hanging processes, nfs, automount

[Ronald Klop]

On Tue, 31 Mar 2015 10:24:05 +0200, Dominic Fandrey <kamikaze at bsdforen.de>  
wrote:

> I don't know where to ask this, so I'm putting it out here.
>
> Introductions first:
> FreeBSD AprilRyan.norad 10.1-STABLE FreeBSD 10.1-STABLE #2 r280373: Mon  
> Mar 23 17:43:09 CET 2015      
> root at AprilRyan.norad:/usr/obj/S403/amd64/usr/src/sys/S403  amd64
>
> I have an nfs-share that I access via automount in /net/<host>/<share>.
>
> The following scenario:
> - cd /net/<host>/<share> inside tmux
> - Do stuff
> - S4 the laptop and take it to work the next morning
> - Resume and the tmux session hangs infinitely
> - Kill all shells belonging to the tmux session
> - tmux server process hangs in state "rpcrec"
>
> It's bad enough that there apparently is no path for NFS to fail
> and reanimate the shell that accesses the share.
>
> But the really bad problem is, that tmux in rpcrec state is not
> killable. I.e. not kill -9 able. And as long as that process is around
> any new tmux server I try to start just hangs.

For this killable stuff there is the 'intr' mount option. See also the  
'soft' mount option if you don't want NFS to wait indefinitely for a lost  
server.
man mount_nfs describes this all.

Regards,
Ronald.


>
> My only recourse so far is a reboot.