NFS Share suddenly stopped being recognized by PVE

[vrocco]

I had an NFS share to my Qnap all set up and working fine. However, all of the sudden I can't see any of the details in PVE. It keeps giving me the error:

mkdir /mnt/pve/nas_storage/private: Read-only file system at /usr/share/perl5/PVE/Storage/Plugin.pm line 1322. (500)

I don't know what I did or how to fix it and I'm hoping someone can help out. Strangely, the container I am currently using this share for (Plex) still has the share mounted and I can access it no problem from the container. I just can't access it from PVE for any backups or ISOs or anything.

Can someone help me get this fixed? Thank you in advance!

 

[bbgeek17]

Could be:
https://pve.proxmox.com/wiki/Roadmap

Known Issues

Setups mounting a QNAP NFS share could not be mounted with NFS version 4.1 with kernel pve-kernel-5.15.30-2-pve - the issue has been mitigated in kernels pve-kernel-5.15.35-2 and above.
If your QNAP NFS share cannot be mounted upgrade the kernel and reboot.
As an alternative mitigation you can explicitly set the NFS version to 4 (not 4.1 or auto).

Or, its possible, the device really went r/o due to some condition. When you run "df" do you see it mounted to path in /mnt/pve/[storagename] ?
Can you "cd" there and "touch testfile;rm testfile" ?
What does "mount|egrep [storagename]" show?

---

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

 

[vrocco]

This might be an issue. I did try to set the NFS version to 4 explicitly, but that didn't seem to make a difference.

I seem to be on 5.13.35-1, but not sure how to get to 35-2. When I try to install, I get:

root@prox:~# uname -r
5.15.35-1-pve
root@prox:~# apt install pve-kernel-5.15
Reading package lists... Done
Building dependency tree... Done
Reading state information... Done
pve-kernel-5.15 is already the newest version (7.2-3).
pve-kernel-5.15 set to manually installed.
0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded.

Hmmm it does seem to be r/o from PVE......not sure how it got that way. When I try and touch a file from the PVE console, I get the following:

root@prox:/mnt/pve/nas_storage# touch testfile
touch: cannot touch 'testfile': Read-only file system

However, if I run the same thing from the console in the Plex LXC, I can create and delete the testfile.


the command you specified to run produces:

root@prox:/mnt/pve/nas_storage# mount | egrep nas_storage
192.168.1.51:/Media on /mnt/pve/nas_storage type nfs4 (rw,relatime,vers=4.2,rsize=1048576,wsize=1048576,namlen=255,hard,proto=tcp,timeo=600,retrans=2,sec=sys,clientaddr=192.168.1.90,local_lock=none,addr=192.168.1.51)

 

[bbgeek17]

You can examine your logs to find out when the mount went R/O: journalctl -b . That may help to understand why.
Generally, there could be many reasons. Google "nfs read only" or similar.
As for the container - you are most likely mounting it there directly and not via bind-mount from host.
You can try to force a remount: systemctl try-reload-or-restart pvedaemon pveproxy pvestatd
Or it may require a reboot.

---

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

 

[vrocco]

I'm an idiot. Checked the host permissions in NFS again and my Proxmox node wasn't added as R/W. I have been moving stuff around, so IPs must have gotten mixed up. Fixed now. Thanks for your patience.

 

[bbgeek17]

However, all of the sudden I can't see any of the details in PVE

I'm an idiot. I have been moving stuff around, so IPs must have gotten mixed up

Good for you! That's a serious self-progress in a span of 90 minutes

---

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