[Wien] error in LAPW1 (parallel calculation)

[Khuong P. Ong]

Dear Wien users,

  Could someone help me to overcome the following error:

     cycle 1     (Mon Jun  6 13:13:48 SGT 2005)  (40/40 to go)

 >   lapw0 -p    (13:13:48) starting parallel lapw0 at Mon Jun  6 13:13:49 
SGT 2005
--------
running lapw0 in single mode
33.120u 0.220s 0:35.84 93.0%    0+0k 0+0io 2323pf+0w
 >   lapw1  -up -p       (13:14:24) starting parallel lapw1 at Mon Jun  6 
13:14:24 SGT 2005
->  starting parallel LAPW1 jobs at Mon Jun  6 13:14:24 SGT 2005
running LAPW1 in parallel mode (using .machines)
8 number_of_parallel_jobs
**  LAPW1 crashed!
0.570u 0.920s 3:39:01.13 0.0%   0+0k 0+0io 122254pf+0w

 >   stop error

I checked the error files and obtained:
**  Error in Parallel LAPW1
**  LAPW1 STOPPED at Mon Jun 6 16:53:25 SGT 2005
**  check ERROR FILES!
  Cholesky INFO =  4903
  'SECLR4' - POTRF (Scalapack/LAPACK) failed.

This error happen at only one of eight machines with error

  Cholesky INFO =  4903
  'SECLR4' - POTRF (Scalapack/LAPACK) failed.

No error were reported for the last 7 machines.

This error sometime appears at Cycle 3, sometime at cycle 1 as in this case.

Many thanks for help.

Regards,

Khuong



-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://zeus.theochem.tuwien.ac.at/pipermail/wien/attachments/20050606/ba9a21d6/attachment.html