I just tried the same with Ubuntu 24.04, same error. Just for the record, this is the bootstrap command: root@tentacle-ubuntu24:~# cephadm --image quay.io/ceph/ceph:v20.2.0 bootstrap --mon-ip 192.168.124.172 --single-host-defaults --allow-mismatched-release --skip-firewalld --allow-overwrite --skip-monitoring --log-to-file Zitat von Eugen Block <eblock@nde.ag>:
Hi,
I tried bootstrapping a fresh test cluster on a virtual machine (openSUSE Leap 16), and it fails:
---snip--- Setting orchestrator backend to cephadm... Non-zero exit code 95 from /usr/bin/podman run --rm --ipc=host --stop-signal=SIGTERM --net=host --entrypoint /usr/bin/ceph --init -e CONTAINER_IMAGE=quay.io/ceph/ceph:v20.2.0 -e NODE_NAME=tentacle -v /var/log/ceph/c31e1b94-c921-11f0-b1cf-fa163e7037b6:/var/log/ceph:z -v /tmp/ceph-tmpqdv5t793:/etc/ceph/ceph.client.admin.keyring:z -v /tmp/ceph-tmp615joe66:/etc/ceph/ceph.conf:z quay.io/ceph/ceph:v20.2.0 orch set backend cephadm /usr/bin/ceph: stderr Error ENOTSUP: Warning: due to ceph-mgr restart, some PG states may not be up to date /usr/bin/ceph: stderr Module 'orchestrator' is not enabled/loaded (required by command 'orch set backend'): use `ceph mgr module enable orchestrator` to enable it RuntimeError: Failed command: /usr/bin/podman run --rm --ipc=host --stop-signal=SIGTERM --net=host --entrypoint /usr/bin/ceph --init -e CONTAINER_IMAGE=quay.io/ceph/ceph:v20.2.0 -e NODE_NAME=tentacle -v /var/log/ceph/c31e1b94-c921-11f0-b1cf-fa163e7037b6:/var/log/ceph:z -v /tmp/ceph-tmpqdv5t793:/etc/ceph/ceph.client.admin.keyring:z -v /tmp/ceph-tmp615joe66:/etc/ceph/ceph.conf:z quay.io/ceph/ceph:v20.2.0 orch set backend cephadm: Error ENOTSUP: Warning: due to ceph-mgr restart, some PG states may not be up to date Module 'orchestrator' is not enabled/loaded (required by command 'orch set backend'): use `ceph mgr module enable orchestrator` to enable it ---snip---
Setting it afterwards succeeds, but then you end up with a semi-broken cluster:
[ceph: root@tentacle /]# ceph orch set backend cephadm [ceph: root@tentacle /]# ceph orch host ls 0 hosts in cluster
I created a tracker without checking if anything similar already exists:
https://tracker.ceph.com/issues/73974
I haven't tried bootstrapping with other distros yet, have others experienced something similar on a different OS?
Thanks, Eugen