Please Help !!! : Error "Stale file handle" when Backup (Schedule or Manual).

[jean1]

Hello everyone.

I have a Proxmox 8.1.4 VM and when I Start a Schedule Backup I had this error, I set to 3 backups limit when the all 3 are completed the next backup fails, even with only manual default backup limit or manual backup I had the same error, sometimes the backups works and sometime doesnt, its like a random error.


I attach the error file.


Please Help me !!!!!

 

[bbgeek17]

If the "CHANGITO-ROOT3" is NFS, that error usually means that the NFS server has rebooted or otherwise restarted. That, in turn, caused the host to be out of sync with NFS server regarding file handles that the host may have cached.
One way to address this is to avoid restarting NFS server. Otherwise, you may need to restart some services on PVE before backup. However, I am not sure which one.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[jean1]

Hello!!!.

Thanks bbgeek17 for the answer, I m going to make more test.

 

[crashtest]

If the "CHANGITO-ROOT3" is NFS, that error usually means that the NFS server has rebooted or otherwise restarted. That, in turn, caused the host to be out of sync with NFS server regarding file handles that the host may have cached.
One way to address this is to avoid restarting NFS server. Otherwise, you may need to restart some services on PVE before backup. However, I am not sure which one.

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

I have been having this issue as well, except I'm trying to get baackups to work on PBS... I was very careful to boot up the NFS server and then the PBS. In my research on ESTALE, it appears that this can be resolved if the software just does another lookup. My guess is that PBS is checking the NFS then failing out after one error.

 

[bbgeek17]

In my research on ESTALE, it appears that this can be resolved if the software just does another lookup.

I dont know if your issue is related to Ops. That said there is a big difference between "stale file handle" and "stale mount handle". One can be refreshed, the other needs to be forcefully remounted.

https://unix.stackexchange.com/questions/433051/mount-nfs-stale-file-handle-error-cannot-umount

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

 

[crashtest]

I dont know if your issue is related to Ops. That said there is a big difference between "stale file handle" and "stale mount handle". One can be refreshed, the other needs to be forcefully remounted.

https://unix.stackexchange.com/questions/433051/mount-nfs-stale-file-handle-error-cannot-umount

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox

I've also tried forcefully remounting then attempting a backup immediately after. This error no matter what order, clean boot, etc etc I have tried

INFO: Error: unable to update manifest blob - unable to load blob '"/mnt/pve-backup/ct/701/2024-03-25T22:00:35Z/index.json.blob"' - Stale file handle (os error 116)

That timestamp directory is never there either the times I went looking for the index.json.blob file

 

[bbgeek17]

I've also tried forcefully remounting then attempting a backup immediately after. This error no matter what order, clean boot, etc etc I have tried

Its possible you have a network issue or your NFS server is broken.

My guess is that PBS is checking

OP does not mention using PBS server. Although the two products come from the same company, they are not the same.

That timestamp directory is never there either the times I went looking for the index.json.blob file

You should open your own thread and probably in a different section of the forum, the one dedicated to PBS

good luck

---

Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox