[systemd-devel] failing boot start jobs delay reboot

[Lennart Poettering]

On Wed, 28.01.15 15:12, Michael Biebl (mbiebl at gmail.com) wrote:

> 2015-01-28 2:05 GMT+01:00 Lennart Poettering <lennart at poettering.net>:
> > On Tue, 20.01.15 11:24, Andrei Borzenkov (arvidjaar at gmail.com) wrote:
> >> - stop in emergency shell when mount fails
> >> - fix /etc/fstab
> >> - ^D to continue boot
> >>
> >> no more works, because /etc/fstab is not reevaluated so systemd will
> >> still try the same broken mount again.
> >
> > Yeah, systemd will require a "systemctl daemon-reload" first. It might
> > be a good idea if distros would mention that in a comment in the
> > default fstab file they install.
> 
> 
> Hm, I vaguely remember that we discussed quite a while ago. Wasn't the
> idea to trigger the daemon-reload automatically when exiting the
> emergency/rescue shell?
> 
> Could be as simple as changing emergency.service from
> 
>  ExecStart=-/bin/sh -c "/sbin/sulogin; /bin/systemctl --fail --no-block default"
> 
> to
>  ExecStart=-/bin/sh -c "/sbin/sulogin; /bin/systemctl daemon-reload;
> /bin/systemctl --fail --no-block default"

I am not convinced this would be a good idea. I mean, I fail to see
why we should add an automatism in this specific case while we don't
have any automatism when you edit the file in any other context, for
example when the system is fully up.

I think this is really a documentation issue: the fstab file should
contain a nice hint to run "systemctl daemon-reload" to make systemd
consider it. And we could even update the rescue mode help text that
is shown right before sulogin to say the same. But otherwise, I am not
convinced anything should happen automatically here...

Lennart

-- 
Lennart Poettering, Red Hat