RE: AMBER: sander MPI fails included tests

From: Sergio Wong <swong.mccammon.ucsd.edu>
Date: Mon, 25 Jun 2007 12:06:21 -0700 (PDT)

HI;

>> forrtl: severe (174): SIGSEGV, segmentation fault occurred
>> While performing the test under
>> AMBERHOME/test/tgtmd/change_target.ntr.
>> It was configured : ./configure linux_em64t ifort mpich_gm pubfft
>
> This is pmemd 9.0 yes?
>
> The configure line you give above is for pmemd (not sander) but then the
> targetted MD test case will not run with PMEMD since it is not supported...
> So I am confused what you actually ran here.

My apologies. You're correct, that's the config line for pmemd. I am,
however, testing sander.MPI (version 9). The configure command was:

# Generated via command: ./configure -mpich -static ifort_x86_64

> What did you have DO_PARALLEL set to?

export DO_PARALLEL="mpirun -np 2 "

> What did you have TESTsander set to (if at all)?

I didn't set TESTsander

> What commands did you issue to run the test cases?

make test.parallel
I also manually ran the Run.** script from the appropriate directory
and got the same error.

> Did you compile sander in parallel (sander.MPI etc...)? As well as pmemd?

At that time I only had compiled sander.MPI

Thanks!!!

-Sergio
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber.scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo.scripps.edu
Received on Wed Jun 27 2007 - 06:07:23 PDT
Custom Search