Bug #45737
closedModule 'cephadm' has failed: cannot send (already closed?)
0%
Description
Hi,
I have a development cluster running on 4 VM. They're all running CentOS8 Stream and were bootstraped using cephadm (octopus).
We were running tests, and I stopped VMs randomly to see how the cluster reacts.
When I siwtching VMs back on, the cluster state was Warning, saying that 2 nodes didn't passed the checks. So I went and run `ceph cephadm check-host HOST`, host1 was fine, but host2 gave a python stack trace, unfortunatly I've not copied it.
Now ceph cephadm says : Error EIO: Module 'cephadm' has experienced an error and cannot handle commands: cannot send (already closed?)
ceph -v
ceph version 15.2.2 (0c857e985a29d90501a285f242ea9c008df49eb8) octopus (stable)
cat /etc/centos-release
CentOS Linux release 8.1.1911 (Core)
Updated by Sebastian Wagner almost 4 years ago
- Project changed from Ceph to Orchestrator
- Category deleted (
ceph cli)
Updated by Sebastian Wagner almost 4 years ago
- Related to Bug #45627: cephadm: frequently getting `1 hosts fail cephadm check` added
Updated by Sebastian Wagner almost 4 years ago
- Status changed from New to Duplicate
Updated by Alain Deleglise almost 4 years ago
Hi,
So besides the fact this is a duplicate of an issue, that is waiting to have a fix reviewed, what should I do in the mean time ?
I mean, is it fixable in some way ? Should I wait for the fixe to be released and update components ? Right now my cluster is a developement cluster, so it's not crucial, but one can easily imagine this happening in production, what is the stance to have in such situation ?
Thanks