Stale file handle on nfs share in my main pool

[peterge]

root@paperless-ngx:~# docker-compose -f /root/docker-compose.yml exec -T webserver touch /mnt/paperless-truenas/opa
touch: setting times of '/mnt/paperless-truenas/opa': Stale file handle

root@paperless-ngx:~# cat /etc/fstab
# UNCONFIGURED FSTAB FOR BASE SYSTEM
10.0.12.100:/mnt/main/Home_Share_Dataset/peterge/Dokumente/paperless /mnt/paperless-truenas nfs bg 0 0
root@paperless-ngx:~#

Running latest SCALE. NFS Client is a debian 11 lxc

 

[anodos]

Sometimes permissions issues can present as a stale file handle. Maybe check permissions across full path to the export for the maproot user that you have selected (defaults to nobody:nogroup) (presuming you're using root for the NFS mount).

Or at least verify that the user in question can `cd` into the path.

 

[peterge]

yes, that user can cd in that dir and create files etc. I have mapall user and mapall group set too for this share.

 

[anodos]

yes, that user can cd in that dir and create files etc. I have mapall user and mapall group set too for this share.

Client is not hosted on the NAS, correct?

If not, then try increasing NFSD debug levels on our server.

Code:

root@truenas[~]# midclt call nfs.get_debug
{"NFS": ["NONE"], "NFSD": ["NONE"], "NLM": ["NONE"], "RPC": ["NONE"]}
root@truenas[~]# midclt call nfs.set_debug '["NFSD", "NLM", "RPC"]' '["ALL"]'
null
root@truenas[~]# midclt call nfs.get_debug                                  
{"NFS": ["NONE"], "NFSD": ["ALL"], "NLM": ["ALL"], "RPC": ["ALL"]}

And check logs. This might give some clues about what is going wrong.
Maybe do same on client.

 

[peterge]

No, client is running on a seperate host in a proxmox lxc container.
Which file stores nfs logs? grep -rni nfs in /var/log returns a lot of matches...

 

[peterge]

Solved by using autofs