[MOAB-dev] Parallel Test Failure
Rajeev Jain
jain at mcs.anl.gov
Mon Oct 27 12:30:14 CDT 2014
Did all your tests pass?
Are you building hdf5 with mpi support?
This could be because due to serial hdf5. I see you have:
FAIL: parallel_hdf5_test
Rajeev Jain
630-252-3176 / 630-252-5986 (fax)
jain at mcs.anl.gov
On Monday, October 27, 2014 12:24 PM, "Grindeanu, Iulian R." <iulian at mcs.anl.gov> wrote:
Hi Friedrich,
What are you trying to do?
Rajeev and Vijay have experience running / building nek5000 with moab.
Iulian
________________________________
From: Grabner, Friedrich [F.M.Grabner at warwick.ac.uk]
Sent: Monday, October 27, 2014 12:17 PM
To: Grindeanu, Iulian R.
Subject: Re: Parallel Test Failure
Hi Iulian,
I have managed to install it although I believe there is some issue when installing on hpc.
However now when I try to run my moab enabled nek5000 I can only do so if I secure my resources on the system and then manually start my job.
nekmpi [casename] [np]
When I submit a job to the queue I receive the following error:
-bash: BASH_FUNC_module(): line 0: syntax error near unexpected token `)'
-bash: BASH_FUNC_module(): line 0: `BASH_FUNC_module() () { eval $($LMOD_CMD bash "$@");'
-bash: error importing function definition for `BASH_FUNC_module'
/home/eng/esumgy/.bashrc: line 29: module: command not found
mv: cannot stat `pipe.log.8': No such file or directory
mv: cannot stat `pipe.sch': No such file or directory
node6.1259Error re-mmapping shared memory: Cannot allocate memory (err=9)
node6.1259Error re-mmapping shared memory: Cannot allocate memory
[node6:01259] Open MPI detected an unexpected PSM error in opening an endpoint: Error re-mmapping shared memory: Cannot allocate memory
--------------------------------------------------------------------------
mpiexec has exited due to process rank 5 with PID 1259 on
node node6 exiting without calling "finalize". This may
have caused other processes in the application to be
terminated by signals sent by mpiexec (as reported here).
--------------------------------------------------------------------------
Could this be as I haven't correctly installed moab or any of it's prerequisites?
Regards Friedrich
________________________________
From: Grindeanu, Iulian R. <iulian at mcs.anl.gov>
Sent: 27 October 2014 16:54
To: Grabner, Friedrich
Cc: MOAB dev
Subject: RE: Parallel Test Failure
Hi Friedrich,
Thanks for your report.
I could not reproduce your error, so I don't know what is going on.
Are these the only errors you see in your make check?
Can you try a newer version of moab, or hdf5?
We have several nightly builds on our systems that resemble you configuration, for example this one:
http://gnep.mcs.anl.gov:8010/builders/moab-par
more configurations are here:
http://gnep.mcs.anl.gov:8010/waterfall?category=moab
thanks,
Iulian
________________________________
From: Grabner, Friedrich [F.M.Grabner at warwick.ac.uk]
Sent: Friday, October 24, 2014 8:03 AM
To: Grindeanu, Iulian R.
Subject: Re: Parallel Test Failure
Hi Iulian,
See attached!
Friedrich
________________________________
From: Grindeanu, Iulian R. <iulian at mcs.anl.gov>
Sent: 24 October 2014 13:14
To: Grabner, Friedrich; moab-announce at mcs.anl.gov
Subject: RE: Parallel Test Failure
can you send your config.log file?
Thanks,
Iulian
________________________________
From: moab-announce-bounces at mcs.anl.gov [moab-announce-bounces at mcs.anl.gov] on behalf of Grabner, Friedrich [F.M.Grabner at warwick.ac.uk]
Sent: Friday, October 24, 2014 4:22 AM
To: moab-announce at mcs.anl.gov
Subject: [MOAB-announce] Parallel Test Failure
Hi All,
I am trying to install moab-4.6.3 on my HPC system following the instructions on following webpage:
http://trac.mcs.anl.gov/projects/ITAPS/wiki/BuildingAndUsingMOAB
I have run everything exactly as specified on my desktop and successfully installed it however on hpc I receive error:
PASS: pcomm_unit
FAIL: parallel_unit_tests
PASS: uber_parallel_test
FAIL: scdtest
PASS: pcomm_serial
PASS: par_spatial_locator_test
FAIL: parallel_hdf5_test
PASS: mhdf_parallel
PASS: parallel_write_test
================================
3 of 9 tests failed
See test/parallel/test-suite.log
================================
make[4]: *** [test-suite.log] Error 1
make[4]: Leaving directory `/gpfs/home/eng/esumgy/moab-4.6.3/test/parallel'
make[3]: *** [check-TESTS] Error 2
make[3]: Leaving directory `/gpfs/home/eng/esumgy/moab-4.6.3/test/parallel'
make[2]: *** [check-am] Error 2
make[2]: Leaving directory `/gpfs/home/eng/esumgy/moab-4.6.3/test/parallel'
make[1]: *** [check-recursive] Error 1
make[1]: Leaving directory `/gpfs/home/eng/esumgy/moab-4.6.3/test
I have attached my test-suite.log file.
Thanks in advance, Friedrich
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/moab-dev/attachments/20141027/7ae872cc/attachment-0001.html>
More information about the moab-dev
mailing list