CephFS deletion when using RBD

[sviccc]

Hi there. Is it safe to remove CephFS, when im using RBD only? I removed all cephfs pools, mgrs and now i need to remove CephFS.
Is "ceph fs rm cephfs --yes-i-really-mean-it" sufficient? Wont it destroy something needed for RBD to work?
Thank you for answers

 

[djardim]

I believe CephFS is a feature enabled by the use of metadata servers and specially created data and metadata pools. The goal is to provide a mountable filesystem that can be used as a data storage by the Proxmox cluster, or as a remote filesystem/storage for other systems, without the hassle of creating a block device (e.g., virtual hard drive), formatting it and using a bogus OS just to share the storage space. So CEPH already makes that easy by providing the CephFS feature.

Once you remove those ingredients, CephFS ceases to work, and Ceph is left with it's regular internal workings (monitors, managers, OSDs, etc). And then I believe no further work is needed to remove CephFS.

On the other hand, if you're talking about removing Ceph itself, then you can check this post.

 

[Drallas]

Hi there. Is it safe to remove CephFS, when im using RBD only? I removed all cephfs pools, mgrs and now i need to remove CephFS.
Is "ceph fs rm cephfs --yes-i-really-mean-it" sufficient? Wont it destroy something needed for RBD to work?
Thank you for answers

Yes it's safe to remove, the default Proxmox Ceph installation also comes without a CephFS, Ceph_data and Ceph_metadata pools. It's a little a bit of a pain to do so, https://pve.proxmox.com/pve-docs/chapter-pveceph.html#_destroy_cephfs