Actions
Feature #62647
open[RFE] Provide OSD's projected %usage in the cluster log or in ceph pg query
Status:
New
Priority:
Normal
Assignee:
-
Category:
-
Target version:
-
% Done:
0%
Source:
Tags:
Backport:
Reviewed:
Affected Versions:
Pull request ID:
Description
With reference to [[https://tracker.ceph.com/issues/62248]]. There is no way to know why PGs are stuck in backfill_toofull state even though all OSDs '%USE' is less than '50%'.
We can definitely document this behavior for the awareness but it would be nice to have projected usage of the OSD based on backfill space reservation for backfilling PGs logged to either to ceph cluster log file or in the ceph pg query.
Actions