[mpich-discuss] mpdman: invalid attempt to open 2 simultaneous pmi connections

[Richard Jacobsen]

Hello everyone,

I'm trying to port Parflow (a groundwater flow model) to OS X/mpich2 and 
I'm encountering a few errors when running it that I can't decode, was 
wondering if someone could point me in the right direction...

Parflow has a tcl script which configures various things for the model 
and I have it eventually start up:
mpirun -bnr -machinefile ./.hostfile -np 4 
~/src/parflow/parflow/bin/parflow ow

mpdman: invalid attempt to open 2 simultaneous pmi connections
  client=/Users/elcangrejo/src/parflow/parflow/bin/parflow  
cwd=/Users/elcangrejo/src/parflow/parflow.darwin/src/test/ow10
mpdman: invalid attempt to open 2 simultaneous pmi connections
  client=/Users/elcangrejo/src/parflow/parflow/bin/parflow  
cwd=/Users/elcangrejo/src/parflow/parflow.darwin/src/test/ow10
mpdman: invalid attempt to open 2 simultaneous pmi connections
  client=/Users/elcangrejo/src/parflow/parflow/bin/parflow  
cwd=/Users/elcangrejo/src/parflow/parflow.darwin/src/test/ow10
mpdman: invalid attempt to open 2 simultaneous pmi connections
  client=/Users/elcangrejo/src/parflow/parflow/bin/parflow  
cwd=/Users/elcangrejo/src/parflow/parflow.darwin/src/test/ow10
mpirun -bnr -machinefile ./.hostfile -np 4 
/Users/elcangrejo/src/parflow/parflow/bin/parflow ow
rank 3 in job 13  vh242-fogg2.lawr.ucdavis.edu_52093   caused collective 
abort of all ranks
  exit status of rank 3: killed by signal 9
rank 2 in job 13  vh242-fogg2.lawr.ucdavis.edu_52093   caused collective 
abort of all ranks
  exit status of rank 2: killed by signal 9
rank 1 in job 13  vh242-fogg2.lawr.ucdavis.edu_52093   caused collective 
abort of all ranks
  exit status of rank 1: killed by signal 9
rank 0 in job 13  vh242-fogg2.lawr.ucdavis.edu_52093   caused collective 
abort of all ranks
  exit status of rank 0: killed by signal 9

Various important versions:
uname: 9.2.2 Darwin Kernel Version 9.2.2: Tue Mar  4 21:17:34 PST 2008; 
root:xnu-1228.4.31~1/RELEASE_I386 i386
mpich2-1.0.7
gcc-4.3.1

I can compile and run cpilog/etc just fine... Googling the error message 
that mpdman spits out doesn't appear fruitful.

Parflow runs fine with mpich1 and Linux, and I have the non-mpi version 
working fine on OS X.  I can't get mpich1 to properly configure fortran 
support with gfortran, which the model needs.

Thanks,
Richard


Thanks,
Richard





MPICH as far as I know is working properly, cpilog/etc run fine.


-- 
Richard Jacobsen
Land, Air, and Water Resources, UC Davis
148 Hoagland Hall
530.752.8297