atd stopped starting on boot after 11.3 -> 12.2 upgrade

English Install/Boot/Login
1

Title says it all… I use ‘atd’ pretty regularly, and had it starting in runlevels 3 and 5. After upgrading to 12.2, I noticed it was not starting, even though it was still selected for those runlevels. If I tried to start it by hand, I got:

/etc/init.d/atd start

redirecting to systemctl

but then:

/etc/init.d/atd status

redirecting to systemctl
atd.service - Execution Queue Daemon
Loaded: loaded (/lib/systemd/system/atd.service; disabled)
Active: failed (Result: timeout) since Wed, 24 Apr 2013 20:56:19 -0500; 1min 56s ago
Process: 3595 ExecStart=/usr/sbin/atd (code=exited, status=0/SUCCESS)
CGroup: name=systemd:/system/atd.service

so it looks like something in systemd is disabling atd.

I can run ‘startproc /usr/sbin/atd’, and it runs and works OK, so there does not appear to be anything wrong with the underlying daemon, just something in the way that systemd calls or interacts with it. I am not familiar enough with systemd to know how best to debug or work around this problem (other than just starting atd manually on boot).

Any suggestions on getting atd working automatically again would be appreciated.

2

Try enabling the service

systemctl enable atd.service

It should then be started next boot.

3

Well, this is weird. I did the following:

systemctl enable atd.service

and rebooted the machine. I saw /usr/sbin/atd in the process tree after boot. Yay!

But after about 2 minutes, **systemd kills it! ** From /var/log/messages:

Apr 24 21:52:41 grond systemd[1]: Reloading.
Apr 24 21:56:03 grond atd[2598]: Removing stale lockfile for pid 720
Apr 24 21:57:33 grond systemd[1]: atd.service stopping timed out. Killing.
Apr 24 21:57:33 grond systemd[1]: Unit atd.service entered failed state.

systemctl status atd.service

atd.service - Execution Queue Daemon
Loaded: loaded (/lib/systemd/system/atd.service; enabled)
Active: failed (Result: timeout) since Wed, 24 Apr 2013 21:57:33 -0500; 3min 14s ago
Process: 2596 ExecStart=/usr/sbin/atd (code=exited, status=0/SUCCESS)
CGroup: name=systemd:/system/atd.service

Apr 24 21:56:03 grond atd[2598]: Removing stale lockfile for pid 720

The same thing happens if I run systemctl start atd.service by hand… apparent success, followed several minutes later by systemd shutting atd down. When I start atd on its own, it runs indefinitely. What the hell is systemd doing?

4

Well, there was an error in atd’s service file on 12.2:
https://bugzilla.novell.com/show_bug.cgi?id=780259
But that should be fixed and released as online update since December!

This looks exactly like comment #8 in that bug.
Please check that you have the latest at package from the update repo installed!

To fix it, do:

sudo zypper in -f at-3.1.8-1110.4.1
5

Yes, that appears to do the trick. Thanks, all.

One last question: how did you find that bug report? When I searched for ‘atd’ ( https://bugzilla.novell.com/buglist.cgi?quicksearch=atd ) I got 3 results, but not this one - and none on this topic.

Thanks again–

6

The quick search only looks for open bugs, but this one is marked as “RESOLVED FIXED”, so it wasn’t found.
You would have to go to the Advanced search and additionally select “RESOLVED” in the status list.

Btw, my search terms have been: “atd not start”…

7

On 2013-04-25 04:06, dcfleck2 wrote:>
> Title says it all… I use ‘atd’ pretty regularly, and had it starting
> in runlevels 3 and 5. After upgrading to 12.2, I noticed it was not
> starting, even though it was still selected for those runlevels. If I
> tried to start it by hand, I got:

On 2013-04-25 14:16, dcfleck2 wrote:
>
> Yes, that appears to do the trick. Thanks, all.

This means that your upgrade procedure was incorrect, you did not apply
the updates.

–
Cheers / Saludos,

Carlos E. R.
(from 12.1 x86_64 “Asparagus” at Telcontar)