[OMPI users] OpenMPI job launch failures
On our cluster we are noticing intermediate job launch failure when using OpenMPI. We are currently using OpenMPI-1.6.1 on our cluster and it is integrated with Torque-4.1.3. It failes even for a simple MPI hello world applications. The issue is that orted gets launched on all the nodes but there are a bunch of nodes that dont launch the actual MPI application. There are no errors reported when the job gets killed because the walltime expires. Enabling --debug-daemons doesnt show any errors either. The only difference being that successful runs have MPI_proctable listed and for failures this is absent. Any help in debugging this issue is greatly appreciated. -- Bharath smime.p7s Description: S/MIME Cryptographic Signature
Re: [OMPI users] OpenMPI job launch failures
Sounds like the orteds aren't reporting back to mpirun after launch. The MPI_proctable observation just means that the procs didn't launch in those cases where it is absent, which is something you already observed. Set "-mca plm_base_verbose 5" on your cmd line. You should see each orted report back to mpirun after it launches. If not, then it is likely that something is blocking it. You could also try updating to 1.6.3/4 in case there is some race condition in 1.6.1, though we haven't heard of it to-date. On Feb 14, 2013, at 7:21 AM, Bharath Ramesh wrote: > On our cluster we are noticing intermediate job launch failure when using > OpenMPI. We are currently using OpenMPI-1.6.1 on our cluster and it is > integrated with Torque-4.1.3. It failes even for a simple MPI hello world > applications. The issue is that orted gets launched on all the nodes but > there are a bunch of nodes that dont launch the actual MPI application. There > are no errors reported when the job gets killed because the walltime expires. > Enabling --debug-daemons doesnt show any errors either. The only difference > being that successful runs have MPI_proctable listed and for failures this is > absent. Any help in debugging this issue is greatly appreciated. > > -- > Bharath > > > ___ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users
Re: [OMPI users] OpenMPI job launch failures
Is there any way to prevent the output of more than one node written to the same line. I tried setting --output-filename, which didnt help. For some reason only stdout was written to the files. Making it little bit hard to read close to a 6M output file. -- Bharath On Thu, Feb 14, 2013 at 07:35:02AM -0800, Ralph Castain wrote: > Sounds like the orteds aren't reporting back to mpirun after launch. The > MPI_proctable observation just means that the procs didn't launch in those > cases where it is absent, which is something you already observed. > > Set "-mca plm_base_verbose 5" on your cmd line. You should see each orted > report back to mpirun after it launches. If not, then it is likely that > something is blocking it. > > You could also try updating to 1.6.3/4 in case there is some race condition > in 1.6.1, though we haven't heard of it to-date. > > > On Feb 14, 2013, at 7:21 AM, Bharath Ramesh wrote: > > > On our cluster we are noticing intermediate job launch failure when using > > OpenMPI. We are currently using OpenMPI-1.6.1 on our cluster and it is > > integrated with Torque-4.1.3. It failes even for a simple MPI hello world > > applications. The issue is that orted gets launched on all the nodes but > > there are a bunch of nodes that dont launch the actual MPI application. > > There are no errors reported when the job gets killed because the walltime > > expires. Enabling --debug-daemons doesnt show any errors either. The only > > difference being that successful runs have MPI_proctable listed and for > > failures this is absent. Any help in debugging this issue is greatly > > appreciated. > > > > -- > > Bharath > > > > > > ___ > > users mailing list > > us...@open-mpi.org > > http://www.open-mpi.org/mailman/listinfo.cgi/users > > > ___ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users smime.p7s Description: S/MIME cryptographic signature
[OMPI users] Very high latency with openib btl
Hi, I have a strange case here. The application is "plink" (http://pngu.mgh.harvard.edu/~purcell/plink/download.shtml). The computation/communication pattern of the application is the following : 1- MPI_Init 2- Some single rank computation 3- MPI_Bcast 4- Some single rank computation 5- MPI_Barrier 6- rank 0 sends data to each other rank with MPI_Ssend, one rank at a time. 6- other ranks use MPI_Recv 7- Some single rank computation 8- other ranks send result to rank 0 with MPI_Ssend 8- rank 0 receives data with MPI_Recv 9- rank 0 analyses result 10- MPI_Finalize The amount of data being sent is of the order of the kilobytes, and we have IB. The problem we observe is in step 6. I've output dates before and after each MPI operation. With the openib btl, the behavior I observe is that : - rank 0 starts sending - rank n receives almost instantly, and MPI_Recv returns. - rank 0's MPI_Ssend often returns _minutes_. It looks like the acknowledgement from rank n takes minutes to reach rank 0. Now, the tricky part is that if I disable the openib btl to use instead tcp over IB, there is no such latency and the acknowledgement comes back within a fraction of a second. Also, if rank 0 and rank n are on the same node, the acknowledgement is also quasi-instantaneous (I guess it goes through the SM btl instead of openib). I tried to reproduce this in a simple case, but I observed no such latency. The duration that I got for the whole communication is of the order of milliseconds. Does anyone have an idea of what could cause such very high latencies when using the OpenIB BTL ? Also, I tried replacing step 6 by explicitly sending a confirmation : - rank 0 does MPI_Isend to rank n followed by MPI_Recv from rank n - rank n does MPI_Recv from rank 0 followed by MPI_Isend to rank 0 In this case also, rank n's MPI_Isend executes quasi-instantaneously, and rank 0's MPI_Recv only returns a few minutes later. Thanks, Maxime Boissonneault
[OMPI users] process binding to NUMA node on Opteron 6xxx series CPUs?
Hi, is it possible to bind MPI processes to a NUMA node somehow on Opteron 6xxx series CPUs (e.g. --bind-to-NUMAnode) *without* the usage of a rankfile? Opteron 6xxx have two NUMA nodes per CPU(-socket) so --bind-to-socket doesn't work as I want. This is a 4 socket Opteron 6344 (12 CPUs per CPU(-socket)): root@node01:~> numactl --hardware | grep cpus node 0 cpus: 0 1 2 3 4 5 node 1 cpus: 6 7 8 9 10 11 node 2 cpus: 12 13 14 15 16 17 node 3 cpus: 18 19 20 21 22 23 node 4 cpus: 24 25 26 27 28 29 node 5 cpus: 30 31 32 33 34 35 node 6 cpus: 36 37 38 39 40 41 node 7 cpus: 42 43 44 45 46 47 root@node01:~> /opt/openmpi/1.6.3/gcc/bin/mpirun --report-bindings -np 8 --bind-to-socket --bysocket sleep 1s [node01.cluster:21446] MCW rank 1 bound to socket 1[core 0-11]: [. . . . . . . . . . . .][B B B B B B B B B B B B][. . . . . . . . . . . .][. . . . . . . . . . . .] [node01.cluster:21446] MCW rank 2 bound to socket 2[core 0-11]: [. . . . . . . . . . . .][. . . . . . . . . . . .][B B B B B B B B B B B B][. . . . . . . . . . . .] [node01.cluster:21446] MCW rank 3 bound to socket 3[core 0-11]: [. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .][B B B B B B B B B B B B] [node01.cluster:21446] MCW rank 4 bound to socket 0[core 0-11]: [B B B B B B B B B B B B][. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .] [node01.cluster:21446] MCW rank 5 bound to socket 1[core 0-11]: [. . . . . . . . . . . .][B B B B B B B B B B B B][. . . . . . . . . . . .][. . . . . . . . . . . .] [node01.cluster:21446] MCW rank 6 bound to socket 2[core 0-11]: [. . . . . . . . . . . .][. . . . . . . . . . . .][B B B B B B B B B B B B][. . . . . . . . . . . .] [node01.cluster:21446] MCW rank 7 bound to socket 3[core 0-11]: [. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .][B B B B B B B B B B B B] [node01.cluster:21446] MCW rank 0 bound to socket 0[core 0-11]: [B B B B B B B B B B B B][. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .] So each process is bound to *two* NUMA nodes, but I wan't to bind to *one* NUMA node. What I want is more like this: root@node01:~> cat rankfile rank 0=localhost slot=0-5 rank 1=localhost slot=6-11 rank 2=localhost slot=12-17 rank 3=localhost slot=18-23 rank 4=localhost slot=24-29 rank 5=localhost slot=30-35 rank 6=localhost slot=36-41 rank 7=localhost slot=42-47 root@node01:~> /opt/openmpi/1.6.3/gcc/bin/mpirun --report-bindings -np 8 --rankfile rankfile sleep 1s [node01.cluster:21505] MCW rank 1 bound to socket 0[core 6-11]: [. . . . . . B B B B B B][. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .] (slot list 6-11) [node01.cluster:21505] MCW rank 2 bound to socket 1[core 0-5]: [. . . . . . . . . . . .][B B B B B B . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .] (slot list 12-17) [node01.cluster:21505] MCW rank 3 bound to socket 1[core 6-11]: [. . . . . . . . . . . .][. . . . . . B B B B B B][. . . . . . . . . . . .][. . . . . . . . . . . .] (slot list 18-23) [node01.cluster:21505] MCW rank 4 bound to socket 2[core 0-5]: [. . . . . . . . . . . .][. . . . . . . . . . . .][B B B B B B . . . . . .][. . . . . . . . . . . .] (slot list 24-29) [node01.cluster:21505] MCW rank 5 bound to socket 2[core 6-11]: [. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . B B B B B B][. . . . . . . . . . . .] (slot list 30-35) [node01.cluster:21505] MCW rank 6 bound to socket 3[core 0-5]: [. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .][B B B B B B . . . . . .] (slot list 36-41) [node01.cluster:21505] MCW rank 7 bound to socket 3[core 6-11]: [. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . B B B B B B] (slot list 42-47) [node01.cluster:21505] MCW rank 0 bound to socket 0[core 0-5]: [B B B B B B . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .] (slot list 0-5) Actually I'm dreaming of mpirun --bind-to-NUMAnode --bycore ... or mpirun --bind-to-NUMAnode --byNUMAnode ... Is there any workaround execpt rankfiles for this? Regards, Oliver Weihe
Re: [OMPI users] process binding to NUMA node on Opteron 6xxx series CPUs?
Sure - use the 1.7 branch or the developer's trunk. We have the --bind-to numa option there. On Feb 14, 2013, at 8:54 AM, Oliver Weihe wrote: > Hi, > > is it possible to bind MPI processes to a NUMA node somehow on Opteron 6xxx > series CPUs (e.g. --bind-to-NUMAnode) *without* the usage of a rankfile? > Opteron 6xxx have two NUMA nodes per CPU(-socket) so --bind-to-socket doesn't > work as I want. > > This is a 4 socket Opteron 6344 (12 CPUs per CPU(-socket)): > > root@node01:~> numactl --hardware | grep cpus > node 0 cpus: 0 1 2 3 4 5 > node 1 cpus: 6 7 8 9 10 11 > node 2 cpus: 12 13 14 15 16 17 > node 3 cpus: 18 19 20 21 22 23 > node 4 cpus: 24 25 26 27 28 29 > node 5 cpus: 30 31 32 33 34 35 > node 6 cpus: 36 37 38 39 40 41 > node 7 cpus: 42 43 44 45 46 47 > > root@node01:~> /opt/openmpi/1.6.3/gcc/bin/mpirun --report-bindings -np 8 > --bind-to-socket --bysocket sleep 1s > [node01.cluster:21446] MCW rank 1 bound to socket 1[core 0-11]: [. . . . . . > . . . . . .][B B B B B B B B B B B B][. . . . . . . . . . . .][. . . . . . . > . . . . .] > [node01.cluster:21446] MCW rank 2 bound to socket 2[core 0-11]: [. . . . . . > . . . . . .][. . . . . . . . . . . .][B B B B B B B B B B B B][. . . . . . . > . . . . .] > [node01.cluster:21446] MCW rank 3 bound to socket 3[core 0-11]: [. . . . . . > . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .][B B B B B B B > B B B B B] > [node01.cluster:21446] MCW rank 4 bound to socket 0[core 0-11]: [B B B B B B > B B B B B B][. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . . > . . . . .] > [node01.cluster:21446] MCW rank 5 bound to socket 1[core 0-11]: [. . . . . . > . . . . . .][B B B B B B B B B B B B][. . . . . . . . . . . .][. . . . . . . > . . . . .] > [node01.cluster:21446] MCW rank 6 bound to socket 2[core 0-11]: [. . . . . . > . . . . . .][. . . . . . . . . . . .][B B B B B B B B B B B B][. . . . . . . > . . . . .] > [node01.cluster:21446] MCW rank 7 bound to socket 3[core 0-11]: [. . . . . . > . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .][B B B B B B B > B B B B B] > [node01.cluster:21446] MCW rank 0 bound to socket 0[core 0-11]: [B B B B B B > B B B B B B][. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . . > . . . . .] > > So each process is bound to *two* NUMA nodes, but I wan't to bind to *one* > NUMA node. > > What I want is more like this: > root@node01:~> cat rankfile > rank 0=localhost slot=0-5 > rank 1=localhost slot=6-11 > rank 2=localhost slot=12-17 > rank 3=localhost slot=18-23 > rank 4=localhost slot=24-29 > rank 5=localhost slot=30-35 > rank 6=localhost slot=36-41 > rank 7=localhost slot=42-47 > root@node01:~> /opt/openmpi/1.6.3/gcc/bin/mpirun --report-bindings -np 8 > --rankfile rankfile sleep 1s > [node01.cluster:21505] MCW rank 1 bound to socket 0[core 6-11]: [. . . . . . > B B B B B B][. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . . > . . . . .] (slot list 6-11) > [node01.cluster:21505] MCW rank 2 bound to socket 1[core 0-5]: [. . . . . . . > . . . . .][B B B B B B . . . . . .][. . . . . . . . . . . .][. . . . . . . . > . . . .] (slot list 12-17) > [node01.cluster:21505] MCW rank 3 bound to socket 1[core 6-11]: [. . . . . . > . . . . . .][. . . . . . B B B B B B][. . . . . . . . . . . .][. . . . . . . > . . . . .] (slot list 18-23) > [node01.cluster:21505] MCW rank 4 bound to socket 2[core 0-5]: [. . . . . . . > . . . . .][. . . . . . . . . . . .][B B B B B B . . . . . .][. . . . . . . . > . . . .] (slot list 24-29) > [node01.cluster:21505] MCW rank 5 bound to socket 2[core 6-11]: [. . . . . . > . . . . . .][. . . . . . . . . . . .][. . . . . . B B B B B B][. . . . . . . > . . . . .] (slot list 30-35) > [node01.cluster:21505] MCW rank 6 bound to socket 3[core 0-5]: [. . . . . . . > . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .][B B B B B B . . > . . . .] (slot list 36-41) > [node01.cluster:21505] MCW rank 7 bound to socket 3[core 6-11]: [. . . . . . > . . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . B > B B B B B] (slot list 42-47) > [node01.cluster:21505] MCW rank 0 bound to socket 0[core 0-5]: [B B B B B B . > . . . . .][. . . . . . . . . . . .][. . . . . . . . . . . .][. . . . . . . . > . . . .] (slot list 0-5) > > > Actually I'm dreaming of > mpirun --bind-to-NUMAnode --bycore ... > or > mpirun --bind-to-NUMAnode --byNUMAnode ... > > Is there any workaround execpt rankfiles for this? > > Regards, > Oliver Weihe > ___ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users
Re: [OMPI users] OpenMPI job launch failures
I don't think this is documented anywhere, but it is an available trick (not sure if it is in 1.6.1, but might be): if you set OPAL_OUTPUT_STDERR_FD=N in your environment, we will direct all our error outputs to that file descriptor. If it is "0", then it goes to stdout. Might be worth a try? On Feb 14, 2013, at 8:38 AM, Bharath Ramesh wrote: > Is there any way to prevent the output of more than one node > written to the same line. I tried setting --output-filename, > which didnt help. For some reason only stdout was written to the > files. Making it little bit hard to read close to a 6M output > file. > > -- > Bharath > > On Thu, Feb 14, 2013 at 07:35:02AM -0800, Ralph Castain wrote: >> Sounds like the orteds aren't reporting back to mpirun after launch. The >> MPI_proctable observation just means that the procs didn't launch in those >> cases where it is absent, which is something you already observed. >> >> Set "-mca plm_base_verbose 5" on your cmd line. You should see each orted >> report back to mpirun after it launches. If not, then it is likely that >> something is blocking it. >> >> You could also try updating to 1.6.3/4 in case there is some race condition >> in 1.6.1, though we haven't heard of it to-date. >> >> >> On Feb 14, 2013, at 7:21 AM, Bharath Ramesh wrote: >> >>> On our cluster we are noticing intermediate job launch failure when using >>> OpenMPI. We are currently using OpenMPI-1.6.1 on our cluster and it is >>> integrated with Torque-4.1.3. It failes even for a simple MPI hello world >>> applications. The issue is that orted gets launched on all the nodes but >>> there are a bunch of nodes that dont launch the actual MPI application. >>> There are no errors reported when the job gets killed because the walltime >>> expires. Enabling --debug-daemons doesnt show any errors either. The only >>> difference being that successful runs have MPI_proctable listed and for >>> failures this is absent. Any help in debugging this issue is greatly >>> appreciated. >>> >>> -- >>> Bharath >>> >>> >>> ___ >>> users mailing list >>> us...@open-mpi.org >>> http://www.open-mpi.org/mailman/listinfo.cgi/users >> >> >> ___ >> users mailing list >> us...@open-mpi.org >> http://www.open-mpi.org/mailman/listinfo.cgi/users > ___ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users
Re: [OMPI users] OpenMPI job launch failures
After manually fixing some of the issues I see that the failed nodes never receive commands to launch the local processes. I am going to request the admins to look into the logs for any dropped connections. On Thu, Feb 14, 2013 at 07:35:02AM -0800, Ralph Castain wrote: > Sounds like the orteds aren't reporting back to mpirun after launch. The > MPI_proctable observation just means that the procs didn't launch in those > cases where it is absent, which is something you already observed. > > Set "-mca plm_base_verbose 5" on your cmd line. You should see each orted > report back to mpirun after it launches. If not, then it is likely that > something is blocking it. > > You could also try updating to 1.6.3/4 in case there is some race condition > in 1.6.1, though we haven't heard of it to-date. > > > On Feb 14, 2013, at 7:21 AM, Bharath Ramesh wrote: > > > On our cluster we are noticing intermediate job launch failure when using > > OpenMPI. We are currently using OpenMPI-1.6.1 on our cluster and it is > > integrated with Torque-4.1.3. It failes even for a simple MPI hello world > > applications. The issue is that orted gets launched on all the nodes but > > there are a bunch of nodes that dont launch the actual MPI application. > > There are no errors reported when the job gets killed because the walltime > > expires. Enabling --debug-daemons doesnt show any errors either. The only > > difference being that successful runs have MPI_proctable listed and for > > failures this is absent. Any help in debugging this issue is greatly > > appreciated. > > > > -- > > Bharath > > > > > > ___ > > users mailing list > > us...@open-mpi.org > > http://www.open-mpi.org/mailman/listinfo.cgi/users > > > ___ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users -- Bharath smime.p7s Description: S/MIME cryptographic signature
Re: [OMPI users] OpenMPI job launch failures
When I set the OPAL_OUTPUT_STDERR_FD=0 I receive a whole bunch of mca_oob_tcp_message_recv_complete: invalid message type errors and the job just hangs even when all the nodes have fired off the MPI application. -- Bharath On Thu, Feb 14, 2013 at 09:51:50AM -0800, Ralph Castain wrote: > I don't think this is documented anywhere, but it is an available trick (not > sure if it is in 1.6.1, but might be): if you set OPAL_OUTPUT_STDERR_FD=N in > your environment, we will direct all our error outputs to that file > descriptor. If it is "0", then it goes to stdout. > > Might be worth a try? > > > On Feb 14, 2013, at 8:38 AM, Bharath Ramesh wrote: > > > Is there any way to prevent the output of more than one node > > written to the same line. I tried setting --output-filename, > > which didnt help. For some reason only stdout was written to the > > files. Making it little bit hard to read close to a 6M output > > file. > > > > -- > > Bharath > > > > On Thu, Feb 14, 2013 at 07:35:02AM -0800, Ralph Castain wrote: > >> Sounds like the orteds aren't reporting back to mpirun after launch. The > >> MPI_proctable observation just means that the procs didn't launch in those > >> cases where it is absent, which is something you already observed. > >> > >> Set "-mca plm_base_verbose 5" on your cmd line. You should see each orted > >> report back to mpirun after it launches. If not, then it is likely that > >> something is blocking it. > >> > >> You could also try updating to 1.6.3/4 in case there is some race > >> condition in 1.6.1, though we haven't heard of it to-date. > >> > >> > >> On Feb 14, 2013, at 7:21 AM, Bharath Ramesh wrote: > >> > >>> On our cluster we are noticing intermediate job launch failure when using > >>> OpenMPI. We are currently using OpenMPI-1.6.1 on our cluster and it is > >>> integrated with Torque-4.1.3. It failes even for a simple MPI hello world > >>> applications. The issue is that orted gets launched on all the nodes but > >>> there are a bunch of nodes that dont launch the actual MPI application. > >>> There are no errors reported when the job gets killed because the > >>> walltime expires. Enabling --debug-daemons doesnt show any errors either. > >>> The only difference being that successful runs have MPI_proctable listed > >>> and for failures this is absent. Any help in debugging this issue is > >>> greatly appreciated. > >>> > >>> -- > >>> Bharath > >>> > >>> > >>> ___ > >>> users mailing list > >>> us...@open-mpi.org > >>> http://www.open-mpi.org/mailman/listinfo.cgi/users > >> > >> > >> ___ > >> users mailing list > >> us...@open-mpi.org > >> http://www.open-mpi.org/mailman/listinfo.cgi/users > > ___ > > users mailing list > > us...@open-mpi.org > > http://www.open-mpi.org/mailman/listinfo.cgi/users > > > ___ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users smime.p7s Description: S/MIME cryptographic signature
Re: [OMPI users] OpenMPI job launch failures
Rats - sorry. I seem to recall fixing something in 1.6 that might relate to this - a race condition in the startup. You might try updating to the 1.6.4 release candidate. On Feb 14, 2013, at 11:04 AM, Bharath Ramesh wrote: > When I set the OPAL_OUTPUT_STDERR_FD=0 I receive a whole bunch of > mca_oob_tcp_message_recv_complete: invalid message type errors > and the job just hangs even when all the nodes have fired off the > MPI application. > > > -- > Bharath > > On Thu, Feb 14, 2013 at 09:51:50AM -0800, Ralph Castain wrote: >> I don't think this is documented anywhere, but it is an available trick (not >> sure if it is in 1.6.1, but might be): if you set OPAL_OUTPUT_STDERR_FD=N in >> your environment, we will direct all our error outputs to that file >> descriptor. If it is "0", then it goes to stdout. >> >> Might be worth a try? >> >> >> On Feb 14, 2013, at 8:38 AM, Bharath Ramesh wrote: >> >>> Is there any way to prevent the output of more than one node >>> written to the same line. I tried setting --output-filename, >>> which didnt help. For some reason only stdout was written to the >>> files. Making it little bit hard to read close to a 6M output >>> file. >>> >>> -- >>> Bharath >>> >>> On Thu, Feb 14, 2013 at 07:35:02AM -0800, Ralph Castain wrote: Sounds like the orteds aren't reporting back to mpirun after launch. The MPI_proctable observation just means that the procs didn't launch in those cases where it is absent, which is something you already observed. Set "-mca plm_base_verbose 5" on your cmd line. You should see each orted report back to mpirun after it launches. If not, then it is likely that something is blocking it. You could also try updating to 1.6.3/4 in case there is some race condition in 1.6.1, though we haven't heard of it to-date. On Feb 14, 2013, at 7:21 AM, Bharath Ramesh wrote: > On our cluster we are noticing intermediate job launch failure when using > OpenMPI. We are currently using OpenMPI-1.6.1 on our cluster and it is > integrated with Torque-4.1.3. It failes even for a simple MPI hello world > applications. The issue is that orted gets launched on all the nodes but > there are a bunch of nodes that dont launch the actual MPI application. > There are no errors reported when the job gets killed because the > walltime expires. Enabling --debug-daemons doesnt show any errors either. > The only difference being that successful runs have MPI_proctable listed > and for failures this is absent. Any help in debugging this issue is > greatly appreciated. > > -- > Bharath > > > ___ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users ___ users mailing list us...@open-mpi.org http://www.open-mpi.org/mailman/listinfo.cgi/users >>> ___ >>> users mailing list >>> us...@open-mpi.org >>> http://www.open-mpi.org/mailman/listinfo.cgi/users >> >> >> ___ >> users mailing list >> us...@open-mpi.org >> http://www.open-mpi.org/mailman/listinfo.cgi/users > ___ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users
[OMPI users] qsub error
I'm encountering an error using qsub that none of us can figure out. MPI C++ programs seem to run fine when executed from the command line, but for some reason when I submit them through the queue I get a strange error message .. [compute-3-12.local][[58672,1],0][btl_tcp_endpoint.c:638:mca_btl_tcp_endpoint_complete_connect] connect() to 2002:8170:6c2f:b:21d:9ff:fefd:7d94 failed: Permission denied (13) the compute node 3-12 doesn't matter (the error can generate from any of the nodes, and I'm guessing that 3-12 is the parent node here). To check if there was some problem with my own code, I created a simple 'hello world' program (see attached files). Again, the program runs fine from the command line but fails in qsub with the same sort of error message. I have included (i) the code (ii) the job script for qsub, and (iii) the ".o" file from qsub for the "hello world" program. These don't look like MPI errors, but rather some conflict with, maybe, secure communication accross nodes. Is there something simple I can do to fix this? Thanks, Erik Nelson Howard Hughes Medical Institute 6001 Forest Park Blvd., Room ND10.124 Dallas, Texas 75235-9050 p : 214 645 5981 f : 214 645 5948 #include #include "/opt/openmpi/include/mpi.h" #define bufdim128 int main(int argc, char *argv[]) { char buffer[bufdim]; char id_str[32]; // mpi : MPI::Init(argc,argv); MPI::Status status; int size; int rank; int tag; size=MPI::COMM_WORLD.Get_size(); rank=MPI::COMM_WORLD.Get_rank(); tag=0; if (rank==0) { printf("%d: we have %d processors\n",rank,size); int i; i=1; for ( ;i hello.job Description: Binary data hello.job.o5822590 Description: Binary data
Re: [OMPI users] OpenMPI job launch failures
I ran 15 test jobs using 1.6.4rc3 all of them successful. Unlike 1.6.1 where I would have around 40% of my jobs fail. Thanks for the help really appreciate it. -- Bharath On Thu, Feb 14, 2013 at 11:59:06AM -0800, Ralph Castain wrote: > Rats - sorry. > > I seem to recall fixing something in 1.6 that might relate to this - a race > condition in the startup. You might try updating to the 1.6.4 release > candidate. > > > On Feb 14, 2013, at 11:04 AM, Bharath Ramesh wrote: > > > When I set the OPAL_OUTPUT_STDERR_FD=0 I receive a whole bunch of > > mca_oob_tcp_message_recv_complete: invalid message type errors > > and the job just hangs even when all the nodes have fired off the > > MPI application. > > > > > > -- > > Bharath > > > > On Thu, Feb 14, 2013 at 09:51:50AM -0800, Ralph Castain wrote: > >> I don't think this is documented anywhere, but it is an available trick > >> (not sure if it is in 1.6.1, but might be): if you set > >> OPAL_OUTPUT_STDERR_FD=N in your environment, we will direct all our error > >> outputs to that file descriptor. If it is "0", then it goes to stdout. > >> > >> Might be worth a try? > >> > >> > >> On Feb 14, 2013, at 8:38 AM, Bharath Ramesh wrote: > >> > >>> Is there any way to prevent the output of more than one node > >>> written to the same line. I tried setting --output-filename, > >>> which didnt help. For some reason only stdout was written to the > >>> files. Making it little bit hard to read close to a 6M output > >>> file. > >>> > >>> -- > >>> Bharath > >>> > >>> On Thu, Feb 14, 2013 at 07:35:02AM -0800, Ralph Castain wrote: > Sounds like the orteds aren't reporting back to mpirun after launch. The > MPI_proctable observation just means that the procs didn't launch in > those cases where it is absent, which is something you already observed. > > Set "-mca plm_base_verbose 5" on your cmd line. You should see each > orted report back to mpirun after it launches. If not, then it is likely > that something is blocking it. > > You could also try updating to 1.6.3/4 in case there is some race > condition in 1.6.1, though we haven't heard of it to-date. > > > On Feb 14, 2013, at 7:21 AM, Bharath Ramesh wrote: > > > On our cluster we are noticing intermediate job launch failure when > > using OpenMPI. We are currently using OpenMPI-1.6.1 on our cluster and > > it is integrated with Torque-4.1.3. It failes even for a simple MPI > > hello world applications. The issue is that orted gets launched on all > > the nodes but there are a bunch of nodes that dont launch the actual > > MPI application. There are no errors reported when the job gets killed > > because the walltime expires. Enabling --debug-daemons doesnt show any > > errors either. The only difference being that successful runs have > > MPI_proctable listed and for failures this is absent. Any help in > > debugging this issue is greatly appreciated. > > > > -- > > Bharath > > > > > > ___ > > users mailing list > > us...@open-mpi.org > > http://www.open-mpi.org/mailman/listinfo.cgi/users > > > ___ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users > >>> ___ > >>> users mailing list > >>> us...@open-mpi.org > >>> http://www.open-mpi.org/mailman/listinfo.cgi/users > >> > >> > >> ___ > >> users mailing list > >> us...@open-mpi.org > >> http://www.open-mpi.org/mailman/listinfo.cgi/users > > ___ > > users mailing list > > us...@open-mpi.org > > http://www.open-mpi.org/mailman/listinfo.cgi/users > > > ___ > users mailing list > us...@open-mpi.org > http://www.open-mpi.org/mailman/listinfo.cgi/users smime.p7s Description: S/MIME cryptographic signature