[GE users] lam_tight_qrsh-Integration, lamstart Problem

Sebastian ChiefShadow at gmx.de
Wed Nov 16 14:33:57 GMT 2005


Hello,

i have a problem with the integration of lam. I made all like Reuti wrote in
his HowTo and tried to test his settings with this script:

#!/bin/bash
#
#$ -pe lam_tight_qrsh 2
#$ -e /scratch/sge_mpi.err
#$ -o /scratch/sge_mpi.out
#$ -S /bin/bash

export PATH=/opt/lam706/bin:$PATH
echo "Starting Job!"
mpirun C /scratch/mpihello.c
echo "Job finished!"
echo "Sleeping"
sleep 60
echo "Sleep finished"

I have made the new PE lam_tight_qrsh with the correct options and made the
changes in hboot.c and changed the lamd binary.

But when sge will start the LAM-Universe with more than one node, following
error is reported:

ERROR: LAM/MPI unexpectedly received the following on stderr:
*** glibc detected *** free(): invalid pointer: 0x0814f160 ***

The full error-log is attached.
What is wrong, i think the problem is when sge tries to start the
lam-universe.

Thx, Kind Regards, Sebastian 

-- 
Highspeed-Freiheit. Bei GMX superg?nstig, z.B. GMX DSL_Cityflat,
DSL-Flatrate f?r nur 4,99 Euro/Monat*  http://www.gmx.net/de/go/dsl


    [ Part 2: "Attached Text" ]

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe at gridengine.sunsource.net
For additional commands, e-mail: users-help at gridengine.sunsource.net



More information about the gridengine-users mailing list