[systemd-devel] systemd: Program terminated with signal SIGFPE, Arithmetic exception.
paulepanter at users.sourceforge.net
Mon Jan 5 14:38:42 PST 2015
Am Montag, den 05.01.2015, 17:57 +0100 schrieb Lennart Poettering:
> On Tue, 30.12.14 16:51, Paul Menzel (paulepanter at users.sourceforge.net) wrote:
> > using Debian Sid/unstable with systemd 215-8, I still have problems that
> > there are time-outs between systemd and D-Bus, even with the current
> > D-Bus 1.8.12 . Experiencing this again, where
> > `systemd-logind.service` was not started, I was able to log in on tty1
> > but in the end I was unable to run any `sudo systemctl` commands as they
> > timed out. I also was unable to reboot or halt the system.
> > Somewhere in that situation systemd also crashed with an arithmetic
> > exception; logged as `Caught <FPE>` in `/var/log/syslog`.
> > I created ticket #87349 in the Freedesktop.org Bugzilla bug tracker .
> > The backtrace is pasted in the original upstream bug report and in this
> > message at the end.
> Is there any known, easy way to reproduce the issue?
Sorry, I do not know of one and I was not able to reproduce it another
> Is it possible that something in your boot causes the systemd
> configuration to be reloaded?
Sorry, I do not know. Any idea, how I can check that?
> I now made some changes to git
> (9c3349e23b14db27e7ba45f82cf647899c563ea9) that add an explicit
> assert() around the line where the devision by zero happens. Most
> likely the counting of running jobs got confused, hence I reworked
> that code too. Not sure if it fixes the issue, but I can't test this,
> since I have no idea how to trigger it...
> Anyway, would be good if you could check current git, to see if things
> are better for you...
Thank you. Do you think it would be good for distributions to backport
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 181 bytes
Desc: This is a digitally signed message part
More information about the systemd-devel