, and then some services were affected.
So we want to know, what conditions will trigger pg to transition from peering to incomplete?
See the attachment for log details.
Updated by Greg Farnum almost 3 years ago
- Project changed from Ceph to RADOS
- Status changed from New to Closed
This kind of question is best served on the ceph-users@ceph.io mailing list if you can't find the answer in the documentation.
Generally speaking though, PGs go incomplete when there are not enough running OSDs to satisfy min_size or the PG might be missing data.
Actions