On 11/25/24 14:09, Filip Schauer wrote:
When mounting volumes as read-only, certain mount options like
"discard", "lazytime", and "noatime" are either ignored or can cause the
mount to fail. For example, attempting to mount with "-t zfs" and
"-o ro,discard" leads to an error: filesystem cannot be mounted due to
invalid option 'discard'.

This commit ensures that only valid mount options, such as "nodev",
"noexec", and "nosuid", are applied to read-only mounts, avoiding
potential mount failures.

Signed-off-by: Filip Schauer <f.scha...@proxmox.com>

I could reproduce the issue when backing up a container with at least one root disk or mountpoint, which has at least the mount option "discard" applied, and backing it up to a PBS instance (to stay close to the bug report I used v3.2.10-1 for this), while all images were stored on a ZFS pool.

After applying the patch, the container could be backed up to the same PBS instance without any trouble.

Reviewed-by: Daniel Kral <d.k...@proxmox.com>
Tested-by: Daniel Kral <d.k...@proxmox.com>


_______________________________________________
pve-devel mailing list
pve-devel@lists.proxmox.com
https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel

Reply via email to