[molpro-user] matrop problem (again)

Glen Jenness grj3+ at pitt.edu
Tue Aug 21 17:16:15 BST 2007

Dear Users,
I'm having a recurring problem with the Matrop utility in Molpro2006.1
Here's a brief description of the problem:  Whenever I try to run the
program in parallel, Matrop will start, but then stop and will run forever
unless I kill it (this step should only take a couple seconds, yet I've
had it run over several days to see if it would clear up on it's own. 
When it's obvious that it's not doing anything (and never will), I kill
the calculation, upon which it prints:


1:Terminate signal was sent, status=: 15
  1: ARMCI aborting 15 (0xf).
tmp =
 Creating: host=node068, user=glen,

 Recover dump record  2100.2  Wavefunction type=RKS from file scw_a.wfu

 Copying EIG/CANONICAL  Set=1
 Copying DENSITY/CHARGE  Set=1
 Copying FOCK/TOTAL  Set=1
 Copying FOCK/OPEN  Set=2
 Copying GROUP/RHF  Set=1
 Copying OCC/RHF  Set=1
0:SigIntHandler: interrupt signal was caught: 2
  0: ARMCI aborting 2 (0x2).

So, when I kill the job, it reads in the information after the terminate
signal was sent!
Furthermore, when I specify more than one processor in my pbs script for
our queue (but the program is still working in serial), the same problem
still persists!
I have installed a patch I was informed about that supposedly fixes this
problem (patch 39 for bug1264), and even with this patch, the problem
I've attached a tar ball with all my inputs (unfortunately due to size
restraints I couldn't send the wfu files) so if anyone wants to have a
look, or maybe run it on their machine to see if it's a problem on our end
or if it's a problem in the code, they can.  The output included in the
tar ball however, doesn't list this patch ever being installed (even
though it was), so this patch was applied to a backup copy of the code and
was ran that way-however the outputs between the two are identical, so
I've only included the one.
This is the third time I've posted about this problem, and I would greatly
appreciate any help.
Thank you in advance,
Glen Jenness
Jordan Group
Department of Chemistry
University of Pittsburgh
-------------- next part --------------
A non-text attachment was scrubbed...
Name: matrop_prob.tar.gz
Type: application/x-gzip-compressed
Size: 10190 bytes
Desc: not available
URL: <http://www.molpro.net/pipermail/molpro-user/attachments/20070821/73eefe8d/attachment.bin>

More information about the Molpro-user mailing list