[Debichem-devel] Bug#680397: Bug#680397: Bug#680397: aces3: FTBFS: Times out

[Michael Banck]

Hi,

On Fri, Jul 06, 2012 at 01:46:10PM +0200, Michael Banck wrote:
> On Thu, Jul 05, 2012 at 06:50:40PM +0200, Kurt Roeckx wrote:
> > Source: aces3
> > Version: 3.0.6-4
> > Severity: serious
> > 
> > Your package fails to build on a few arches, timing out in the
> > test suite:
> > (cd tests; OMPI_MCA_plm_rsh_agent=/bin/false csh ./runscript)
> > Running 1.1.1.1
> > mpirun: killing job...
> > 
> > E: Caught signal 'Terminated': terminating immediately
> > --------------------------------------------------------------------------
> > mpirun noticed that process rank 0 with PID 9855 on node murphy
> > exited on signal 0 (Unknown signal 0).
> > --------------------------------------------------------------------------
> > mpirun: clean termination accomplished
> > 
> > make: *** [build-arch] Terminated
> > make[1]: *** [override_dh_auto_test] Terminated
> > Build killed with signal TERM after 150 minutes of inactivity
> 
> On i386, this might indeed be a problem we have to investigate. On arm*
> I would guess the autobuilders are just too slow to finish the test
> suite on time for now.  Unfortunately, the ACESIII developers really
> used rather heavy stuff for their test suite.

I have confirmed this on armhf (harris.d.o's unstable chroot) at least
now.  The test suite is running, and a modified (much smaller) testcase
finishes fine.

So I assume this is a machine-dependent issue on the arm* autobuilders
for now.

On my home i386 notebook however, the first testcase starts up fine but
then appears to spin at 100% CPU without a lot of memory usage way too
early in the job.


Michael