<meta http-equiv="Content-Type" content="text/html; charset=utf-8"><div dir="ltr"><div class="gmail_default" style="font-family:tahoma,sans-serif;font-size:small">No, not at all. I have removed openmpi on these machines. they only have mpich. But I was wondering what transport layer MPICH chooses by default. These machines have both infiniband and tcp on them. My guess is that it is trying to run on infiniband, and for some reason fails. Do you know how can I force MPICH to only use TCP?</div><div class="gmail_default" style="font-family:tahoma,sans-serif;font-size:small"><br></div><div class="gmail_default" style="font-family:tahoma,sans-serif;font-size:small">Thanks</div></div><div class="gmail_extra"><br clear="all"><div><div class="gmail_signature"><div dir="ltr">Amin Hassani,<br>CIS department at UAB,<br>
Birmingham, AL, USA.</div></div></div>
<br><div class="gmail_quote">On Tue, Nov 25, 2014 at 10:36 PM, Bland, Wesley B. <span dir="ltr"><<a href="mailto:wbland@anl.gov" target="_blank">wbland@anl.gov</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div dir="auto">
<div>Any chance you're using MPICH on one side and Open MPI on the other? You can get some weird situations when mixing the two. <br>
<br>
<br>
</div><div><div class="h5">
<div><br>
On Nov 25, 2014, at 11:33 PM, Amin Hassani <<a href="mailto:ahassani@cis.uab.edu" target="_blank">ahassani@cis.uab.edu</a>> wrote:<br>
<br>
</div>
<blockquote type="cite">
<div>
<div dir="ltr">
<div class="gmail_default">
<div class="gmail_default"><font face="tahoma, sans-serif">Here you go!</font></div>
<div class="gmail_default"><font face="tahoma, sans-serif"><br>
</font></div>
<div class="gmail_default"><font face="tahoma, sans-serif">$ mpirun -hostfile hosts-hydra -np 2 hostname</font></div>
<div class="gmail_default" style="font-family:tahoma,sans-serif;font-size:small">
oakmnt-0-a</div>
<div class="gmail_default" style="font-family:tahoma,sans-serif;font-size:small">
oakmnt-0-b</div>
<div style="font-family:tahoma,sans-serif;font-size:small"><br>
</div>
<div style="font-family:tahoma,sans-serif;font-size:small">Thanks.</div>
</div>
</div>
<div class="gmail_extra"><br clear="all">
<div>
<div>
<div dir="ltr">Amin Hassani,<br>
CIS department at UAB,<br>
Birmingham, AL, USA.</div>
</div>
</div>
<br>
<div class="gmail_quote">On Tue, Nov 25, 2014 at 10:31 PM, Lu, Huiwei <span dir="ltr">
<<a href="mailto:huiweilu@mcs.anl.gov" target="_blank">huiweilu@mcs.anl.gov</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
I can run your simplest code on my machine without a problem. So I guess there is some problem in cluster connection. Could you give me the output of the following?<br>
<br>
$ mpirun -hostfile hosts-hydra -np 2 hostname<br>
<br>
—<br>
<span><font color="#888888">Huiwei<br>
</font></span>
<div>
<div><br>
> On Nov 25, 2014, at 10:24 PM, Amin Hassani <<a href="mailto:ahassani@cis.uab.edu" target="_blank">ahassani@cis.uab.edu</a>> wrote:<br>
><br>
> Hi,<br>
><br>
> the code that I gave you had more stuff in it that I didn't want to distract you. here is the simpler send/recv test that I just ran and it failed.<br>
><br>
> which mpirun: specific directory that I install my MPIs<br>
> /nethome/students/ahassani/usr/mpi/bin/mpirun<br>
><br>
> mpirun with no argument:<br>
> $ mpirun<br>
> [mpiexec@oakmnt-0-a] set_default_values (../../../../src/pm/hydra/ui/mpich/utils.c:1528): no executable provided<br>
> [mpiexec@oakmnt-0-a] HYD_uii_mpx_get_parameters (../../../../src/pm/hydra/ui/mpich/utils.c:1739): setting default values failed<br>
> [mpiexec@oakmnt-0-a] main (../../../../src/pm/hydra/ui/mpich/mpiexec.c:153): error parsing parameters<br>
><br>
><br>
><br>
> #include <mpi.h><br>
> #include <stdio.h><br>
> #include <malloc.h><br>
> #include <unistd.h><br>
> #include <stdlib.h><br>
><br>
> int skip = 10;<br>
> int iter = 30;<br>
><br>
> int main(int argc, char** argv)<br>
> {<br>
> int rank, size;<br>
> int i, j, k;<br>
> double t1, t2;<br>
> int rc;<br>
><br>
> MPI_Init(&argc, &argv);<br>
> MPI_Comm world = MPI_COMM_WORLD, newworld, newworld2;<br>
> MPI_Comm_rank(world, &rank);<br>
> MPI_Comm_size(world, &size);<br>
> int a = 0, b = 1;<br>
> if(rank == 0){<br>
> MPI_Send(&a, 1, MPI_INT, 1, 0, world);<br>
> }else{<br>
> MPI_Recv(&b, 1, MPI_INT, 0, 0, world, MPI_STATUS_IGNORE);<br>
> }<br>
><br>
> printf("b is %d\n", b);<br>
> MPI_Finalize();<br>
><br>
> return 0;<br>
> }<br>
><br>
> Thank you.<br>
><br>
><br>
> Amin Hassani,<br>
> CIS department at UAB,<br>
> Birmingham, AL, USA.<br>
><br>
> On Tue, Nov 25, 2014 at 10:20 PM, Lu, Huiwei <<a href="mailto:huiweilu@mcs.anl.gov" target="_blank">huiweilu@mcs.anl.gov</a>> wrote:<br>
> Hi, Amin,<br>
><br>
> Could you quickly give us the output of the following command: "which mpirun"<br>
><br>
> Also, your simplest code couldn’t compile correctly: "error: ‘t_avg’ undeclared (first use in this function)”. Can you fix it?<br>
><br>
> —<br>
> Huiwei<br>
><br>
> > On Nov 25, 2014, at 2:58 PM, Amin Hassani <<a href="mailto:ahassani@cis.uab.edu" target="_blank">ahassani@cis.uab.edu</a>> wrote:<br>
> ><br>
> > This is the simplest code I have that doesn't run.<br>
> ><br>
> ><br>
> > #include <mpi.h><br>
> > #include <stdio.h><br>
> > #include <malloc.h><br>
> > #include <unistd.h><br>
> > #include <stdlib.h><br>
> ><br>
> > int main(int argc, char** argv)<br>
> > {<br>
> > int rank, size;<br>
> > int i, j, k;<br>
> > double t1, t2;<br>
> > int rc;<br>
> ><br>
> > MPI_Init(&argc, &argv);<br>
> > MPI_Comm world = MPI_COMM_WORLD, newworld, newworld2;<br>
> > MPI_Comm_rank(world, &rank);<br>
> > MPI_Comm_size(world, &size);<br>
> ><br>
> > t2 = 1;<br>
> > MPI_Allreduce(&t2, &t_avg, 1, MPI_DOUBLE, MPI_SUM, world);<br>
> > t_avg = t_avg / size;<br>
> ><br>
> > MPI_Finalize();<br>
> ><br>
> > return 0;<br>
> > }<br>
> ><br>
> > Amin Hassani,<br>
> > CIS department at UAB,<br>
> > Birmingham, AL, USA.<br>
> ><br>
> > On Tue, Nov 25, 2014 at 2:46 PM, "Antonio J. Peña" <<a href="mailto:apenya@mcs.anl.gov" target="_blank">apenya@mcs.anl.gov</a>> wrote:<br>
> ><br>
> > Hi Amin,<br>
> ><br>
> > Can you share with us a minimal piece of code with which you can reproduce this issue?<br>
> ><br>
> > Thanks,<br>
> > Antonio<br>
> ><br>
> ><br>
> ><br>
> > On 11/25/2014 12:52 PM, Amin Hassani wrote:<br>
> >> Hi,<br>
> >><br>
> >> I am having problem running MPICH, on multiple nodes. When I run an multiple MPI processes on one node, it totally works, but when I try to run on multiple nodes, it fails with the error below.<br>
> >> My machines have Debian OS, Both infiniband and TCP interconnects. I'm guessing it has something do to with the TCP network, but I can run openmpi on these machines with no problem. But for some reason I cannot run MPICH on multiple nodes. Please let me
know if more info is needed from my side. I'm guessing there are some configuration that I am missing. I used MPICH 3.1.3 for this test. I googled this problem but couldn't find any solution.<br>
> >><br>
> >> In my MPI program, I am doing a simple allreduce over MPI_COMM_WORLD.<br>
> >><br>
> >> my host file (hosts-hydra) is something like this:<br>
> >> oakmnt-0-a:1<br>
> >> oakmnt-0-b:1 <br>
> >><br>
> >> I get this error:<br>
> >><br>
> >> $ mpirun -hostfile hosts-hydra -np 2 test_dup<br>
> >> Assertion failed in file ../src/mpi/coll/helper_fns.c at line 490: status->MPI_TAG == recvtag<br>
> >> Assertion failed in file ../src/mpi/coll/helper_fns.c at line 490: status->MPI_TAG == recvtag<br>
> >> internal ABORT - process 1<br>
> >> internal ABORT - process 0<br>
> >><br>
> >> ===================================================================================<br>
> >> = BAD TERMINATION OF ONE OF YOUR APPLICATION PROCESSES<br>
> >> = PID 30744 RUNNING AT oakmnt-0-b<br>
> >> = EXIT CODE: 1<br>
> >> = CLEANING UP REMAINING PROCESSES<br>
> >> = YOU CAN IGNORE THE BELOW CLEANUP MESSAGES<br>
> >> ===================================================================================<br>
> >> [mpiexec@vulcan13] HYDU_sock_read (../../../../src/pm/hydra/utils/sock/sock.c:239): read error (Bad file descriptor)<br>
> >> [mpiexec@vulcan13] control_cb (../../../../src/pm/hydra/pm/pmiserv/pmiserv_cb.c:199): unable to read command from proxy<br>
> >> [mpiexec@vulcan13] HYDT_dmxu_poll_wait_for_event (../../../../src/pm/hydra/tools/demux/demux_poll.c:76): callback returned error status<br>
> >> [mpiexec@vulcan13] HYD_pmci_wait_for_completion (../../../../src/pm/hydra/pm/pmiserv/pmiserv_pmci.c:198): error waiting for event<br>
> >> [mpiexec@vulcan13] main (../../../../src/pm/hydra/ui/mpich/mpiexec.c:344): process manager error waiting for completion<br>
> >><br>
> >> Thanks.<br>
> >> Amin Hassani,<br>
> >> CIS department at UAB,<br>
> >> Birmingham, AL, USA.<br>
> >><br>
> >><br>
> >> _______________________________________________<br>
> >> discuss mailing list<br>
> >> <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a><br>
> >><br>
> >> To manage subscription options or unsubscribe:<br>
> >><br>
> >> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
> ><br>
> ><br>
> > --<br>
> > Antonio J. Peña<br>
> > Postdoctoral Appointee<br>
> > Mathematics and Computer Science Division<br>
> > Argonne National Laboratory<br>
> > 9700 South Cass Avenue, Bldg. 240, Of. 3148<br>
> > Argonne, IL 60439-4847<br>
> ><br>
> > <a href="mailto:apenya@mcs.anl.gov" target="_blank">apenya@mcs.anl.gov</a><br>
> > <a href="http://www.mcs.anl.gov/~apenya" target="_blank">www.mcs.anl.gov/~apenya</a><br>
> ><br>
> > _______________________________________________<br>
> > discuss mailing list <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a><br>
> > To manage subscription options or unsubscribe:<br>
> > <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
> ><br>
> > _______________________________________________<br>
> > discuss mailing list <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a><br>
> > To manage subscription options or unsubscribe:<br>
> > <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
><br>
> _______________________________________________<br>
> discuss mailing list <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a><br>
> To manage subscription options or unsubscribe:<br>
> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
><br>
> _______________________________________________<br>
> discuss mailing list <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a><br>
> To manage subscription options or unsubscribe:<br>
> <a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br>
<br>
_______________________________________________<br>
discuss mailing list <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a><br>
To manage subscription options or unsubscribe:<br>
<a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a></div>
</div>
</blockquote>
</div>
<br>
</div>
</div>
</blockquote>
<blockquote type="cite">
<div><span>_______________________________________________</span><br>
<span>discuss mailing list <a href="mailto:discuss@mpich.org" target="_blank">discuss@mpich.org</a></span><br>
<span>To manage subscription options or unsubscribe:</span><br>
<span><a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a></span></div>
</blockquote>
</div></div></div>
<br>_______________________________________________<br>
discuss mailing list <a href="mailto:discuss@mpich.org">discuss@mpich.org</a><br>
To manage subscription options or unsubscribe:<br>
<a href="https://lists.mpich.org/mailman/listinfo/discuss" target="_blank">https://lists.mpich.org/mailman/listinfo/discuss</a><br></blockquote></div><br></div>