The ABINIT Users Mailing List ( CLOSED )

[PGanesh <pganesh@ciw.edu>]

Dear All,

I was previously running v5.4.2 on our dual core AMD opteron
cluster.  Lately I downloaded the v5.4.4 seq.+parall. binary.  The 
parallel jobs end right after this step with a seg. fault message:

ITER STEP NUMBER     1
vtorho : nnsclo_now=  2, note that nnsclo,dbl_nnsclo,istep=  0 0  1

and even before this step it prints out the following lines:

-P-0000 
================================================================================
-P-0000 == DATASET  1 
==================================================================
-P-0000
dtsetcopy : copying area  algalch    the actual size (^C^@^@^@) of the 
index (^A^@^@^@)  differs from its standard size (^@^@^@^@)
dtsetcopy : copying area  kberry     the actual size (^T^@^@^@) of the 
index (^B^@^@^@)  differs from its standard size (^A^@^@^@)
dtsetcopy : copying area  nband      the actual size (^R^@^@^@) of the 
index (^A^@^@^@)  differs from its standard size (^A^@^@^@)
 dtsetcopy : allocated densty= T
dtsetcopy : copying area  mixalch    the actual size (^C^@^@^@) of the 
index (^A^@^@^@)  differs from its standard size (^@^@^@^@)
dtsetcopy : copying area  mixalch    the actual size (^C^@^@^@) of the 
index (^B^@^@^@)  differs from its standard size (^@^@^@^@)
dtsetcopy : copying area  occ_orig   the actual size (<88>^B^@^@) of the 
index (^A^@^@^@)  differs from its standard size (<9E>^A^@^@)
dtsetcopy : copying area  shiftk     the actual size ^@^@^@) of the 
index (^B^@^@^@)  differs from its standard size (^A^@^@^@)

Sequential jobs print out the same message as above but run smoothly 
afterwards.

I don't understand this problem.  Does anyone have an idea of why 
parallel jobs are dying?

Thank you very much.

P. Ganesh