AMBER Archive (2008)

Subject: RE: AMBER: AMBER10: QMMM parallel test error

From: Ross Walker (ross_at_rosswalker.co.uk)
Date: Wed Nov 05 2008 - 19:41:27 CST


Hi Yang,

Can you give me some more info here please. What version of mvapich are you
using? What compilers and which versions? What options did you pass to
configure_amber?

Also are you using MKL? If so then which version (exactly) is it?

Could you also please go into the test/qmmm2/1NLN_test_diagonalizers
directory and send me the output file / partial output file produced by the
Run.1NLN_dsyevr script.

All the best
Ross

> -----Original Message-----
> From: owner-amber_at_scripps.edu [mailto:owner-amber_at_scripps.edu] On Behalf
> Of Yang, Ping
> Sent: Wednesday, November 05, 2008 3:44 PM
> To: amber_at_scripps.edu
> Subject: AMBER: AMBER10: QMMM parallel test error
>
> Greetings,
>
> I compiled the serial version and parallel version of Amber using mvapich
> and intel compiler for Barcelona processors with no error.
> The test for serial version passed smoothly along with the
> "test.parallel". However, the "test.parallel.QMMM" quited with the
> following error message:
>
>
> ==============================================================
> cd qmmm2/1NLN_test_diagonalizers && ./Run.1NLN_dsyevd
> diffing mdout.1NLN_dsyevd.save with mdout.1NLN_dsyevd
> PASSED
> ==============================================================
> cd qmmm2/1NLN_test_diagonalizers && ./Run.1NLN_dsyevr
> forrtl: severe (174): SIGSEGV, segmentation fault occurred
> Image PC Routine Line Source
> libpthread.so.0 000000353C70C4F0 Unknown Unknown
> Unknown
> libmkl_lapack.so 0000002A9B4C631F Unknown Unknown
> Unknown
> rank 0 in job 150 cu0login1_55120 caused collective abort of all ranks
> exit status of rank 0: killed by signal 9
> ./Run.1NLN_dsyevr: Program error
> make[1]: *** [test.sander.QMMM] Error 1
> make[1]: Leaving directory `/home/amber/amber10/amber10/test'
> make: *** [test.sander.QMMM.MPI] Error 2
>
>
> I'd appreciate any help to solve this problem and please let me know if
> further information is needed for diagnosis.
>
> Thank you,
>
> -Ping
>
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
> to majordomo_at_scripps.edu

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" (in the *body* of the email)
      to majordomo_at_scripps.edu