Bug#800707: systemd: Sytem hangs on shutdown when nfs-shares are mounted via autofs

[Jürgen Bausa]

Am Dienstag, 12. März 2019, 03:49:44 CET schrieb Michael Biebl:
> Control: tags -1 + moreinfo
> 
> On Fri, 02 Oct 2015 21:01:07 +0200 =?utf-8?q?J=C3=BCrgen_Bausa?=
> 
> <Juergen.Bausa at online.de> wrote:
> > Package: systemd
> > Version: 215-17+deb8u2
> > Severity: normal
> > 
> > Dear Maintainer,
> > 
> > I am running jessie on a laptop with wlan (Networkmanager) and want to use
> > autofs to mount nfs shares, as the shares are only available, when I am at
> > home.
> > 
> > The problem is, that when I am actually accessing the share (and it is
> > therefore mounted by autofs), the system hangs on shutdown. At first I get
> > the message
> > 
> >   [*****]A stop job is running for /mnt/nfs/share (xxs / 1mn 30s)
> > 
> > and the systems counts down 90 s. After that I see som more messages which
> > end up with
> > 
> >   Reached target shutdown
> > 
> > After that, the system hangs forever (at least for a long time, I think I
> > waited some minutes).
> > 
> > When I am not accessing an nfs share during the session, and autofs does
> > not need to actually mount it, shutdown works fine.
> > 
> > I think the reason for this problem is, that systemd stops Networkmanager
> > before autofs had a chance to unmount the shares. After that unmounting
> > doesnt work and later stopping autofs also fails. Thats my theory.
> 
> Is this still an issue with stretch or buster?
> Can you share more details about your network configuration, (wifi,
> ethernet,...)

I am running stretch now and it is not an issue anymore. As I remember I found 
some workaround for jessie. Most lilkely decreasing some timeouts, but I am 
not sure.

At the time of reporting the bug I used jessie with network-manager, and a wifi 
connection.

Juergen