I stood up a disposable ceph cluster under a nested VMware ESXi host (running under VMware Workstation) so I could learn the in’s and out’s of how Ceph functions. I stood up a single node Ceph monitor (Ubuntu 22.04 Desktop) and 3x ceph OSD nodes (allocated with 100GB of space via /dev/sdb), all running Ubuntu 22.04 server. I followed a 3rd party steps on setting up a Ceph cluster using Ansible (https://computingforgeeks.com/install-ceph-storage-cluster-on-ubuntu-linux-servers/). The cluster (i.e., monitors and ceph nodes) is up and running fine.
My question is I would like to use Juju to connect to said existing Ceph cluster so that I can use Juju to help manage it. I thought that since Juju acts more like an orchestrator, that you could just add already deployed instances to its database and then manage it from there. Juju has such a large learning curve that I why I found it was easier just to stand up a ceph cluster via an automation tool like Ansible (I’ve seen versions of doing it with Chef as well).
I can install the juju client on either a standalone Ubuntu server or one (or all) of the monitor nodes.
Is there any how-to document or steps available on how to do this?