DAPL errors on Azure RDMA-enabled SLES cluster -

-

i set 2 azure a8 vms in availability set running sles-hpc 12 (following tutorial here: https://azure.microsoft.com/en-us/documentation/articles/virtual-machines-linux-cluster-rdma/).

when run intel mpi pingpong test, getting dapl errors:

azureuser@sshvm0:~> /opt/intel/impi/5.0.3.048/bin64/mpirun -hosts 10.0.0.4,10.0.0.5 -ppn 1 -n 2 -env i_mpi_fabrics=shm:dapl -env i_mpi_dynamic_connection=0 -env i_mpi_dapl_provider=ofa-v2-ib0 /opt/intel/impi/5.0.3.048/bin64/imb-mpi1 pingpong sshvm1:d28:bef0eb40: 12930 us(12930 us):  dapl_rdma_accept: err -1 input/output error sshvm1:d28:bef0eb40: 12946 us(16 us):  dapl err accept input/output error [1:10.0.0.5][../../src/mpid/ch3/channels/nemesis/netmod/dapl/dapl_conn_rc.c:622] error(0x40000): ofa-v2-ib0: not accept dapl connection request: dat_internal_error() assertion failed in file ../../src/mpid/ch3/channels/nemesis/netmod/dapl/dapl_conn_rc.c @ line 622: 0 internal abort - process 0

similar errors when running 1 of osu mpi microbenchmarks (compiled impi compiler):

azureuser@sshvm0:~> /opt/intel/impi/5.0.3.048/bin64/mpirun -hosts 10.0.0.4,10.0.0.5 -ppn 1 -n 2 -env i_mpi_fabrics=shm:dapl -env i_mpi_dynamic_connection=0 -env i_mpi_dapl_provider=ofa-v2-ib0 /opt/intel/impi/5.0.3.048/bin64/imb-mpi1 pingpong sshvm1:d28:bef0eb40: 12930 us(12930 us):  dapl_rdma_accept: err -1 input/output error sshvm1:d28:bef0eb40: 12946 us(16 us):  dapl err accept input/output error [1:10.0.0.5][../../src/mpid/ch3/channels/nemesis/netmod/dapl/dapl_conn_rc.c:622] error(0x40000): ofa-v2-ib0: not accept dapl connection request: dat_internal_error() assertion failed in file ../../src/mpid/ch3/channels/nemesis/netmod/dapl/dapl_conn_rc.c @ line 622: 0 internal abort - process 0

what cause of these errors? how fix , run these microbenchmarks? help!

i did verify ssh connectivity between 2 nodes running "mpiexec -machinefile machinefile -n 2 hostname"

you need update rdma drivers. have updated documentation follow link below https://azure.microsoft.com/en-us/documentation/articles/virtual-machines-linux-cluster-rdma/

please go section update linux rdma drivers sles 12

please follow instructions , update rdma drivers. please update drivers if have provisioned vm's in 1 of following regions east north central south central north europe


Comments

Post a Comment

Popular posts from this blog

c++ - llvm function pass ReplaceInstWithInst malloc -

-
#include "llvm/pass.h" #include "llvm/ir/module.h" #include "llvm/ir/function.h" #include "llvm/support/raw_ostream.h" #include "llvm/ir/legacypassmanager.h" #include "llvm/ir/instrtypes.h" #include "llvm/transforms/ipo/passmanagerbuilder.h" #include "llvm/ir/irbuilder.h" #include "llvm/transforms/utils/basicblockutils.h" using namespace llvm; namespace { struct replacepass : public functionpass { static char id; replacepass() : functionpass(id) {} virtual bool runonfunction(function &f) { allocainst* insttoreplace = ??? basicblock::iterator ii(insttoreplace); replaceinstwithinst(insttoreplace->getparent()->getinstlist(), ii, new allocainst(type::int32ty, 0, insttoreplace)); return true; } }; } char replacepass::id = 0; static void registerreplacepass(const passmanagerbuilder &, legacy::passmanagerbase &pm) { pm.add(new re...
Read more

Cross-Compiling Linux Kernel for Raspberry Pi - ${CCPREFIX}gcc -v does not work -

-
i'm trying follow this guide . i'm running both ubuntu 12.04.5 lts (gnu/linux 3.13.0-74-generic x86_64) on "real" hardware , 14.04.1 via virtualbox on mac. problem don't past step 1: hoffmann@angl99:~$ export ccprefix=/home/hoffmann/tools/arm-bcm2708/gcc-linaro-arm-linux-gnueabihf-raspbian/bin/arm-linux-gnueabihf- hoffmann@angl99:~$ ${ccprefix}gcc -v i'm getting following error: -bash: /home/hoffmann/tools/arm-bcm2708/gcc-linaro-arm-linux-gnueabihf-raspbian/bin/arm-linux-gnueabihf-gcc: no such file or directory however, file i'm told missing there: hoffmann@angl99:~$ less /home/hoffmann/tools/arm-bcm2708/gcc-linaro-arm-linux-gnueabihf-raspbian/bin/arm-linux-gnueabihf-gcc "/home/hoffmann/tools/arm-bcm2708/gcc-linaro-arm-linux-gnueabihf-raspbian/bin/arm-linux-gnueabihf-gcc" may binary file. see anyway? this result of basic error/misconception. suggest solution? thanks! sebastian ok - i've worked out (with of person po...
Read more

java.lang.NoClassDefFoundError When Creating New Android Project -

-
i've tried installing (and re-installing) android studio several times now, , each time, studio isn't able create new android project. when attempt initial android studio window (ie. basic 'create', 'open', 'import' etc options), i'm asked of questions in wizard, brief progress dialog, followed nothing @ all. if open incomplete project process creates choosing 'open' list, can try choosing create new android project menu. this has same issue, error message in output console time... java.lang.noclassdeffounderror: com/sun/xml/internal/bind/v2/model/annotation/annotationreader (wrong name: com/sun/xml/internal/bind/v0/model/qnnotation/annotationreader) note: when installed android studio, installed incompatible jdk (1.8, not 1.7). have had go default project settings change jdk uses 1.8. does know how can avoid this? there place should downloading jxb classes? missing in java paths? this appears have been caused cor...
Read more