Proxmox NFS Storage Status Unknown on Second Node: Fix It

Proxmox NFS storage status unknown on the second node is almost always a mount issue. The first node added the NFS export fine, but the other node can’t reach it or doesn’t have the right options. Here’s the fix.

Why the Second Node Shows Unknown

When you add NFS storage in the Proxmox GUI, it writes the config to /etc/pve/storage.cfg and tries to mount it on the node you’re working on. Other nodes in the cluster pick up the config, but they don’t automatically mount the share. If the mount fails on a node, that node shows status: unknown. The storage is still defined, but Proxmox can’t see the actual NFS export.

Common causes: the NFS server doesn’t allow that node’s IP in its export list, the mount options are wrong, or the network path is broken. Sometimes it’s just a typo in the server path.

Check the Config and the Mount

First, look at /etc/pve/storage.cfg on the second node. You’ll see something like this:

nfs: mynfs
    export /mnt/tank/nfs
    path /mnt/pve/mynfs
    server 192.168.1.50
    content images,iso,backup
    options vers=4.2

Then try mounting it manually to see the actual error:

mount -t nfs 192.168.1.50:/mnt/tank/nfs /mnt/test

If it fails, you’ll get something like “access denied by server” or “connection timed out”. That tells you where to look. On the NFS server (like TrueNAS), check the export permissions. Make sure the second node’s IP is allowed, and that the export is set for the right NFS version.

The Fix

Once you’ve fixed the server side or the mount options, go back to the Proxmox GUI on the second node. Select the storage, click Edit, and just hit OK. That forces Proxmox to remount it. Or from the CLI:

pvesm set mynfs --server 192.168.1.50 --export /mnt/tank/nfs

That updates the config and triggers a remount. Check the status again — it should show active now.

If you’re using NFSv4, make sure the export on the server is set for NFSv4 and the mount option matches. Sometimes vers=4.2 causes issues if the server only supports 4.1. Drop it to vers=4 or vers=3 and see if it mounts.

My Take

This is a common cluster annoyance. The storage config is shared, but the mount isn’t. It’s not a bug, just how Proxmox handles NFS. Keep the export list on your NFS server up to date with all node IPs, and test from each node after adding storage. If you’re running TrueNAS, the export settings are under Sharing > NFS. Make sure “All dirs” is checked if you need subdirectory mounts.

For more on Proxmox and TrueNAS, check out Proxmox iSCSI Multipath: Setup with TrueNAS and Failover Test.