[mpich-discuss] MPICH2 (or MPI_Init) limitation | scalability

Darius Buntinas buntinas at mcs.anl.gov
Thu Jan 19 11:03:44 CST 2012


Can you also send me the output when using verbose and eth2, and without specifying -iface:

    mpiexec -verbose -iface eth2 -n 64 bin/basic_test
    mpiexec -verbose -n 64 bin/basic_test

Thanks,
-d

On Jan 19, 2012, at 3:16 AM, Bernard Chambon wrote:

> Hi,
> Le 18 janv. 2012 à 19:04, Darius Buntinas a écrit :
> 
>> Hmm.  Is eth2 available on all machines?
>> 
>> -d
> 
> Yes eth2 is available    (if not we would have got a'unable to find interface …'
> 
> In fact the failure appears as soon as I specify -iface (eth0 or eth2) as with 
> 
> >mpiexec -verbose -iface eth0 -n 64 bin/basic_test  (basic_test = MPI_Init + MPI_Finalize
> (without -iface, it's always ok)
> ...
> [mpiexec at ccwpge0061] [pgid: 0] got PMI command: cmd=put kvsname=kvs_5277_0 key=P0-businesscard value=description#????*$port#39147$ifname#134.158.175.61$
> [mpiexec at ccwpge0061] PMI response to fd 6 pid 0: cmd=put_result rc=0 msg=success
> [mpiexec at ccwpge0061] [pgid: 0] got PMI command: cmd=put kvsname=kvs_5277_0 key=P1-businesscard value=description#??.?;+$port#40895$ifname#134.158.175.61$
> [mpiexec at ccwpge0061] PMI response to fd 6 pid 6: cmd=put_result rc=0 msg=success
> [mpiexec at ccwpge0061] [pgid: 0] got PMI command: cmd=put kvsname=kvs_5277_0 key=P10-businesscard value=description#?????+$port#42577$ifname#134.158.175.61$
> [mpiexec at ccwpge0061] PMI response to fd 6 pid 36: cmd=put_result rc=0 msg=success
> [mpiexec at ccwpge0061] [pgid: 0] got PMI command: cmd=put kvsname=kvs_5277_0 key=P20-businesscard value=description#???,v+$port#36031$ifname#134.158.175.61$
> [mpiexec at ccwpge0061] PMI response to fd 6 pid 66: cmd=put_result rc=0 msg=success
> [mpiexec at ccwpge0061] [pgid: 0] got PMI command: cmd=put kvsname=kvs_5277_0 key=P29-businesscard value=description#"?$
> [mpiexec at ccwpge0061] PMI response to fd 6 pid 93: cmd=put_result rc=0 msg=success
> [mpiexec at ccwpge0061] control_cb (./pm/pmiserv/pmiserv_cb.c:215): assert (!closed) failed
> [mpiexec at ccwpge0061] HYDT_dmxu_poll_wait_for_event (./tools/demux/demux_poll.c:77): callback returned error status
> [mpiexec at ccwpge0061] HYD_pmci_wait_for_completion (./pm/pmiserv/pmiserv_pmci.c:181): error waiting for event
> [mpiexec at ccwpge0061] main (./ui/mpich/mpiexec.c:405): process manager error waiting for completion
> 
> 
> Must we pay attention to the special characters like ??.?;+  after  value=description  ?
> 
> 
> Best regards
> 
> 
> (*) minimal code :
>   if (MPI_Init(NULL, NULL) != MPI_SUCCESS ) {
>    printf("Error calling MPI_Init !!, => exiting \n") ; fflush(stdout);
>    return(1);
>   } else {
>    MPI_Finalize();  
>    printf("It's OK \n") ; fflush(stdout);
>    return(0);
>   }
> 
> ---------------
> Bernard CHAMBON
> IN2P3 / CNRS
> 04 72 69 42 18
> 
> _______________________________________________
> mpich-discuss mailing list     mpich-discuss at mcs.anl.gov
> To manage subscription options or unsubscribe:
> https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss



More information about the mpich-discuss mailing list