[SOLVED] PVE 7.4-3 - Scanning for NFS share doesn't return a mount path

[patrick.reinboth]

Hi all!

I am currently having trouble in mounting a NFS share I created on a Synology in PVE 7.4-3.
The scan for the external paths returns nothing, but I am 100% sure that the NFS share is configured correctly on the synology as I have other services already connected to NFS shares there.

I even tried to mount the share directly via cli on the server to see if there maybe is a connection issue via

mount -t nfs IP:/volume1/pve_test /mnt/pve_test

but that works just fine! Adding it directly in

/etc/pve/storage.cfg

didn't work though.

I also tried to run

pvesm nfsscan IP

but that just returned

clnt_create: RPC: Timed out

Do you have any ideas what the issue might be in this case?

 

[bbgeek17]

What happens when you run:
/sbin/showmount --no-headers --exports [nfs_ip]

---

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

 

[patrick.reinboth]

What happens when you run:
/sbin/showmount --no-headers --exports [nfs_ip]

---

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

Then I get the following output:

clnt_create: RPC: Unknown host

 

[bbgeek17]

The usual Linux convention is that a string inside brackets, not prefaced with single or double dash, needs to be replaced with an appropriate value. In your case an IP of the NFS server...

Square Brackets [ ]

The square brackets ( [ ] ) indicate that the enclosed element (parameter, value, or information) is optional. You can choose one or more items or no items. Do not type the square brackets themselves in the command line.

---

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

 

[patrick.reinboth]

The usual Linux convention is that a string inside brackets, not prefaced with single or double dash, needs to be replaced with an appropriate value. In your case an IP of the NFS server...

Square Brackets [ ]

The square brackets ( [ ] ) indicate that the enclosed element (parameter, value, or information) is optional. You can choose one or more items or no items. Do not type the square brackets themselves in the command line.

---

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

The more you know.
In that case, I get the same timeout message I got before by running pvesm nfsscan:

clnt_create: RPC: Timed out

 

[bbgeek17]

Can you run this command from any other host?

In summary, Proxmox is using native Linux (Kernel and Userland) NFS functionality. It uses basic probing commands, such as showmount and rpcinfo to confirm that NFS storage is available. In your case these commands fail (time out). There could be many reasons:
- mtu mismatch
- bad cable / packet loss
- firewall at pve/switch/synology
- misconfiguration of synology
just to name a few.
It can be debated whether the particular probing that PVE does is universal enough, but it is what it is today. Until you have showmount and rpcinfo working you will continue having issues at PVE level as it relies on them.

---

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

 

[patrick.reinboth]

Can you run this command from any other host?

In summary, Proxmox is using native Linux (Kernel and Userland) NFS functionality. It uses basic probing commands, such as showmount and rpcinfo to confirm that NFS storage is available. In your case these commands fail (time out). There could be many reasons:
- mtu mismatch
- bad cable / packet loss
- firewall at pve/switch/synology
- misconfiguration of synology
just to name a few.
It can be debated whether the particular probing that PVE does is universal enough, but it is what it is today. Until you have showmount and rpcinfo working you will continue having issues at PVE level as it relies on them.

---

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

Thank you for your answer which pointed me into the right direction!
I tried the same command on a different host, which had active NFS mounts and I got a timeout there as well. So I checked the synology again and saw that the firewall dropped port 111.

We didn't need it open till now, but your explanation, helped me to understand while it is needed in this case.