Hi all,
We're upgrading a cluster from luminous to nautilus. The monitors and managers are running a non-release version of luminous (12.2.12-642-g5ff3e8e) and we're upgrading them to 14.2.9.
We've upgraded one monitor and it's happily in quorum as a peon. However, when a ceph status hits the nautilus mon it has trouble talking to the manager apparently, and it returns a status output with no pg stats and garbage usage numbers. From the mon log:
2020-05-13 10:41:43.121 7fa1e6fdf700 0 mon.ceph-mon5@4(peon) e25 handle_command mon_command({"prefix": "status"} v 0) v1
2020-05-13 10:41:43.121 7fa1e6fdf700 0 log_channel(audit) log [DBG] : from='client.? v1:130.246.x.x:0/3261311028' entity='client.admin' cmd=[{"prefix": "status"}]: dispatch
2020-05-13 10:41:43.443 7fa1e6fdf700 -1 mon.ceph-mon5@4(peon).mgrstat failed to decode mgrstat state; luminous dev version? buffer::end_of_buffer
2020-05-13 10:41:44.397 7fa1e6fdf700 1 mon.ceph-mon5@4(peon) e25 dropping unexpected mon_health( e 0 r 0 ) v1
Is this expected for a luminous to nautilus upgrade or could this be due to the odd luminous version we are running, or something else entirely?
Cheers,
Tom
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
