Le 29/11/2023 à 11:44:57-0500, Adam King a écrit Hi,
I think I remember a bug that happened when there was a small mismatch between the cephadm version being used for bootstrapping and the container. In this case, the cephadm binary used for bootstrap knows about the ceph-exporter service and the container image being used does not. The ceph-exporter was removed from quincy between 17.2.6 and 17.2.7 so I'd guess the cephadm binary here is a bit older and it's pulling hte 17.2.7 image. For now, I'd say just workaround this by running bootstrap with `--skip-monitoring-stack` flag. If you want the other services in the monitoring stack after bootstrap you can just run `ceph orch apply <service>` for services alertmanager, prometheus, node-exporter, and grafana and it would get you in the same spot as if you didn't provide the flag and weren't hitting the issue.
For an extra note, this failed bootstrap might be leaving things around that could cause subsequent bootstraps to fail. If you run `cephadm ls` and see things listed, you can grab the fsid from the output of that command and run `cephadm rm-cluster --force --fsid <fsid>` to clean up the env before bootstrapping again.
I run into the same problem few weeks ago. My mistake was to use cepahdm 17.2.6 and trying to install the 17.2.7 version because I forget to remove the cephadm 17.2.6 on my server (currently I try to fit the install of ceph «into» our puppet config. Soon I remove the old version of cephadm and install the 17.2.7 version everything work fine again. Regards. -- Albert SHIH 🦫 🐸 France Heure locale/Local time: mer. 29 nov. 2023 22:06:35 CET