Unable to activate storage - cifs mount dies randomly

[autumnwalker]

Hello,

I've been running into this issue recently where my cifs mount for "backup" randomly goes offline. The icon in the GUI has a question mark and I start seeing entries like this in the log:

Oct 11 11:41:05 vmh01 pvestatd[1371]: unable to activate storage 'backup' - directory '/mnt/pve/backup' does not exist or is unreachable

I have other cifs mounts which are working fine attached to the same NAS device - so Proxmox is able to reach the device fine. Uses the same user account / credentials as well. It's just this one mount named 'backup'.

In order to recover I have to deactivate the storage and from the console, umount /mnt/pve/backup, and then reactivate the storage in the GUI.

Any thoughts on why this happens?

 

[autumnwalker]

Further - when I navigate in the console to /mnt/pve and run ls I get: "ls: cannot access 'backup': Stale file handle".

 

[autumnwalker]

I ran umount to fix it yesterday and overnight it went bad again. Any ideas?

 

[Ectropian]

I'm having the same issue.. my network storage goes unreachable

 

[nojah]

I have observed this too, but only for shares on my Unraid NAS.
It seems to occur when I add new files to a share that uses the cache disk, and when the files gets moved from the cache disk to the array.
If I disable the of the use chache disk for the same share, the issue seems to go away.

 

[autumnwalker]

I have observed this too, but only for shares on my Unraid NAS.
It seems to occur when I add new files to a share that uses the cache disk, and when the files gets moved from the cache disk to the array.
If I disable the of the use chache disk for the same share, the issue seems to go away.

Yeah, I have been playing around with this and notice the same behavior. NFS seems to be more tolerant, but still had the disconnect issue. Switching off cache on the share seems to resolve the problem. I've also opened a thread on the Unraid forum (https://forums.unraid.net/topic/86707-cifs-smb-stale-file-handle/), with no definitive solution.

 

[kalasnikov]

Yeah, I have been playing around with this and notice the same behavior. NFS seems to be more tolerant, but still had the disconnect issue. Switching off cache on the share seems to resolve the problem. I've also opened a thread on the Unraid forum (https://forums.unraid.net/topic/86707-cifs-smb-stale-file-handle/), with no definitive solution.

Where did you switch off the cache?

 

[autumnwalker]

Where did you switch off the cache?

On the share in Unraid.

 

[karloldreyes]

Bumping this thread. Any resolution here? Weird that we can't use the cache drive.

 

[Neobin]

Bumping this thread. Any resolution here? Weird that we can't use the cache drive.

Reading the thread [1] from @autumnwalker on the Unraid forum, the same problem also occurs for several people inside Linux-VMs and even on bare-metal Linux installations (non-PVE).

So in my humble opinion, this is not a Proxmox-specific problem at all.

[1] https://forums.unraid.net/topic/86707-cifs-smb-stale-file-handle

 

[snmp]

I am very new to Proxmox and found this thread when searching for the same issue.

I have the same issue, the only difference is the mounts that are going stale are on a different device than the ones that don't exhibit this behaviour.

But even the mounts that are going stale in Proxmox, none of the other devices, like my windows client has the issue. They can still access the shares.

Update: just thought i'd highlight i am not using Unraid. I have ubuntu running on a pi4 that is serving an attached storage as NAS.
I have another WD 18tb connected to the pi that doesn't have the same issue