[mvapich-discuss] PMGR_COLLECTIVE ERROR: MPIRUN_RANK

Sayantan Sur surs at cse.ohio-state.edu
Fri May 14 11:10:18 EDT 2010


Hi Senthil,

On Fri, May 14, 2010 at 10:08 AM,  <senthil.natesan at ndsu.edu> wrote:
>
> Dear MVAPICH developers/users,
>
> I compiled Amber PMEMD (molecular dynamics) program (www.ambermd.org)with
> MVAPICH-1.1.0 successfully, but when I am submitting the job, it gives the
> following error for every node.
>
> PMGR_COLLECTIVE ERROR: unitialized MPI task: Missing required environment
> variable: MPIRUN_RANK
>
> My system is craycx1 with RHEL 5.3 (x86_64), with both ethernet and
> infiniband hybrid cables/switches. (6 nodes, each node intel xeon Quad
> processors).
>

Thanks for the report. Just checking - did you compile MVAPICH with
the ch_gen2 interface or sockets? If you did compile MVAPICH with
sockets, could you recompile with ch_gen2?

It could be that you have compiled Amber with one version of MVAPICH y
and trying to run it with another.

MVAPICH-1.1.0 version is old, may we request you go update to later
MVAPICH-1.2-rc1 or even better MVAPICH2-1.4.1.


MVAPICH2-1.4.1 user guide:
http://mvapich.cse.ohio-state.edu/support/user_guide_mvapich2-1.4.1.html
MVAPICH-1.2-rc1 user guide:
http://mvapich.cse.ohio-state.edu/support/user_guide_mvapich2-1.5rc1.html

> Any suggestions would be appreciated.
>
> Thanks,
>
> Senthil Natesan
>
> _______________________________________________
> mvapich-discuss mailing list
> mvapich-discuss at cse.ohio-state.edu
> http://mail.cse.ohio-state.edu/mailman/listinfo/mvapich-discuss
>
>



-- 
Sayantan Sur

Research Scientist
Department of Computer Science
The Ohio State University.


More information about the mvapich-discuss mailing list