I have a strange issue with a new installation of Proxmox 7. I installed three servers that clustered just fine as a data center and with ceph. I went to set up NFS for the data center to connect to a NAS, and one of three that were installed from the same USB flash failed. Testing revealed that the NFS kernel module was missing, and a comparison of the working nodes suggested that there were more than a few kernel modules missing. I found no errors in the apt logs when I updated the problem machine. I'm running the latest 7.2-11 with kernel 5.15.64-1-pve that I installed and updated yesterday (11/10/2022). I know reinstallation might fix this problem, but I thought I would see if anyone had another suggestion. I'm going to look at what it takes to rebuild the Proxmox kernel modules, but that doesn't look like it's going to be easy. Any advice is welcome.
Missing kernel modules in new Proxmox 7 installation
- Thread starter J.C. Utter
- Start date
It would be better if you provided direct command output.
- pveversion -v
- uname -a
- showmount -e nfs_server
- showmount -a nfs_server
- mount nfs_server:/export_path /tmp
- any commands that you ran
- any errors you encountered
Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
- pveversion -v
- uname -a
- showmount -e nfs_server
- showmount -a nfs_server
- mount nfs_server:/export_path /tmp
- any commands that you ran
- any errors you encountered
Blockbridge : Ultra low latency all-NVME shared storage for Proxmox - https://www.blockbridge.com/proxmox
I started to get you the info you requested, but when I went to run modinfo nfs the nfs module is loaded now and the Proxmox web UI is no longer showing a ? over the NFS share. I basically worked on this problem for several hours this morning, and when I came back from lunch, the kernel modules that were missing are loaded. I've been working with Linux over 25 years now, and this is one of the strangest new installation issues I can remember. I've never seen a kernel module issue fix itself, so I'm still scratching my head.
While I don't like to reboot Linux to fix problems because it rarely fixes anything, I think that is exactly what happened. I rebooted the problem server, and the Proxmox UI still had a question mark over the NFS storage on the problem node, but I must have needed to refresh the browser. The reboot must have fixed it because it is indeed working now. I've never seen anything like it...