On Tuesday 21 July 2015 23:09:10 Marcus Meissner wrote:
On Tue, Jul 21, 2015 at 07:30:15PM +0200, Jan Engelhardt wrote:
On Tuesday 2015-07-21 19:08, Adrian Schröter wrote:
[ 26s] [ 5.482103] mce: Unable to init device /dev/mcelog (rc: -5) [ 26s] [ 5.556674] systemd: segfault at 8 ip 00007f3e5d6a4793 sp 00007ffe80f3b4f8 error 4 in libc-2.21.so[7f3e5d626000+199000] [ 26s] [ 5.559622] systemd: Caught <SEGV>, dumped core as pid 69. [ 26s] [ 5.559773] systemd: Freezing execution. [ 325s] ### WATCHDOG TRIGGERED, KILLING VM ###
Hm, you have also an own systemd in your project. So there are good chances that this one is breaking here.
You need to debug this using a local "osc build --vm-type=kvm" build IMHO.
No idea how to debug systemd though, I suppose starting it via a gdb is not working?>
Well, you can start systemd via systemd-nspawn, though that is not exactly the same as a domU.
Staging:E is actually the systemd version upgrade. Jan or Cristian need to debug it.
I actually had two concerns here:
1. Broken package, in this case systemd Happens, but will eventually be dealt with by the package maintainers
2. Stuck worker About 20 workers were in a busy state for about 8 hours, without doing anything useful. I would have expected the build process to fail and the worker commencing with the next workload.