Cannot kill lzop (NFS) - Proxmox 6

[filipealvarez]

Preventing NFS stupidity
Mounting an NFS filesystem with the soft option will help prevent stuck processes when a network connection is lost.

showmount -e remote_nfs_server
mount remote_ls
mount -o soft nfs_server:/path /media/mount_point

Credit: http://www.noah.org/wiki/Kill_-9_does_not_work

Can this help with stuck of lzop when problem occours with NFS mount in proxmox?

Proxmox Version: 6.0-7
Linux 5.0.21-2-pve #1 SMP PVE 5.0.21-3 (Thu, 05 Sep 2019 13:56:01 +0200) x86_64 GNU/Linux

# cat /proc/mounts|grep nfs
192.168.1.101:/nfs /mnt/pve/nfs-server nfs4 rw,relatime,vers=4.2,rsize=1048576,wsize=1048576,namlen=255,hard,proto=tcp,timeo=600,retrans=2,sec=sys,clientaddr=192.168.1.2,local_lock=none,addr=192.168.1.101 0 0

root 27189 0.0 0.0 5216 1736 ? D 00:14 0:00 lzop


Tried to kill without sucess.

Can I try umount -l nfs-mount-point ?

Can this umount cause instability to Proxmox ? (Its a production server)

Thank you

 

[Stoiko Ivanov]

hanging NFS-mounts are indeed very unpleasant and cause quite severe problems on the NFS-Clients (the PVE-node in this case).
a NFS server which becomes unreachable causes processes, which do I/O to that export to go into uninterruptible sleep state (D+) - usually you need to reboot the NFS-Client in those cases (although sometimes it works out if the NFS server becomes reachable again).

while using soft as mount-option mitigates these issues - the man page (`man nfs` ) is quite explicit about the implications:

NB: A so-called "soft" timeout can cause silent data corruption in certain
cases. As such, use the soft option only when client responsiveness is more im‐
portant than data integrity. Using NFS over TCP or increasing the value of the
retrans option may mitigate some of the risks of using the soft option.

so use it at your own risk

I hope this helps!