On Tue, Aug 4, 2009 at 4:03 PM, Yixun Liu <span dir="ltr">&lt;<a href="mailto:enjoywm@cs.wm.edu">enjoywm@cs.wm.edu</a>&gt;</span> wrote:<br><div class="gmail_quote"><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
I&#39;m not familiar with MPI. Are there some easy way to test it?</blockquote><div><br>1) You can tryu using MPICH instead, with --download-mpich<br><br>2) You can try compiling a simple program on your own and running it<br>
<br>3) I suspect that there may be a problem running on this machine. It might<br>    be that the configuration is different than the &quot;compute nodes&quot; since it<br>    appears that you can&#39;t see a library you need.<br>
<br>  Matt<br> </div><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><br>
<div class="im">Matthew Knepley wrote:<br>
&gt; On Tue, Aug 4, 2009 at 3:51 PM, Yixun Liu &lt;<a href="mailto:enjoywm@cs.wm.edu">enjoywm@cs.wm.edu</a><br>
</div><div><div></div><div class="h5">&gt; &lt;mailto:<a href="mailto:enjoywm@cs.wm.edu">enjoywm@cs.wm.edu</a>&gt;&gt; wrote:<br>
&gt;<br>
&gt;     Hi,<br>
&gt;     I encounter the error as I run make test,<br>
&gt;     &gt;make test<br>
&gt;     Running test examples to verify correct installation<br>
&gt;     Possible error running C/C++ src/snes/examples/tutorials/ex19 with<br>
&gt;     1 MPI<br>
&gt;     process<br>
&gt;     See<br>
&gt;     <a href="http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html" target="_blank">http://www.mcs.anl.gov/petsc/petsc-as/documentation/troubleshooting.html</a><br>
&gt;     DAT: library load failure: /usr/lib64/libdaplcma.so.1: undefined<br>
&gt;     symbol:<br>
&gt;     dat_registry_add_provider<br>
&gt;<br>
&gt;<br>
&gt; This looks like a bizarre failure with your OpenMPI compiler wrappers.<br>
&gt; Do they work<br>
&gt; at all?<br>
&gt;<br>
&gt;   Matt<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt;     I list the output of configure. Hope it helpful for diagnosis.<br>
&gt;<br>
&gt;     &gt; ./config/configure.py --download-parmetis<br>
&gt;     ***************************************************************************************************************************************************************<br>
&gt;     TESTING: alternateConfigureLibrary from<br>
&gt;     PETSc.packages.petsc4py(config/PETSc/packages/petsc4py.py:69)<br>
&gt;     Compilers:<br>
&gt;      C Compiler:         mpicc  -Wall -Wwrite-strings<br>
&gt;     -Wno-strict-aliasing -g3<br>
&gt;      Fortran Compiler:   mpif90  -Wall -Wno-unused-variable -g<br>
&gt;     Linkers:<br>
&gt;      Static linker:   /usr/bin/ar cr<br>
&gt;     PETSc:<br>
&gt;      **<br>
&gt;      ** Before running &quot;make&quot; your PETSC_ARCH must be specified with:<br>
&gt;      **  ** setenv PETSC_ARCH linux-gnu-c-debug (csh/tcsh)<br>
&gt;      **  ** PETSC_ARCH=linux-gnu-c-debug; export PETSC_ARCH (sh/bash)<br>
&gt;      **<br>
&gt;      PETSC_DIR: /home/scratch/yixun/petsc-3.0.0-p3<br>
&gt;      **<br>
&gt;      ** Now build the libraries with &quot;make all&quot;<br>
&gt;      **<br>
&gt;      Clanguage: C<br>
&gt;      PETSc shared libraries: disabled<br>
&gt;      PETSc dynamic libraries: disabled<br>
&gt;      Scalar type:real<br>
&gt;     MPI:<br>
&gt;      Includes: -I/usr/lib64/mpi/gcc/openmpi/include<br>
&gt;     -I/usr/lib64/mpi/gcc/openmpi/lib64<br>
&gt;     X11:<br>
&gt;      Includes: [&#39;&#39;]<br>
&gt;      Library: [&#39;-lX11&#39;]<br>
&gt;     BLAS/LAPACK: -llapack -lblas<br>
&gt;     ParMetis:<br>
&gt;      Includes: -I/usr/lib64/mpi/gcc/openmpi/include<br>
&gt;     -I/usr/lib64/mpi/gcc/openmpi/lib64<br>
&gt;     -I/home/scratch/yixun/petsc-3.0.0-p3/linux-gnu-c-debug/include<br>
&gt;      Library:<br>
&gt;     -Wl,-rpath,/home/scratch/yixun/petsc-3.0.0-p3/linux-gnu-c-debug/lib<br>
&gt;     -L/home/scratch/yixun/petsc-3.0.0-p3/linux-gnu-c-debug/lib -lparmetis<br>
&gt;     -lmetis<br>
&gt;     *******************************************************************************************************************************************************************<br>
&gt;<br>
&gt;     Yixun<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt; --<br>
&gt; What most experimenters take for granted before they begin their<br>
&gt; experiments is infinitely more interesting than any results to which<br>
&gt; their experiments lead.<br>
&gt; -- Norbert Wiener<br>
<br>
</div></div></blockquote></div><br><br clear="all"><br>-- <br>What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.<br>
-- Norbert Wiener<br>