Actions
Support #38152
closedrocksdb problem in the 108 down osds
% Done:
0%
Tags:
Reviewed:
Affected Versions:
Pull request ID:
Description
My cluster is luminous 12.2.10, upgrade from jewel. so the cluster has 108 osds down in 3 hosts, 216 osds is up. I want to upgrade to mimic(13.2.4) to solve the rocksdb problem( the issue: http://tracker.ceph.com/issues/25146) in the 108 down osds. but my cluster doesn't satisfied the purged_snapdirs flags. I wait for more than 48 hours, and execute operation:ceph pg dump pgs_brief | cut -d " " -f 1 | xargs -n1 ceph pg scrub . It doestn't work.
If i upgrade to mimic without the purged_snapdirs flags, my mon can't work.
Updated by Patrick Donnelly about 5 years ago
- Status changed from New to Rejected
Please seek help on ceph-users.
Actions