December 2012 - discuss - mailman3.cels.anl.gov

Error while executing mpi code
by Save seecs 25 Dec '12

25 Dec '12

I am getting error while executing my parallel code on 4 number of processors. librdmacm: couldn't read ABI version. librdmacm: assuming: 4 CMA: unable to get RDMA device list -------------------------------------------------------------------------- [[55255,1],0]: A high-performance Open MPI point-to-point messaging module was unable to find any relevant network interfaces: Module: OpenFabrics (openib) Host: barq.seecs.edu.pk Another transport will be used instead, although this may result in lower performance. -------------------------------------------------------------------------- librdmacm: couldn't read ABI version. librdmacm: assuming: 4 CMA: unable to get RDMA device list librdmacm: couldn't read ABI version. librdmacm: assuming: 4 CMA: unable to get RDMA device list librdmacm: couldn't read ABI version. librdmacm: assuming: 4 CMA: unable to get RDMA device list Can any one guide me that how can I fix it ? ~Regards Saqib Khan

1 0

Re: [mpich-discuss] Paramesh with MPICH2 anyone?
by Pavan Balaji 24 Dec '12

24 Dec '12

On 12/24/2012 10:41 AM US Central Time, Angel de Vicente wrote: > Pavan Balaji <balaji(a)mcs.anl.gov> writes: >> There should not be any difference in which MPI is used for any >> high-level library. Did you try running MPICH with simpler programs? >> Does that work correctly? > > yes, MPICH2 for simpler problems is working OK. Actually it is working > OK for my own developed version of our code with MPI. But when using > Paramesh it is totally broken. I haven't looked under the hood to try to > find what could be causing these problems, so I was hoping perhaps there > was someone reading this list who had already tried it and had some idea > of what was wrong. > > I'm going to be using Paramesh extensively during the coming months, so > I might end up looking into the details myself, but any help will be > appreciated. Sorry, we haven't played with Paramesh. Did you try contacting the paramesh folks? -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0

Re: [mpich-discuss] Paramesh with MPICH2 anyone?
by Angel de Vicente 24 Dec '12

24 Dec '12

Hi Pavan, Pavan Balaji <balaji(a)mcs.anl.gov> writes: > There should not be any difference in which MPI is used for any > high-level library. Did you try running MPICH with simpler programs? > Does that work correctly? yes, MPICH2 for simpler problems is working OK. Actually it is working OK for my own developed version of our code with MPI. But when using Paramesh it is totally broken. I haven't looked under the hood to try to find what could be causing these problems, so I was hoping perhaps there was someone reading this list who had already tried it and had some idea of what was wrong. I'm going to be using Paramesh extensively during the coming months, so I might end up looking into the details myself, but any help will be appreciated. Cheers, -- Ángel de Vicente http://www.iac.es/galeria/angelv/ --------------------------------------------------------------------------------------------- ADVERTENCIA: Sobre la privacidad y cumplimiento de la Ley de Protecci�n de Datos, acceda a http://www.iac.es/disclaimer.php WARNING: For more information on privacy and fulfilment of the Law concerning the Protection of Data, consult http://www.iac.es/disclaimer.php?lang=en

1 0

Re: [mpich-discuss] Paramesh with MPICH2 anyone?
by Pavan Balaji 24 Dec '12

24 Dec '12

On 12/24/2012 03:16 AM US Central Time, behzad moradi wrote: > yes,I do running MPICH with simpler program. > no,It doesn't work correctly. Doesn't work how? Can you give some more details? -- Pavan -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0

unsubscribe
by Nachiket Gokhale 24 Dec '12

24 Dec '12

Unsubscribe On Sun, Dec 23, 2012 at 10:11 PM, Pavan Balaji <balaji(a)mcs.anl.gov> wrote: > > There should not be any difference in which MPI is used for any > high-level library. Did you try running MPICH with simpler programs? > Does that work correctly? > > -- Pavan > > On 08/30/2012 02:41 AM US Central Time, Angel de Vicente wrote: >> Hi, >> >> last days I have been trying Paramesh >> (http://www.physics.drexel.edu/~olson/paramesh-doc/Users_manual/amr.html) I'm >> using version 4.1, and I managed to make it work with one of the newest >> Intel compilers and OpenMPI, but with the latest MPICH2 version, the >> code works OK only if I run it in one processor, giving bogus results or >> aborting whenever I try to run it with more processors. >> >> By any chance, anyone here has any experience with Paramesh+MPICH2? >> >> Thanks, >> >> >> >> _______________________________________________ >> mpich-discuss mailing list mpich-discuss(a)mcs.anl.gov >> To manage subscription options or unsubscribe: >> https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss >> > > -- > Pavan Balaji > http://www.mcs.anl.gov/~balaji > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] Paramesh with MPICH2 anyone?
by behzad moradi 24 Dec '12

24 Dec '12

yes,I do running MPICH with simpler program. no,It doesn't work correctly. On Dec 24, 2012, at 5:11 AM, Pavan Balaji wrote: > > There should not be any difference in which MPI is used for any > high-level library. Did you try running MPICH with simpler programs? > Does that work correctly? > > -- Pavan > > On 08/30/2012 02:41 AM US Central Time, Angel de Vicente wrote: >> Hi, >> >> last days I have been trying Paramesh >> (http://www.physics.drexel.edu/~olson/paramesh-doc/Users_manual/amr.html) I'm >> using version 4.1, and I managed to make it work with one of the newest >> Intel compilers and OpenMPI, but with the latest MPICH2 version, the >> code works OK only if I run it in one processor, giving bogus results or >> aborting whenever I try to run it with more processors. >> >> By any chance, anyone here has any experience with Paramesh+MPICH2? >> >> Thanks, >> >> >> >> _______________________________________________ >> mpich-discuss mailing list mpich-discuss(a)mcs.anl.gov >> To manage subscription options or unsubscribe: >> https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss >> > > -- > Pavan Balaji > http://www.mcs.anl.gov/~balaji > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss

1 0

Re: [mpich-discuss] Error in c.txt nemesis does not exist
by Pavan Balaji 23 Dec '12

23 Dec '12

I'm cc'ing our Windows expert, Jayesh. He might know what's going on. -- Pavan On 12/05/2012 03:27 PM US Central Time, Ayesha Ghani wrote: > Hi Dave > > > Thanks for your help! I have tried to install a few versions of MPICH2 > on cygwin now. > > For version 2-1.4, the c.txt file seems fine (no errors) but the m.txt > file goes on for a while 'Entering directory' and 'Leaving directory' > and then at the end outputs an error. > > make[1]: *** No rule to make target > '/home/agha023/libraries/mpich2-1.4/src/mpl/Makefile.am' needed by > '/home/agha023/libraries/mpich2-1.4/src/mpl/Makefile.in'. Stop > make: *** [all-redirect] Error 1 > > > From searching up forums this error occurs because make couldnt find > instructions in the Makefile.in to build the target. However Makefile.am > and Makefile.in both exist in the directory above. > > Again however I can't find a solution to this problem and would greatly > appreciate any help. > > Should I email the m.txt and c.txt files? > > > Kind Regards > Ayesha Ghani > > > > > > > > > On 6 December 2012 08:03, Dave Goodell <goodell(a)mcs.anl.gov > <mailto:[email protected]>> wrote: > > Try using mpich2-1.4 instead. It would not be surprising if there > was a problem with 1.5 on cygwin. > > -Dave > > On Dec 4, 2012, at 8:48 PM PST, Ayesha Ghani wrote: > > > Hello > > > > > > I was working through the installation instructions for mpich2 > very carefully this time and my .c file has an error at the end: > > 'immediate ERROR: nemesis does not exist' > > > > I would appreciate any assistance. > > > > > > Kind Regards > > Ayesha > > <c.txt>_______________________________________________ > > discuss mailing list discuss(a)mpich.org <mailto:[email protected]> > > To manage subscription options or unsubscribe: > > https://lists.mpich.org/mailman/listinfo/discuss > > _______________________________________________ > discuss mailing list discuss(a)mpich.org <mailto:[email protected]> > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > > > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0

Re: [mpich-discuss] Error when running
by Pavan Balaji 23 Dec '12

23 Dec '12

The problem seems to be that the proxy is created, but it is providing an ID which is outside the expected range. The likely problem here is that you are running this on a combination of 32-bit and 64-bit machines, in which case mpiexec would get very confused. Is this the case? -- Pavan On 11/15/2012 08:12 AM US Central Time, Will Dixon wrote: > I am making a cluster for Francis Marion University and can't figure out > how to get rid of this error. I understand what it is stating. A proxy > is created but it can't find it once created. I think it isn't creating > a proxy. The other errors seem to roll off errors from the proxy error. > I tried looking up PMI but could find anything relative. > I get this error every time I try to use mpirun -np 6 ./hello I have 6 > ibook g3's that have powerpc processors and a headnode that is a tower > that runs intel. I installed mpich2 on the headnode and on one of the > slaves. I copied the slave's mpich to the others. I am able to ssh > without any password to the slaves using rsa. my HYDRA_HOST_FILE > environment variable is /mpich2/hosts > > [mpiexec@headnode] HYD_pmcd_pmiserv_proxy_init_cb > (./pm/pmiserv/pmiserv_cb.c:529): cannot find proxy with ID 33554432 > [mpiexec@headnode] HYDT_dmxu_poll_wait_for_event > (./tools/demux/demux_poll.c:77): callback returned error status > [mpiexec@headnode] HYD_pmci_wait_for_completion > (./pm/pmiserv/pmiserv_pmci.c:196): error waiting for event > [mpiexec@headnode] main (./ui/mpich/mpiexec.c:325): process manager > error waiting for completion > > I am able using c3 to run cexecs 'date' and they all return a date no > problem. my /mpich2/hosts looks like: > > node01 > node02 > node03 > node04 > node05 > node07 > > The headnode has two Ethernet ports one going to the internet and the > other to a switch with the slaves connected to them. They are all > running Debian Linux 6.0.6. > Help would be much appreciated. > Sincerely, > Will Dixon > > > > _______________________________________________ > discuss mailing list discuss(a)mpich.org > To manage subscription options or unsubscribe: > https://lists.mpich.org/mailman/listinfo/discuss > -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0

Re: [mpich-discuss] Error with MPMD using MPE
by Pavan Balaji 23 Dec '12

23 Dec '12

MPE is not fully supported at this time. Did you try using a different performance monitoring tool, such as Tau or HPCToolkit? -- Pavan On 11/12/2012 03:37 AM US Central Time, Edward Smith wrote: > > Thank you for the reply. > > The same MPMD program works correctly without MPE. > Each of the programs (a.exe and b.exe) also work indivdually using MPE. > > > _______________________________________________ > mpich-discuss mailing list mpich-discuss(a)mcs.anl.gov > To manage subscription options or unsubscribe: > https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss > -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0

Re: [mpich-discuss] Bug in mpiexec -configfile
by Pavan Balaji 23 Dec '12

23 Dec '12

Thanks John. I can reproduce this error. I've created a ticket for this issue. http://trac.mpich.org/projects/mpich/ticket/1773 -- Pavan On 11/08/2012 02:13 PM US Central Time, John Fettig wrote: > When I run > > mpiexec -configfile myConfigFile > > the input taken from the lines of myConfigFile is truncated at 1024 > characters. If the executable doesn't appear in the first 1024 > characters, mpiexec will complain that you didn't specify an > executable. I see this because I set -env LD_LIBRARY_PATH which in my > case is quite large. > > Here is the output of mpiexec -info > > $ mpiexec -info > HYDRA build details: > Version: 1.5 > Release Date: Mon Oct 8 14:00:48 CDT 2012 > CC: icc -L/usr/lib64 -lscif > CXX: icpc -L/usr/lib64 -lscif > F77: ifort -L/usr/lib64 -lscif > F90: ifort -L/usr/lib64 -lscif > Configure options: '--disable-option-checking' > '--prefix=/home/jfe/local' '--disable-romio' '--disable-mpe' 'CC=icc' > 'CXX=icpc' 'F77=ifort' 'FC=ifort' 'LIBS=-lpthread -lscif ' > 'LDFLAGS=-L/usr/lib64 ' '--with-device=ch3:nemesis:scif,tcp' > '--cache-file=/dev/null' '--srcdir=.' 'CFLAGS= -O2' 'CPPFLAGS= > -I/home/jfe/Downloads/mpich2-1.5/src/mpl/include > -I/home/jfe/Downloads/mpich2-1.5/src/mpl/include > -I/home/jfe/Downloads/mpich2-1.5/src/openpa/src > -I/home/jfe/Downloads/mpich2-1.5/src/openpa/src' > Process Manager: pmi > Launchers available: ssh rsh fork slurm ll lsf > sge manual persist > Topology libraries available: hwloc > Resource management kernels available: user slurm ll lsf sge pbs > Checkpointing libraries available: > Demux engines available: poll select > > > Regards, > John > > > > _______________________________________________ > mpich-discuss mailing list mpich-discuss(a)mcs.anl.gov > To manage subscription options or unsubscribe: > https://lists.mcs.anl.gov/mailman/listinfo/mpich-discuss > -- Pavan Balaji http://www.mcs.anl.gov/~balaji

1 0