[mpich-discuss] MPI_Probe exception from Windows cluster

Balaji, Pavan balaji at anl.gov
Mon Mar 6 19:27:29 CST 2017


Sorry, we don't support windows.

> On Mar 7, 2017, at 1:34 AM, Feng Yue <Feng.Yue at rms.com> wrote:
> 
> Hi all,
>  
> We are experiencing frequent exceptions from MPI_Probe recently. Does anybody know what cause might be for this exception and what I can do to track down the cause?
>  
> MPI_Probe(113).......................: MPI_Probe(src=MPI_ANY_SOURCE, tag=MPI_ANY_TAG, comm=0x84000002, status=000000033FA217C8) failed
> MPIDI_CH3I_Progress(402).............: 
> MPID_nem_mpich2_blocking_recv(905)...: 
> MPID_nem_newtcp_module_poll(37)......: 
> MPID_nem_newtcp_module_connpoll(2656): 
> gen_read_fail_handler(1145)..........: read from socket failed - The specified network name is no longer available.
>  
> Our cluster has MPICH2 1.4.1p1 and Windows Server 2008 R2 installed on 22 slave nodes and 1 master node.
>  
> Thanks,
> Feng
>  
> _______________________________________________
> discuss mailing list     discuss at mpich.org
> To manage subscription options or unsubscribe:
> https://lists.mpich.org/mailman/listinfo/discuss

_______________________________________________
discuss mailing list     discuss at mpich.org
To manage subscription options or unsubscribe:
https://lists.mpich.org/mailman/listinfo/discuss


More information about the discuss mailing list