[mpich-discuss] errors while running the test suite of mpich2

[Rajeev Thakur]

If you configure with --disable-aio, many of the I/O ones will go away. Some
of the others may be timing out because you are writing to NFS, which is
slow.

Rajeev  

> -----Original Message-----
> From: owner-mpich-discuss at mcs.anl.gov 
> [mailto:owner-mpich-discuss at mcs.anl.gov] On Behalf Of 
> Kamaraju Kusumanchi
> Sent: Wednesday, October 22, 2008 11:39 AM
> To: mpich-discuss at mcs.anl.gov
> Subject: [mpich-discuss] errors while running the test suite of mpich2
> 
> The cluster configuration is something like
> 
> 
>                master
>                  |
>           ETHERNET SWITCH
>            /   /   \   \
>         /    /       \    \
>      /     /           \     \
>   /      /               \      \
> node1  node2 .......... node16 node17
>   \     \                /      /
>      \    \            /     /
>         \   \        /    /
>            \  \    /   /
>         INFINIBAND SWITCH
> 
> 
> All the compilations have to be performed on the master node. All the
> executions have to be performed on the slave nodes only.
> 
> I configured mpich2 1.0.7 (on the master node) using gfortran 4.3.2,
> gcc 4.3.2 with the following command
> 
> ~/software/unZipped/mpich2-1.0.7/configure --enable-f77 --enable-f90
> --prefix=/home6/raju/software/compiledLibs/mpich2_1.0.7_gcc_4.
3.2_gfortran_4.3.2
> 2>&1 | tee configure.log
> 
> then I compiled the source code (on the master node) using the command
> 
> nohup make > make.log 2>&1 &
> 
> Then I installed the software using
> 
> nohup make install > install.log 2>&1 &
> 
> Next I logged into the slave node and performed installcheck using
> 
> mpd&
> make installcheck 2>&1 | tee installcheck.log
> 
> After that, I ran the testsuite (on the slave node) using
> 
> make testing 2>&1 | tee make_testing.log
> 
> At this stage, I am getting errors such as
> 
> Looking in ./f77/pt2pt/testlist
> Processing directory info
> Looking in ./f77/info/testlist
> Processing directory spawn
> Looking in ./f77/spawn/testlist
> Processing directory io
> Looking in ./f77/io/testlist
> Unexpected output in iwriteatf: mpiexec_node1.ank.mae.cornell.edu
> (handle_sig_occurred 1122): job ending due to env var
> MPIEXEC_TIMEOUT=180
> Program iwriteatf exited without No Errors
> Unexpected output in iwritef: mpiexec_node1.ank.mae.cornell.edu
> (handle_sig_occurred 1122): job ending due to env var
> MPIEXEC_TIMEOUT=180
> Program iwritef exited without No Errors
> Unexpected output in iwriteshf: mpiexec_node1.ank.mae.cornell.edu
> (handle_sig_occurred 1122): job ending due to env var
> MPIEXEC_TIMEOUT=180
> Program iwriteshf exited without No Errors
> Unexpected output in writef: mpiexec_node1.ank.mae.cornell.edu
> (handle_sig_occurred 1122): job ending due to env var
> MPIEXEC_TIMEOUT=180
> Program writef exited without No Errors
> Unexpected output in writeatf: mpiexec_node1.ank.mae.cornell.edu
> (handle_sig_occurred 1122): job ending due to env var
> MPIEXEC_TIMEOUT=180
> Program writeatf exited without No Errors
> Unexpected output in writeallf: mpiexec_node1.ank.mae.cornell.edu
> (handle_sig_occurred 1122): job ending due to env var
> MPIEXEC_TIMEOUT=180
> Program writeallf exited without No Errors
> Unexpected output in writeshf: mpiexec_node1.ank.mae.cornell.edu
> (handle_sig_occurred 1122): job ending due to env var
> MPIEXEC_TIMEOUT=180
> Program writeshf exited without No Errors
> Unexpected output in writeordf: mpiexec_node1.ank.mae.cornell.edu
> (handle_sig_occurred 1122): job ending due to env var
> MPIEXEC_TIMEOUT=180
> Program writeordf exited without No Errors
> Unexpected output in writeatallf: mpiexec_node1.ank.mae.cornell.edu
> (handle_sig_occurred 1122): job ending due to env var
> MPIEXEC_TIMEOUT=180
> Program writeatallf exited without No Errors
> Unexpected output in writeatallbef: mpiexec_node1.ank.mae.cornell.edu
> (handle_sig_occurred 1122): job ending due to env var
> MPIEXEC_TIMEOUT=180
> Program writeatallbef exited without No Errors
> Unexpected output in writeallbef: mpiexec_node1.ank.mae.cornell.edu
> (handle_sig_occurred 1122): job ending due to env var
> MPIEXEC_TIMEOUT=180
> Program writeallbef exited without No Errors
> Unexpected output in writeordbef: mpiexec_node1.ank.mae.cornell.edu
> (handle_sig_occurred 1122): job ending due to env var
> MPIEXEC_TIMEOUT=180
> Program writeordbef exited without No Errors
> Processing directory rma
> Looking in ./f77/rma/testlist
> Processing directory init
> Looking in ./f77/init/testlist
> 
> 
> .....
> 
> 
> Processing directory attr
> Looking in ./cxx/attr/testlist
> Processing directory pt2pt
> Looking in ./cxx/pt2pt/testlist
> Failed to build bsend1cxx; make[3]: Entering directory
> `/home6/raju/software/compileHere/mpich2-1.0.7/test/mpi/cxx/pt2pt'
> /home6/raju/software/compiledLibs/mpich2_1.0.7_gcc_4.3.2_gfort
ran_4.3.2/bin/mpicxx
> -DHAVE_CONFIG_H -I.
> -I/home6/raju/software/unZipped/mpich2-1.0.7/test/mpi/cxx/pt2pt
> -I../../include
> -I/home6/raju/software/unZipped/mpich2-1.0.7/test/mpi/cxx/pt2p
> t/../../include
>   -c 
> /home6/raju/software/unZipped/mpich2-1.0.7/test/mpi/cxx/pt2pt/
> bsend1cxx.cxx
> /home6/raju/software/unZipped/mpich2-1.0.7/test/mpi/cxx/pt2pt/
> bsend1cxx.cxx:
> In function 'int main(int, char**)':
> /home6/raju/software/unZipped/mpich2-1.0.7/test/mpi/cxx/pt2pt/
> bsend1cxx.cxx:81:
> error: 'strcmp' was not declared in this scope
> /home6/raju/software/unZipped/mpich2-1.0.7/test/mpi/cxx/pt2pt/
> bsend1cxx.cxx:91:
> error: 'strcmp' was not declared in this scope
> make[3]: *** [bsend1cxx.o] Error 1
> make[3]: Leaving directory
> `/home6/raju/software/compileHere/mpich2-1.0.7/test/mpi/cxx/pt2pt'
> 
> 
> 
> 
> I am attaching all the log files for your reference. I have tried
> mpich2 1.0.8rc1. I am getting similar errors when running the test
> suite. Could someone please tell me how to fix these problems?
> 
> thanks
> raju
>