[SOLVED] Problem reconnecting NFS share after problem with NFS server

[kmadsen]

I've been successfully using an NFS share from a small NAS as a backup target for one of my PVE cluster nodes. I had a problem during an update on the NAS that interrupted the FNS service. That's been fixed, but I can't reconnect PVE to the NFS share.

Running the df command shows that the share (mserver3) *appears* to be mounted:

Filesystem                          Size  Used Avail Use% Mounted on
udev                                 34G     0   34G   0% /dev
tmpfs                               6.8G  1.8M  6.8G   1% /run
/dev/mapper/pve-root                101G   35G   62G  37% /
tmpfs                                34G   76M   34G   1% /dev/shm
tmpfs                               5.3M     0  5.3M   0% /run/lock
efivarfs                            154k   76k   73k  52% /sys/firmware/efi/efivars
/dev/nvme0n1p2                      1.1G   13M  1.1G   2% /boot/efi
/dev/sdb2                           5.1T  2.9T  2.2T  57% /mnt/pve/sg-backup02
/dev/fuse                           135M   33k  135M   1% /etc/pve
10.0.0.201:/shares/MBPool/Backup01  250G  224G   27G  90% /mnt/pve/mbackup01
mserver3.local:/export/mshare3       63G  3.1G   57G   6% /mnt/pve/mshare3
tmpfs                               6.8G     0  6.8G   0% /run/user/0

However, it is now and the mount directory is empty. The size statistics are also incorrect.

Looking up previous posts I've tried unmount and remounting, rebooting, and deleting/re-creating the NFS storage entry, all to no avail. I am running version 8.4.1.

I would really appreciate some pointers of where to look and/or how to force the system to "refresh" this NFS connection.

Thanks

 

[gfngfn256]

1. Show output for:

cat /etc/pve/storage.cfg

2. What procedure (initially) did you use to connect that NFS share in Proxmox.

3. I see that you have used a DNS name for the NFS server: mserver3.local ; have you considered using the actual IP address (as suggested in the docs)? You seem to have done this with the mount entry above: 10.0.0.201:/shares/MBPool/Backup01 . Are those 2 different servers?

4. If you ping the server from the host using the DNS name : ping mserver3.local do you get a response?

5. Have you checked whether after that update on the NFS server the NFS version may have changed?

6. Have you checked (as in the docs) with pvesm scan nfs mserver3.local which shares are actually exported?

 

[kmadsen]

Thank you for your good suggestions. It turns out to be simpler than I thought (as is sometimes/often the case). I did most of the things suggested, but the problem was actually with the NFS server, not with proxmox. There had been an update problem with the ZFS modules and I had to import the zfs pool that was being shared. Although the NFS share looked like it was OK, it wasn't ok and I had to delete and re-create the share. Then Proxmox picked right up after a umount and everything was back working as intended.