[mpich-discuss] Fwd: intrepid error
Pavan Balaji
balaji at mcs.anl.gov
Sat Mar 19 22:44:00 CDT 2011
Sorry, this is too little information for me to understand what's going
on. Can you stop by my office on Monday and show it?
-- Pavan
On 03/17/2011 09:04 AM, Jayesh Krishna wrote:
> AFAIK, He used Totalview and selected MPICH2 + 1 proc (So my guess is mpiexec -n 1 ./TEST). The test itself does not contain any MPI.
>
> -Jayesh
>
> ----- Original Message -----
> From: "Pavan Balaji"<balaji at mcs.anl.gov>
> To: mpich-discuss at mcs.anl.gov
> Cc: "Jayesh Krishna"<jayesh at mcs.anl.gov>, "Xiabing Xu"<xbxu at mcs.anl.gov>
> Sent: Thursday, March 17, 2011 7:45:53 AM
> Subject: Re: [mpich-discuss] Fwd: intrepid error
>
>
> What's the command-line being used? The error message seems to be coming
> because the UI expects to be able to send some data to STDOUT or STDERR,
> but that socket is closed.
>
> -- Pavan
>
> On 03/16/2011 02:45 PM, Jayesh Krishna wrote:
>> Pavan,
>> One of the developers working in the climate group gets the following error message with mpich2-1.3.2p1 (only "--prefix " used to configure mpich2) on fusion (gcc 4.1.2).
>>
>> ============ error message ======================
>> [mpiexec at flogin1] stdoe_cb (./ui/utils/uiu.c:315): assert (!closed)
>> failed
>> [mpiexec at flogin1] control_cb (./pm/pmiserv/pmiserv_cb.c:229): error
>> in the UI defined callback
>> [mpiexec at flogin1] HYDT_dmxu_poll_wait_for_event (./tools/demux/
>> demux_poll.c:77): callback returned error status
>> [mpiexec at flogin1] HYD_pmci_wait_for_completion (./pm/pmiserv/
>> pmiserv_pmci.c:206): error waiting for event
>> [mpiexec at flogin1] main (./ui/mpich/mpiexec.c:404): process manager
>> error waiting for completion
>>
>> =================================================
>>
>> The test case (part of trilinos - intrepid package) that he is running does not use any MPI but is run using mpiexec (hydra; 1 proc). The test sometimes runs successfully and sometimes fails with the following error message (Obtained when he ran the test multiple times using TotalView).
>> Is this is a known issue ?
>>
>> Regards,
>> Jayesh
>>
>> ----- Forwarded Message -----
>> From: "Xiabing Xu"<xbxu at mcs.anl.gov>
>> To: "Jayesh Krishna"<jayesh at mcs.anl.gov>
>> Sent: Wednesday, March 16, 2011 2:38:42 PM
>> Subject: intrepid error
>>
>> I am using mpich2-1.3.2p1
>>
>>
>> [mpiexec at flogin1] stdoe_cb (./ui/utils/uiu.c:315): assert (!closed)
>> failed
>> [mpiexec at flogin1] control_cb (./pm/pmiserv/pmiserv_cb.c:229): error
>> in the UI defined callback
>> [mpiexec at flogin1] HYDT_dmxu_poll_wait_for_event (./tools/demux/
>> demux_poll.c:77): callback returned error status
>> [mpiexec at flogin1] HYD_pmci_wait_for_completion (./pm/pmiserv/
>> pmiserv_pmci.c:206): error waiting for event
>> [mpiexec at flogin1] main (./ui/mpich/mpiexec.c:404): process manager
>> error waiting for completion
>>
>> gcc 4.1.2
>> _______________________________________________
>> mpich-discuss mailing list
>> mpich-discuss at mcs.anl.gov
>> https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss
>
--
Pavan Balaji
http://www.mcs.anl.gov/~balaji
More information about the mpich-discuss
mailing list