> > - I still have the ceph-crash container, what should I do with it? > If it's the old one, I think you can remove it. Cephadm can deploy its own crash service (`ceph orch apply crash` if it hasn't). You can check if `crash` is listed under `ceph orch ls` and if it is there you can do `ceph orch ps --daemon-type crash` to check what crash daemons have been deployed. There should generally be one on each host. - The new rgw and mds daemons have some random string in their names (like > rgw.opsrgw.controllera.*pkajqw*), is this correct ? > Yes, cephadm names daemon types it might possibly have to co-locate on a node with 6 random chars at the end. - How should I proceed with the monitoring stack (grafana, prometheus, > alermanager and node-exporter)? should I stop then delete the old ones, > then deploy the new ones with ceph orch? > Cephadm has a `cephadm adopt` command that can be used to put prometheus, grafana, and alertmanager daemons under its control. You can do it sort of like how its done for the mon/mgr in https://docs.ceph.com/en/quincy/cephadm/adoption/#adoption-process but just give the name for the prometheus, grafana, or alertmanager daemon. That's all assuming these things have data you don't want to lose. If you just want monitoring under cephadm going forward and don't care about losing historical data, it could be easier to just remove them and deploy the monitoring stack with cephadm. (`ceph orch apply node-exporter`, `ceph orch apply grafana`, `ceph orch apply alertmanager`, `ceph orch apply prometheus`). Note that for node-exporter you'll have to do it this way anyway as there is no adoption process for node-exporter. On Thu, Feb 29, 2024 at 7:52 AM wodel youchi <wodel.youchi@xxxxxxxxx> wrote: > Hi, > > I am in the middle of migration from ceph-ansible to cephadm (version > quincy), so far so good ;-). And I have some questions : > - I still have the ceph-crash container, what should I do with it? > - The new rgw and mds daemons have some random string in their names (like > rgw.opsrgw.controllera.*pkajqw*), is this correct ? > - How should I proceed with the monitoring stack (grafana, prometheus, > alermanager and node-exporter)? should I stop then delete the old ones, > then deploy the new ones with ceph orch? > > Regards. > _______________________________________________ > 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
 |