[OMPI users] Open-mpi running issue and bug
Hi, Currently I was trying to run mpirun to improve the performance of my running. However it keeps on showing the following bug: -- It looks like opal_init failed for some reason; your parallel process is likely to abort. There are many reasons that a parallel process can fail during opal_init; some of which are due to configuration or environment problems. This failure appears to be an internal failure; here's some additional information (which may only be relevant to an Open MPI developer): opal_shmem_base_select failed --> Returned value -1 instead of OPAL_SUCCESS -- [home:07059] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file runtime/orte_init.c at line 79 [home:07059] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file orterun.c at line 694 make: *** [sample-1.fa] Error 255 Can I know how to solve this bug? I got try to google around but I still fail to find related answer. Below is the spec of my server: Linux home 2.6.32-71.el6.x86_64 #1 SMP Wed Sep 1 01:33:01 EDT 2010 x86_64 x86_64 x86_64 GNU/Linux Hope it is useful for you in order to solve the bug. Thanks. best regards Edge
Re: [OMPI users] Open-mpi running issue and bug
Hi, Many thanks. I was downloading the latest version of OMPI, version.1.6.3. I just INSTALL following the "INSTALL" guide. I will sent the log file while installing later on. Hope that it will be useful to solve my bug :( Thanks first. best regards Edge From: Ralph Castain To: Edge Edge ; Open MPI Users Sent: Saturday, December 29, 2012 12:42 AM Subject: Re: [OMPI users] Open-mpi running issue and bug What version of OMPI are you using? How was it configured? On Dec 28, 2012, at 2:16 AM, Edge Edge wrote: Hi, > > >Currently I was trying to run mpirun to improve the performance of my running. >However it keeps on showing the following bug: > > >-- >It looks like opal_init failed for some reason; your parallel process is >likely to abort. There are many reasons that a parallel process can >fail during opal_init; some of which are due to configuration or >environment problems. This failure appears to be an internal failure; >here's some additional information (which may only be relevant to an >Open MPI developer): > > > opal_shmem_base_select failed > --> Returned value -1 instead of OPAL_SUCCESS >-- >[home:07059] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file >runtime/orte_init.c at line 79 >[home:07059] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file orterun.c at >line 694 >make: *** [sample-1.fa] Error 255 > > >Can I know how to solve this bug? >I got try to google around but I still fail to find related answer. > > >Below is the spec of my server: >Linux home 2.6.32-71.el6.x86_64 #1 SMP Wed Sep 1 01:33:01 EDT 2010 x86_64 >x86_64 x86_64 GNU/Linux > > > >Hope it is useful for you in order to solve the bug. > > >Thanks. > > > > >best regards >Edge > > > >___ >users mailing list >us...@open-mpi.org >http://www.open-mpi.org/mailman/listinfo.cgi/users
Re: [OMPI users] Open-mpi running issue and bug
Hi, When I issue the following command as shown in the info listed (http://www.open-mpi.org/community/help/): [home@edge openmpi-1.6.3]$mpirun --bynode --hostfile my_hostfile.txt --tag-output ompi_info -v ompi full --parsable [home@edge openmpi-1.6.3]$ -- It looks like opal_init failed for some reason; your parallel process is likely to abort. There are many reasons that a parallel process can fail during opal_init; some of which are due to configuration or environment problems. This failure appears to be an internal failure; here's some additional information (which may only be relevant to an Open MPI developer): opal_shmem_base_select failed --> Returned value -1 instead of OPAL_SUCCESS -- [home@edge:77658] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file runtime/orte_init.c at line 79 [home@edge:77658] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file orterun.c at line 694 I was downloading "openmpi-1.6.3". If I issue the following command in order to install openmpi-1.6.3. It gives me the following log info after I issue, ./autogen.sh log file: *** Running GNU tools [Running] autom4te --language=m4sh ompi_get_version.m4sh -o ompi_get_version.sh [Running] libtoolize --automake --copy --ltdl - It seems that the execution of "libtoolize --automake --copy --ltdl" has failed. See above for the specific error message that caused it to abort. - error log file: configure.in:2125: installing `confdb/config.guess' configure.in:2125: installing `confdb/config.sub' configure.in:210: installing `confdb/install-sh' configure.in:210: installing `confdb/missing' adio/ad_bgl/Makefile.am: installing `confdb/depcomp' configure.in:96: installing `config/compile' configure.in:34: installing `config/config.guess' configure.in:34: installing `config/config.sub' configure.in:8: installing `config/install-sh' configure.in:8: installing `config/missing' rfg/Makefile.am: installing `config/depcomp' configure.in:10: installing `config/config.guess' configure.in:10: installing `config/config.sub' configure.in:11: installing `config/install-sh' configure.in:11: installing `config/missing' otfauxlib/Makefile.am: installing `config/depcomp' libtoolize: `COPYING.LIB' not found in `/usr/share/libtool/libltdl' Thanks and looking forward to hear from you. best regards Edge From: Jeff Squyres (jsquyres) To: Open MPI Users Cc: Edge Edge Sent: Sunday, December 30, 2012 2:07 AM Subject: Re: [OMPI users] Open-mpi running issue and bug Please sent all the info listed here: http://www.open-mpi.org/community/help/ On Dec 28, 2012, at 11:42 AM, Ralph Castain wrote: > What version of OMPI are you using? How was it configured? > > > On Dec 28, 2012, at 2:16 AM, Edge Edge wrote: > >> Hi, >> >> Currently I was trying to run mpirun to improve the performance of my >> running. >> However it keeps on showing the following bug: >> >> -- >> It looks like opal_init failed for some reason; your parallel process is >> likely to abort. There are many reasons that a parallel process can >> fail during opal_init; some of which are due to configuration or >> environment problems. This failure appears to be an internal failure; >> here's some additional information (which may only be relevant to an >> Open MPI developer): >> >> opal_shmem_base_select failed >> --> Returned value -1 instead of OPAL_SUCCESS >> -- >> [home:07059] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file >> runtime/orte_init.c at line 79 >> [home:07059] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file orterun.c at >> line 694 >> make: *** [sample-1.fa] Error 255 >> >> Can I know how to solve this bug? >> I got try to google around but I still fail to find related answer. >> >> Below is the spec of my server: >> Linux home 2.6.32-71.el6.x86_64 #1 SMP Wed Sep 1 01:33:01 EDT 2010 x86_64 >> x86_64 x86_64 GNU/Linux >> >> Hope it is useful for you in order to solve the bug. >> >> Thanks. >> >> >> best regards >> Edge >> >> >> ___ >> 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] Open-mpi running issue and bug
Hi, Thanks for your fast reply. I'm very appreciate it. It shown the following info when I issue: ./configure configure: error: cannot find install-sh, install.sh, or shtool in config "."/config I was download openmpi-1.6.3.tar.gz and unzip it. Thanks and looking forward to hear from you. best regards Edge From: Ralph Castain To: Edge Edge ; Open MPI Users Sent: Friday, January 4, 2013 11:53 AM Subject: Re: [OMPI users] Open-mpi running issue and bug If you are building from a tarball, you don't need nor want to run autogen. All you want to do is run ./configure. If you attempt to run autogen, you'll have to install all the required tools - including an appropriate libtool. Unless you are adding a component to OMPI, you're better off not doing so. On Jan 3, 2013, at 7:19 PM, Edge Edge wrote: Hi, > > >When I issue the following command as shown in the info listed >(http://www.open-mpi.org/community/help/): >[home@edge openmpi-1.6.3]$mpirun --bynode --hostfile my_hostfile.txt >--tag-output ompi_info -v ompi full --parsable >[home@edge openmpi-1.6.3]$ >-- > >It looks like opal_init failed for some reason; your parallel process is >likely to abort. There are many reasons that a parallel process can >fail during opal_init; some of which are due to configuration or >environment problems. This failure appears to be an internal failure; >here's some additional information (which may only be relevant to an >Open MPI developer): > > > opal_shmem_base_select failed > --> Returned value -1 instead of OPAL_SUCCESS >-- >[home@edge:77658] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file >runtime/orte_init.c at line 79 >[home@edge:77658] [[INVALID],INVALID] ORTE_ERROR_LOG: Error in file orterun.c >at line 694 > > >I was downloading "openmpi-1.6.3". >If I issue the following command in order to install openmpi-1.6.3. >It gives me the following log info after I issue, ./autogen.sh >log file: >*** Running GNU tools >[Running] autom4te --language=m4sh ompi_get_version.m4sh -o ompi_get_version.sh >[Running] libtoolize --automake --copy --ltdl > > >- >It seems that the execution of "libtoolize --automake --copy --ltdl" has >failed. See above for >the specific error message that caused it to abort. >- > > >error log file: >configure.in:2125: installing `confdb/config.guess' >configure.in:2125: installing `confdb/config.sub' >configure.in:210: installing `confdb/install-sh' >configure.in:210: installing `confdb/missing' >adio/ad_bgl/Makefile.am: installing `confdb/depcomp' >configure.in:96: installing `config/compile' >configure.in:34: installing `config/config.guess' >configure.in:34: installing `config/config.sub' >configure.in:8: installing `config/install-sh' >configure.in:8: installing `config/missing' >rfg/Makefile.am: installing `config/depcomp' >configure.in:10: installing `config/config.guess' >configure.in:10: installing `config/config.sub' >configure.in:11: installing `config/install-sh' >configure.in:11: installing `config/missing' >otfauxlib/Makefile.am: installing `config/depcomp' >libtoolize: `COPYING.LIB' not found in `/usr/share/libtool/libltdl' > > >Thanks and looking forward to hear from you. > > >best regards >Edge > > > > > From: Jeff Squyres (jsquyres) >To: Open MPI Users >Cc: Edge Edge >Sent: Sunday, December 30, 2012 2:07 AM >Subject: Re: [OMPI users] Open-mpi running issue and bug > >Please sent all the info listed here: > > http://www.open-mpi.org/community/help/ > > >On Dec 28, 2012, at 11:42 AM, Ralph Castain >wrote: > >> What version of OMPI are you using? How was it configured? >> >> >> On Dec 28, 2012, at 2:16 AM, Edge Edge wrote: >> >>> Hi, >>> >>> Currently I was trying to run mpirun to improve the performance of my >>> running. >>> However it keeps on showing the following bug: >>> >>> -- >>> It looks like opal_init failed for some reason; your parallel process is >>> likely to abort. There are many reasons that a parallel process can >>> fail during opal_init; some of which are due to configuration or >