Ceph storage pool full and other processing methods

1. Check the settings of the current cluster full and nearfull, and query the Mon configuration on the Mon node.

Ceph--admin-daemon / run/ceph/ceph-mon.IP.asok config show | grep full_ratio

two。 Check the setting values of the current cluster full and nearfull, and query the osd configuration on other nodes

Ceph--admin-daemon / run/ceph/ceph-osd.0.asok config show | grep full

3. If you encounter an alarm from the cluster full and the storage cannot write data, emergency configuration steps

1) set OSD to prohibit reading and writing

Ceph osd pause

2) notify Mon to modify the threshold of nearfull and full

Ceph tell mon.* injectargs "--mon-osd-nearfull-ratio 0.96"

Ceph tell mon.* injectargs "--mon-osd-full-ratio 0.97"

Ceph tell osd.* injectargs "--mon-osd-nearfull-ratio 0.96"

Ceph tell osd.* injectargs "--mon-osd-full-ratio 0.97"

3) notify PG to modify the threshold of nearfull and full

Ceph pg set_nearfull_ratio 0.96

Ceph pg set_full_ratio 0.97

4) remove the ban on reading and writing of OSD

Ceph osd unpause

5) organize the cluster, delete some useless data, and expand the cluster.

6) restore the default configuration of the cluster

Ceph tell mon.* injectargs "--mon-osd-nearfull-ratio 0.85"

Ceph tell mon.* injectargs "--mon-osd-full-ratio 0.95"

Ceph tell osd.* injectargs "--mon-osd-nearfull-ratio 0.85"

Ceph tell osd.* injectargs "--mon-osd-full-ratio 0.95"

Ceph pg set_nearfull_ratio 0.85

Ceph pg set_full_ratio 0.95