On Wed, Aug 5, 2009 at 8:35 AM, Yixun Liu <span dir="ltr"><<a href="mailto:enjoywm@cs.wm.edu">enjoywm@cs.wm.edu</a>></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 did as what you said,<br>
mpiexec -n 2 ./ex2<br>
<div class="im">DAT: library load failure: /usr/lib64/libdaplcma.so.1: undefined symbol:<br>
dat_registry_add_provider<br>
DAT: library load failure: /usr/lib64/libdaplcma.so.1: undefined symbol:<br>
dat_registry_add_provider<br>
<br>
</div>WARNING: Failed to open "OpenIB-cma"<br>
[DAT_PROVIDER_NOT_FOUND:DAT_NAME_NOT_REGISTERED].<br>
This may be a real error or it may be an invalid entry in the uDAPL<br>
Registry which is contained in the dat.conf file. Contact your local<br>
System Administrator to confirm the availability of the interfaces in<br>
the dat.conf file.</blockquote><div><br>As it says, you should probably contact your local admin. It does not seem to<br>have much to do with PETSc.<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>
Satish Balay wrote:<br>
> Try running manually - and see if the example works. These messages<br>
> could be verbose - and misleading.<br>
><br>
> i.e<br>
><br>
> cd src/ksp/ksp/examples/tutorials<br>
> make ex2<br>
> mpiexec -n 2 ./ex2<br>
><br>
> Or you can avoid this by installing mpich - instead of openmpi.<br>
><br>
> Satish<br>
<div class="im">><br>
> On Tue, 4 Aug 2009, Yixun Liu wrote:<br>
><br>
><br>
>> I'm not familiar with MPI. Are there some easy way to test it?<br>
</div><div><div></div><div class="h5">>> Matthew Knepley wrote:<br>
>><br>
>>> On Tue, Aug 4, 2009 at 3:51 PM, Yixun Liu <<a href="mailto:enjoywm@cs.wm.edu">enjoywm@cs.wm.edu</a><br>
>>> <mailto:<a href="mailto:enjoywm@cs.wm.edu">enjoywm@cs.wm.edu</a>>> wrote:<br>
>>><br>
>>> Hi,<br>
>>> I encounter the error as I run make test,<br>
>>> >make test<br>
>>> Running test examples to verify correct installation<br>
>>> Possible error running C/C++ src/snes/examples/tutorials/ex19 with<br>
>>> 1 MPI<br>
>>> process<br>
>>> See<br>
>>> <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>
>>> DAT: library load failure: /usr/lib64/libdaplcma.so.1: undefined<br>
>>> symbol:<br>
>>> dat_registry_add_provider<br>
>>><br>
>>><br>
>>> This looks like a bizarre failure with your OpenMPI compiler wrappers.<br>
>>> Do they work<br>
>>> at all?<br>
>>><br>
>>> Matt<br>
>>><br>
>>><br>
>>><br>
>>> I list the output of configure. Hope it helpful for diagnosis.<br>
>>><br>
>>> > ./config/configure.py --download-parmetis<br>
>>> ***************************************************************************************************************************************************************<br>
>>> TESTING: alternateConfigureLibrary from<br>
>>> PETSc.packages.petsc4py(config/PETSc/packages/petsc4py.py:69)<br>
>>> Compilers:<br>
>>> C Compiler: mpicc -Wall -Wwrite-strings<br>
>>> -Wno-strict-aliasing -g3<br>
>>> Fortran Compiler: mpif90 -Wall -Wno-unused-variable -g<br>
>>> Linkers:<br>
>>> Static linker: /usr/bin/ar cr<br>
>>> PETSc:<br>
>>> **<br>
>>> ** Before running "make" your PETSC_ARCH must be specified with:<br>
>>> ** ** setenv PETSC_ARCH linux-gnu-c-debug (csh/tcsh)<br>
>>> ** ** PETSC_ARCH=linux-gnu-c-debug; export PETSC_ARCH (sh/bash)<br>
>>> **<br>
>>> PETSC_DIR: /home/scratch/yixun/petsc-3.0.0-p3<br>
>>> **<br>
>>> ** Now build the libraries with "make all"<br>
>>> **<br>
>>> Clanguage: C<br>
>>> PETSc shared libraries: disabled<br>
>>> PETSc dynamic libraries: disabled<br>
>>> Scalar type:real<br>
>>> MPI:<br>
>>> Includes: -I/usr/lib64/mpi/gcc/openmpi/include<br>
>>> -I/usr/lib64/mpi/gcc/openmpi/lib64<br>
>>> X11:<br>
>>> Includes: ['']<br>
>>> Library: ['-lX11']<br>
>>> BLAS/LAPACK: -llapack -lblas<br>
>>> ParMetis:<br>
>>> Includes: -I/usr/lib64/mpi/gcc/openmpi/include<br>
>>> -I/usr/lib64/mpi/gcc/openmpi/lib64<br>
>>> -I/home/scratch/yixun/petsc-3.0.0-p3/linux-gnu-c-debug/include<br>
>>> Library:<br>
>>> -Wl,-rpath,/home/scratch/yixun/petsc-3.0.0-p3/linux-gnu-c-debug/lib<br>
>>> -L/home/scratch/yixun/petsc-3.0.0-p3/linux-gnu-c-debug/lib -lparmetis<br>
>>> -lmetis<br>
>>> *******************************************************************************************************************************************************************<br>
>>><br>
>>> Yixun<br>
>>><br>
>>><br>
>>><br>
>>><br>
>>> --<br>
>>> What most experimenters take for granted before they begin their<br>
>>> experiments is infinitely more interesting than any results to which<br>
>>> their experiments lead.<br>
>>> -- Norbert Wiener<br>
>>><br>
>><br>
><br>
><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>