normal status:
Fault Status:
Embodiment replacement steps:
1, closed ceph cluster data migration:
osd hard disk failure, the state becomes down. After mod osd down out interval set time interval, ceph to mark out, and the data migration start recovery. In order to reduce data recovery ceph scrub or other operation of the impact on performance, which can be temporarily closed first, replacing the hard disk to be restored and then turn osd:
for i in noout nobackfill norecover noscrub nodeep-scrub;do ceph osd set $i;done
2, locating the fault osd
ceph osd tree | grep -i down
3, enter osd node failure, uninstall osd mount directory
umount /var/lib/ceph/osd/ceph-5
4, removed from the crush map in osd
[root@node1 ~]# ceph osd crush remove osd.5
removed item id 5 name 'osd.5' from crush map
5, delete the key failures of osd
[root@node1 ~]# ceph auth del osd.5
updated
6, delete fault osd
[root@node1 ~]# ceph osd rm 5
removed osd.5
7. After the replacement hard drive, pay attention to the new hard drive letter, and create osd
8, the deployment of the node to switch to user cent, add new osd
[cent@deploy ceph]$ ceph-deploy osd create --data /dev/sdd node3
9. After adding new osd crush map, re-open clusters disable flag
for i in noout nobackfill norecover noscrub nodeep-scrub;do ceph osd unset $i;done
ceph clusters after data migration period of recovery active + clean state
