Hi,
someone else had a similar issue [1], to set the global container
image you can run:
$ ceph config set global container_image my-registry:5000/ceph/ceph:v17.2.6
I usually change that as soon as a cluster is up and running or after
an upgrade so there's no risk of pulling wrong container images (I
assume in your case the local cephadm versions on the hosts differ and
therefore each one pulls a different default image hard-coded in the
cephadm binary).
You should probably be able to start a mgr daemon by changing the
unit.run file temporarily and replace "CONTAINER_IMAGE" with a correct
image version (stop the pod first):
CONTAINER_IMAGE=my-registry/ceph/ceph-quincy@v17.2.6 (this is just an
example).
The same line contains another image reference which you should
change. Then restart that pod (e. g. with systemctl), hopefully you'll
have a MGR up and running to be able to use the orchestrator again.
This procedure helped me in the past.
Regards,
Eugen
[1]
https://lists.ceph.io/hyperkitty/list/ceph-users@xxxxxxx/thread/THAH2JFQNB7B4BPUHTRDPGXJ75WPNSNK/
Zitat von Stefan Kooman <stefan@xxxxxx>:
On 15-09-2023 10:25, Stefan Kooman wrote:
I could just nuke the whole dev cluster, wipe all disks and start
fresh after reinstalling the hosts, but as I have to adopt 17
clusters to the orchestrator, I rather get some learnings from the
not working thing 🙂
There is actually a cephadm "kill it with fire" option to do that
for you, but yeah, make sure you know how to fix it when things do
not go according to plan. It all magically works, until it doesn't 😉.
cephadm rm-cluster --fsid your-fsid-here --force
... ss a last resort (short of wipefs / shred on all disks).
Gr. Stefan
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
