[mpich-discuss] response as soon as possible please

[James Dinan]

In this situation, it can be helpful to attach a debugger and determine 
the state of each process.  You should build your program with debugging 
symbols and attach the debugger to the running process to determine 
that's going on.  With gdb, for example, you would do:

$ gdb --pid=[PID here] ./PPM

  ~Jim.

On 6/14/11 12:55 PM, Irfan Gul wrote:
> I have profiled to sorting algorithms including Parallel Odd Even and
> Parallel Quick sort. I want to compare the results with Parallel Merge
> sort which I have profiled in attached file
> it does not give the compilation error. but
> irfan at hpc <mailto:irfan at hpc>:~> mpicc -o PPM PPMS.c -lmpe -llmpe
> irfan at hpc <mailto:irfan at hpc>:~> mpirun -np 4 ./PPM
> 4 processes mandates root height of 2
> Enter the size of array         Data read and Lines are: 1000000
> 1000000
> Process with ID 3 and time  0.201
> Process with ID 2 and time  0.201
> Process with ID 0 and time  0.201
> Process with ID 1 and time  0.202
>  >it does not end execution.when I press Ctrl+c it gives message
> Ctrl-C caught... cleaning up processes
> [press Ctrl-C again to force abort]
> [mpiexec at hpc] HYDT_bscu_wait_for_completion
> (./tools/bootstrap/utils/bscu_wait.c:99): one of the processes
> terminated badly; aborting
> [mpiexec at hpc] HYDT_bsci_wait_for_completion
> (./tools/bootstrap/src/bsci_wait.c:18): bootstrap device returned error
> waiting for completion
> [mpiexec at hpc] HYD_pmci_wait_for_completion
> (./pm/pmiserv/pmiserv_pmci.c:352): bootstrap server returned error
> waiting for completion
> [mpiexec at hpc] main (./ui/mpich/mpiexec.c:302): process manager error
> waiting for completion
>  > thanks
>
>
>
> _______________________________________________
> mpich-discuss mailing list
> mpich-discuss at mcs.anl.gov
> https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss