Which Process Manager are you using ? If you are using mpd, make sure mpd is running and all the nodes are in the ring. Use mpdtrace -l to check.<br><br>
<div class="gmail_quote">On Tue, May 1, 2012 at 5:27 AM, Albert Spade <span dir="ltr"><<a href="mailto:albert.spade@gmail.com" target="_blank">albert.spade@gmail.com</a>></span> wrote:<br>
<blockquote style="BORDER-LEFT:#ccc 1px solid;MARGIN:0px 0px 0px 0.8ex;PADDING-LEFT:1ex" class="gmail_quote">Hi I want to run my program parallely on the my small cluster. It has 5 nodes one master and 4 compute nodes.<br>
When I run the below program on invidual machine it works fine and give proper output. But if I run it on cluster it gives below error.<br>I disabled firewall.<br><br>OUTPUT....<br>-----------------<br>[root@beowulf ~]# mpiexec -n 4 ./cpi<br>
Process 2 of 4 is on beowulf.master<br>Process 3 of 4 is on beowulf.master<br>Process 1 of 4 is on beowulf.master<br>Process 0 of 4 is on beowulf.master<br>Fatal error in PMPI_Reduce: Other MPI error, error stack:<br>PMPI_Reduce(1270)...............: MPI_Reduce(sbuf=0xbfa66ba8, rbuf=0xbfa66ba0, count=1, MPI_DOUBLE, MPI_SUM, root=0, MPI_COMM_WORLD) failed<br>
MPIR_Reduce_impl(1087)..........:<br>MPIR_Reduce_intra(895)..........:<br>MPIR_Reduce_binomial(144).......:<br>MPIDI_CH3U_Recvq_FDU_or_AEP(380): Communication error with rank 2<br>MPIR_Reduce_binomial(144).......:<br>MPIDI_CH3U_Recvq_FDU_or_AEP(380): Communication error with rank 1<br>
<br>_______________________________________________<br>mpich-discuss mailing list <a href="mailto:mpich-discuss@mcs.anl.gov">mpich-discuss@mcs.anl.gov</a><br>To manage subscription options or unsubscribe:<br><a href="https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss" target="_blank">https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss</a><br>
<br></blockquote></div><br>