Containers on GlusterFS?

Johannes S said:
This will change as soon as the qemu version integraded in ProxmoxVE doesn't have gluster-support anymore. Since qemu deprecated it and announced that they will remove this feature at some point Proxmox developers removed it too:
K

Thread 'Glusterfs is still maintained. Please don't drop support!'

The PVE 9.0 Beta Release Notes states: As GlusterFS is no longer maintained upstream, Proxmox VE 9 drops support for GlusterFS storages.

But this is simply not true. Glusterfs is actively maintained. The last commit and release (11.2) happened just three weeks ago! Furthermore, Debian Trixie has recent packages for gluster server, client, and cli.

While I understand that glusterfs may be in an awkward spot for enterprise since Red Hat EoL-ed Red Hat Gluster Storage, that happened in 2024, and it has not stopped development of the open source project...
  • Like

Beginning with qemu version 11.1 the gluster support was removed: https://qemu-project.gitlab.io/qemu/about/removed-features.html#gluster-backend-removed-in-11-1 With other words: As soon as the integrated qemu version in ProxmoxVE will jump to 11.1 the gluster plugin wouldn't work more anyhow. Right now ProxmoxVE ships version 11.0 with their updates so at some point (maybe tommorow, maybe in a few months) the glusterfs plugin would stop working. To continue to work the Proxmox developers would need to maintain their own fork of qemu which they considered as too much work for not enough benefit since most of their enterprise customers doesn't use glusterfs. So the gluster plugin was removed with the release of PVE9 since at that point (one year ago) PVE8 would still be supported for another year so any users of glusterfs would have enough time to prepare (by migrating to another storage or changing their setup to directory-based mount of gluster).

And whether it worked fine is highly debatable: In the linked thread several people explained why (depending on usecase etc) glusterfs isn't really that great compared to Ceph or other storage solutions.
Click to expand...
Again, I understand that native support for GlusterFS has been removed (qemu removed it, so it really has very little to do with Proxmox itself).

But that still doesn't answer the question as to why the native GlusterFS client didn't support LXC rootdirs natively and why I had to enable that by adding it as a directory storage type instead.

(On an unrelated note, I just re-deployed my three node Proxmox HA cluster using Gluster rather than ceph because ceph's drive capability utilisation is quite poor, especially with EC Ceph. Gluster, with a distribute dispersed gvol, so far, fairs significantly better (8-9X) the drive performance capability utilisation.)
 
You must log in or register to reply here.
Top Bottom
Back