January 30, 20251 yr Good morning, I recently moved my UnRAID from a HP DL380g9 to an Optiplex 3080 SFF. With the old hardware, every once and a while, NFS would seem to hang (mounts would become unreadable on a guest) df -h would hang and data would be inaccessible until UnRAID rebooted UnRAID (10.0.0.20) - Optiplex 3080 PLEX (10.0.0.55) - Proxmox Ubuntu 24.04 LTS VM (Optiplex 3010 i5-13500T) with iGPU passthrough) Since the migration to the 3080, NFS seems to hang every day... frequently. If I attempt to stop the array, it gets to the following and just hang Jan 29 15:21:44 Tower emhttpd: shcmd (1073): /etc/rc.d/rc.nfsd stop Jan 29 15:21:44 Tower rc.nfsd: Stopping NFS server daemon... Jan 29 15:21:44 Tower rc.nfsd: /usr/sbin/rpc.nfsd 0 Diag attached... Happy to share the output of any other commands needed. tower-diagnostics-20250130-0907.zip Edited January 30, 20251 yr by Soup
January 30, 20251 yr Author Thanks, I've increased the number of threads and the number of fuse file descriptors already (16 threads and 500,000 file descriptors). I have no idea if those numbers are appropriate, is there a way to tell if those values are set too low or harm to setting them too high?
January 30, 20251 yr Author The issue just happened again. Host1-VM1 lost access to NFS share Host2-VM2 still had full access to a NFS share Host2-VM2 unmount NFS share, unable to remount NFS share... log only shows Jan 30 14:23:36 Tower rpc.mountd[18574]: v4.2 client attached: 0x33d49064679ae870 from "10.0.0.53:904" Attempted to restart NFS root@Tower:~# /usr/local/etc/rc.d/rc.nfsd restart rc.nfsd: Restarting NFS server daemon... rc.nfsd: Stopping NFS server daemon... Hangs at this.
January 31, 20251 yr Author I changed the fstab entry on my plex server from: 10.0.0.20:/mnt/user/Movies /mnt/user/Movies nfs defaults 0 0 to: 10.0.0.20:/mnt/user/Movies /mnt/user/Movies nfs hard,timeo=50,retrans=5,relatime,rsize=1048576,wsize=1048576 0 0 Have not had a reoccurance of the issue since doing that... will continue to monitor If that does indeed end up being the solution, I'd love for someone (anyone) to try and explain it to me. These NFS options are kinda over my head, got it from the link posted by @JorgeB
January 31, 20251 yr 35 minutes ago, Soup said: I changed the fstab entry on my plex server from: 10.0.0.20:/mnt/user/Movies /mnt/user/Movies nfs defaults 0 0 to: 10.0.0.20:/mnt/user/Movies /mnt/user/Movies nfs hard,timeo=50,retrans=5,relatime,rsize=1048576,wsize=1048576 0 0 Have not had a reoccurance of the issue since doing that... will continue to monitor If that does indeed end up being the solution, I'd love for someone (anyone) to try and explain it to me. These NFS options are kinda over my head, got it from the link posted by @JorgeB I recommend you use UD to manage your remote NFS shares. It is optimized to handle the issues you are seeing.
January 31, 20251 yr Author 7 minutes ago, dlandon said: I recommend you use UD to manage your remote NFS shares. It is optimized to handle the issues you are seeing. Sorry for the stupid question... but what's UD?
January 31, 20251 yr 3 minutes ago, Soup said: I'm mounting the NFS share on a Ubuntu VM. The NFS server is UnRAID. Sorry, I assumed that the fstab entry was on Unraid.
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.