AMBER Archive (2003)

Subject: AMBER: a note on problems with MPI and sander

From: David A. Case (case_at_scripps.edu)
Date: Sat Jul 26 2003 - 13:50:37 CDT

>From time to time (including just recently) there are user reports along
the lines: "MPI_COMM_WORLD rank 0 has terminated without calling
MPI_Finalize()" [or similar messages from other MPI implementations].

I just want to point out that this can happen if there is an error in
the input (e.g. an unrecognized namelist variable). In many cases, the
output file will be empty, (never flushed), so that there is no way to see
the error message.

So: be sure your input works with a non-MPI version of sander. Also, run
the test cases using MPI, since these are known to have good inputs. If you
have run the test suite, make a note of that in your problem report,
particularly for MPI problems; that will make it easier for us to decide
if you are seeing an installation problem, or a problem with the input files.

Of course, there can be other reasons for MPI failures, but this is one
thing to check for first.

..good luck...dac 

-- 

==================================================================
David A. Case                     |  e-mail:      case_at_scripps.edu
Dept. of Molecular Biology, TPC15 |  fax:          +1-858-784-8896
The Scripps Research Institute    |  phone:        +1-858-784-9768
10550 N. Torrey Pines Rd.         |  home page:                   
La Jolla CA 92037  USA            |    http://www.scripps.edu/case
==================================================================
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu