From owner-nwchem-users@emsl.pnl.gov Wed Mar 5 00:49:14 2008 Received: from odyssey.emsl.pnl.gov (localhost [127.0.0.1]) by odyssey.emsl.pnl.gov (8.14.1/8.14.1) with ESMTP id m258nD6x013449 for ; Wed, 5 Mar 2008 00:49:14 -0800 (PST) Received: (from majordom@localhost) by odyssey.emsl.pnl.gov (8.14.1/8.14.1/Submit) id m258nDAM013448 for nwchem-users-outgoing-0915; Wed, 5 Mar 2008 00:49:13 -0800 (PST) X-Authentication-Warning: odyssey.emsl.pnl.gov: majordom set sender to owner-nwchem-users@emsl.pnl.gov using -f X-Ironport-SG: OK_Domains X-Ironport-SBRS: 5.8 X-IronPort-Anti-Spam-Filtered: true X-IronPort-Anti-Spam-Result: AgAAAEvtzUeNFAGSnmdsb2JhbACQcQEBAQEBBgQGBwoYmhI X-IronPort-AV: E=Sophos;i="4.25,449,1199692800"; d="scan'208";a="45181284" X-IronPort-AV: E=Sophos;i="4.25,449,1199660400"; d="scan'208";a="9611882" Message-ID: <47CE5E87.3080806@chemie.hu-berlin.de> Date: Wed, 05 Mar 2008 09:49:11 +0100 From: "Dr. Alexander Hofmann" User-Agent: Thunderbird 2.0.0.12 (Windows/20080213) MIME-Version: 1.0 To: nwchem-users@emsl.pnl.gov Subject: [NWCHEM] V5.1 ga_diag:peigs not interfaced with Infiniband, Open MPI, OpenIB, ofed X-Enigmail-Version: 0.95.6 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-nwchem-users@emsl.pnl.gov Precedence: bulk Dear all, I've encountered the already reported error message 0:0:ga_diag:peigs not interfaced:: 0 1:1:ga_diag:peigs not interfaced:: 0 Last System Error Message from Task 1:: Numerical result out of range [node15:20138] MPI_ABORT invoked on rank 1 in communicator MPI_COMM_WORLD with errorcode 0 0:0:ga_diag:peigs not interfaced:: 0 Last System Error Message from Task 0:: Numerical result out of range [node12:14274] MPI_ABORT invoked on rank 0 in communicator MPI_COMM_WORLD with errorcode 0 for a two cpu test job across different nodes from the nwchem distribution. The call is /opt/openmpi/124_ic91-20061105/bin/mpirun -np 2 -machinefile nodes -mca btl openib $NWCHEM_TOP/bin/LINUX64/nwchem auh2o.nw Staying on one node results the same. I've checked the newslist and found the comments of Dunyou on this subject but it seems I can't follow or it does not work for me ... I'm out of ideas. Any hints are greatly appreciated! Thanks a lot Alexander PS: some details I've compiled NWChem 5.1 with Intel Fortran version ic91_20061105 and Open MPI 1.2.4 (compiled with Intel Fortran 20061105). I also took the new GA 4.1b and 'cp -r'ed it over the tools subdirectory to replace the old GA of the distributed tarball. I also tried the complete replacement of tools with pure GA 4.1b. All the same. My network is OpenIB (Mellanox hardware). Ofed is version 1.2.5.3. TARGET=LINUX64 MPI_INCLUDE=/opt/openmpi/124_ic91-20061105/include MPI_LIB=/opt/openmpi/124_ic91-20061105/lib USE_MPI=yes ARMCI_NETWORK=OPENIB IB_LIB=/opt/ofed/1.2.5.3/lib64 IB_INCLUDE=/opt/ofed/1.2.5.3/include BLAS_I8=yes LARGE_FILES=TRUE NWCHEM=/huge/ah/nwchem-5.1/nwchem-5.1 NWCHEM_BASIS_LIBRARY=/huge/ah/nwchem-5.1/nwchem-5.1/data/libraries/ NWCHEM_TOP=/huge/ah/nwchem-5.1/nwchem-5.1 NWCHEM_NWPW_LIBRARY=/huge/ah/nwchem-5.1/nwchem-5.1/data/ NWCHEM_TARGET=LINUX64 MPIRUN_PATH=/opt/openmpi/124_ic91-20061105/bin I also tried with BLASOPT=-L/opt/goto/114_ic91-20061105 -lgoto_opteron_64 or/and DIAG=PAR (inspired after Dunyou's comments) same result. -- Dr. Alexander Hofmann Humboldt-Universitaet zu Berlin Institut fuer Chemie Arbeitsgruppe Quantenchemie Post: Unter den Linden 6 10099 Berlin Visitors: Brook-Taylor-Strasse 2 12489 Berlin ah chemie.hu-berlin.de @ Tel.: +49-30-2093-7138 Fax.: +49-30-2093-7136 http://www.chemie.hu-berlin.de/ag_sauer/index.html PGP-Key: wwwkeys.de.pgp.net ID: D9D62D35