[molpro-user] Node hanging on h2o_select test job continued

[Dr Seth OLSEN]

Dear Molpro-Users,

After applying the CORLSI patch, I have determined that this bug must not be the reason for the hanging  h2o_select test job.  The node I am compiling (Intel 8.0) and running the package on is a v60x SunFire dual Xeon node with the Fedora Core 2 operating system.  I have not used the -mpp option in the configuration, therefore I am using a serial executable, not a parallel one (and I assume this means that the problem is not linked to GlobalArrays).  The symptomology is as follows.  When I run 'make test' the jobs h20_vdz.test and registry.test run without errors.  However, the execution of h2o_select never finishes.  If I background the test process and examine 'top' it is using most of the %CPU but otherwise does not appear to be doing anything.  The end of the h2o_select output file at this point is: Contracted 2-electron integrals neglected if value below      1.0D-12
 AO integral compression algorithm  1   Integral accuracy      1.0D-12
 
     0.262 MB (compressed) written to integral file (100.0%)
  
 
 NUMBER OF SORTED TWO-ELECTRON INTEGRALS:      13826.     BUFFER LENGTH:  32768
 NUMBER OF SEGMENTS:   1  SEGMENT LENGTH:      13826      RECORD LENGTH: 524288
 
 Memory used in sort:       0.57 MW
 

which leads me to believe that there is a write or read problem going on.  An examination of the disk space reveals that no filesystems are close to full.    Does anyone have any insight into this problem?

Cheers,

Seth Olsen


ccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccms

Dr Seth Olsen, PhD
Postdoctoral Fellow, Computational Systems Biology Group
Centre for Computational Molecular Science
Chemistry Building,
The University of Queensland
Qld 4072, Brisbane, Australia

tel (617) 33653732
fax (617) 33654623
email: s.olsen1 at uq.edu.au
Web: www.ccms.uq.edu.au 

ccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccmsccms