Ceph OSD folder found empty

[PutnamCountyIT]

one of 4 nodes has lost the osd configuration
all nodes are running:
proxmox-ve: 6.0-2 (running kernel: 5.0.21-1-pve)
pve-manager: 6.0-6 (running version: 6.0-6/c71f879f)
ceph: 14.2.2-pve1
ceph-fuse: 14.2.2-pve1
corosync: 3.0.2-pve2

The osd gui screen shows 4 (of 16) osd drives as down and out. When looking the /var/lib/ceph/osd/ceph-0 folder for that node it is empty. (same with ceph-1,ceph-2,ceph-3). The crush map shows those devices. The 4 drives (sda,sdb,sdc,sdd) are still configured as Ceph Journal per fdisk. The usual ownership assignment to the raw disk is not ceph:ceph but is root:root

what is the best method to restore the osd configuration files and get the OSD back up UP and In.

Thank you

 

[dcsapak]

i would check the logs as to why those osds did not start up..
if this cluster is upgraded from luminous, did you do the osd activation step on that host?
see https://pve.proxmox.com/wiki/Ceph_Luminous_to_Nautilus

 

[PutnamCountyIT]

Thank you for the tip. I did miss a step to turn version from 1.0 to 2.0.
I also found that the "ceph-volume simple scan /dev/sdbx" command followed by "ceph-volume simple activate osdnumber uuid" can restore the osd volumes.

Thank you for the help

Bob