Ceph: systemctl start ceph-mon fails after moving to Jewel

[gdi2k]

I followed the tutorial to move Ceph from Hammer to Jewel here:
https://pve.proxmox.com/wiki/Ceph_Hammer_to_Jewel

All the steps went ok aside from starting the monitor daemon;

root@smiles2:~# systemctl start ceph-mon@ceph-mon.1.1500178214.095217502.service
root@smiles2:~# systemctl status ceph-mon@ceph-mon.1.1500178214.095217502.service
● ceph-mon@ceph-mon.1.1500178214.095217502.service - Ceph cluster monitor daemon
   Loaded: loaded (/lib/systemd/system/ceph-mon@.service; enabled)
  Drop-In: /lib/systemd/system/ceph-mon@.service.d
           └─ceph-after-pve-cluster.conf
   Active: failed (Result: start-limit) since Sun 2017-07-16 18:52:18 +08; 30s ago
  Process: 3057 ExecStart=/usr/bin/ceph-mon -f --cluster ${CLUSTER} --id %i --setuser ceph --setgroup ceph (code=exited, status=1/FAILURE)
 Main PID: 3057 (code=exited, status=1/FAILURE)

Jul 16 18:52:08 smiles2 systemd[1]: Unit ceph-mon@ceph-mon.1.1500178214.095217502.service entered failed state.
Jul 16 18:52:18 smiles2 systemd[1]: ceph-mon@ceph-mon.1.1500178214.095217502.service holdoff time over, sched...start.
Jul 16 18:52:18 smiles2 systemd[1]: Stopping Ceph cluster monitor daemon...
Jul 16 18:52:18 smiles2 systemd[1]: Starting Ceph cluster monitor daemon...
Jul 16 18:52:18 smiles2 systemd[1]: ceph-mon@ceph-mon.1.1500178214.095217502.service start request repeated t...start.
Jul 16 18:52:18 smiles2 systemd[1]: Failed to start Ceph cluster monitor daemon.
Jul 16 18:52:18 smiles2 systemd[1]: Unit ceph-mon@ceph-mon.1.1500178214.095217502.service entered failed state.
Hint: Some lines were ellipsized, use -l to show in full.

I was able to start it from the Web GUI and continue, and eventually complete the upgrade (including the ceph osd crush tunables step). But when a node is rebooted, the monitor must be started from the GUI manually each time. This affects all 3 nodes that I upgraded.

Any ideas?

 

[wolfgang]

Hi,

please do what the tutorial tell's you.
You start mon by the old naming schema.

From the upgrade doku

systemctl start ceph-mon@<MON-ID>.service
systemctl enable ceph-mon@<MON-ID>.service

"ceph-mon.1.1500178214.095217502" is not the mon id
"1" is the mon id.

 

[gdi2k]

Ah ok, I had misunderstood the docs (tab complete to get <UNIQUE ID>, not <MON-ID>). Works perfectly with

root@smiles1:~# systemctl start ceph-mon@0.service
root@smiles1:~# systemctl status ceph-mon@0.service
● ceph-mon@0.service - Ceph cluster monitor daemon
   Loaded: loaded (/lib/systemd/system/ceph-mon@.service; enabled)
  Drop-In: /lib/systemd/system/ceph-mon@.service.d
           └─ceph-after-pve-cluster.conf
   Active: active (running) since Mon 2017-07-17 15:59:54 +08; 7s ago
 Main PID: 5951 (ceph-mon)
   CGroup: /system.slice/system-ceph\x2dmon.slice/ceph-mon@0.service
           └─5951 /usr/bin/ceph-mon -f --cluster ceph --id 0 --setuser ceph --setgroup ceph

Jul 17 15:59:54 smiles1 systemd[1]: Started Ceph cluster monitor daemon.
Jul 17 15:59:54 smiles1 ceph-mon[5951]: starting mon.0 rank 0 at 10.15.15.50:6789/0 mon_data /var/lib/ceph/mon...ef05f
Hint: Some lines were ellipsized, use -l to show in full.
root@smiles1:~# systemctl enable ceph-mon@0.service
Created symlink from /etc/systemd/system/ceph-mon.target.wants/ceph-mon@0.service to /lib/systemd/system/ceph-mon@.service.

Many thanks for your help!