<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body text="#000000" bgcolor="#FFFFFF">
    It is not the first time I have this problem and my aim was now to
    try to solve it instead of ignoring tests. The environment seams
    coherent (see below).<br>
    I'll try to run in debug mode to investigate where the code hangs.<br>
    <br>
    Patrick<br>
    <br>
    <tt>[begou@kareline tutorials]$ make ex19</tt><tt><br>
    </tt><tt><b>mpicc</b> -o ex19.o -c -Wall -Wwrite-strings
      -Wno-strict-aliasing -Wno-unknown-pragmas -fvisibility=hidden
      -g3  
      -I/kareline/data/begou/YALES2_1.0.0/PREREQUIS/petsc-git/include
      -I/kareline/data/begou/YALES2_1.0.0/PREREQUIS/petsc-git/GCC48/include   
      `pwd`/ex19.c</tt><tt><br>
    </tt><tt>mpicc -Wall -Wwrite-strings -Wno-strict-aliasing
      -Wno-unknown-pragmas -fvisibility=hidden -g3  -o ex19  ex19.o
      -L/kareline/data/begou/YALES2_1.0.0/PREREQUIS/petsc-git/GCC48/lib 
      -lpetsc -llapack -lblas -lX11 -lhwloc -lssl -lcrypto
      -L/opt/openmpi173-GCC48-node/lib
      -L/opt/GCC48c/lib/gcc/x86_64-unknown-linux-gnu/4.8.1
      -L/opt/GCC48c/lib64 -L/opt/GCC48c/lib -lmpi_usempi -lmpi_mpifh
      -lgfortran -lm -lgfortran -lm -lquadmath -lm -lmpi_cxx -lstdc++
      -L/opt/openmpi173-GCC48-node/lib
      -L/opt/GCC48c/lib/gcc/x86_64-unknown-linux-gnu/4.8.1
      -L/opt/GCC48c/lib64 -L/opt/GCC48c/lib -ldl -lmpi -lgcc_s -lpthread
      -ldl </tt><tt><br>
    </tt><tt>/bin/rm -f ex19.o</tt><tt><br>
      <br>
    </tt><tt>[begou@kareline tutorials]$ <b>which mpiexec</b></tt><b><tt><br>
      </tt></b><b><tt>/opt/openmpi173-GCC48-node/bin/mpiexec</tt></b><tt><br>
      <br>
    </tt><tt>[begou@kareline tutorials]$ <b>mpicc --showme</b></tt><b><tt><br>
      </tt></b><b><tt>gcc -I/opt/openmpi173-GCC48-node/include -pthread
        -L/opt/openmpi173-GCC48-node/lib -lmpi</tt></b><tt><br>
      <br>
    </tt><tt>[begou@kareline tutorials]$ <b>ldd ./ex19</b></tt><tt><br>
    </tt><tt>    linux-vdso.so.1 =>  (0x00007ffe771ea000)</tt><tt><br>
    </tt><tt>    liblapack.so.3 => /usr/lib64/atlas/liblapack.so.3
      (0x00007f5ac8596000)</tt><tt><br>
    </tt><tt>    libblas.so.3 => /usr/lib64/libblas.so.3
      (0x00007f5ac833e000)</tt><tt><br>
    </tt><tt>    libX11.so.6 => /usr/lib64/libX11.so.6
      (0x0000003c66600000)</tt><tt><br>
    </tt><tt>    libhwloc.so.5 => /usr/lib64/libhwloc.so.5
      (0x0000003dde600000)</tt><tt><br>
    </tt><tt>    libssl.so.10 => /usr/lib64/libssl.so.10
      (0x0000003c71600000)</tt><tt><br>
    </tt><tt>    libcrypto.so.10 => /usr/lib64/libcrypto.so.10
      (0x0000003c69a00000)</tt><tt><br>
    </tt><tt>    libmpi_usempi.so.1 => <b>/opt/openmpi173-GCC48-node/lib/libmpi_usempi.so.1</b>
      (0x00007f5ac80a4000)</tt><tt><br>
    </tt><tt>    libmpi_mpifh.so.2 => <b>/opt/openmpi173-GCC48-node/lib/libmpi_mpifh.so.2</b>
      (0x00007f5ac7e5a000)</tt><tt><br>
    </tt><tt>    libgfortran.so.3 =>
      /opt/GCC48c/lib64/libgfortran.so.3 (0x00007f5ac7b43000)</tt><tt><br>
    </tt><tt>    libm.so.6 => /lib64/libm.so.6 (0x0000003c63200000)</tt><tt><br>
    </tt><tt>    libquadmath.so.0 =>
      /opt/GCC48c/lib64/libquadmath.so.0 (0x00007f5ac7907000)</tt><tt><br>
    </tt><tt>    libmpi_cxx.so.1 => <b>/opt/openmpi173-GCC48-node/lib/libmpi_cxx.so.1</b>
      (0x00007f5ac76ed000)</tt><tt><br>
    </tt><tt>    libstdc++.so.6 => /opt/GCC48c/lib64/libstdc++.so.6
      (0x00007f5ac73e4000)</tt><tt><br>
    </tt><tt>    libdl.so.2 => /lib64/libdl.so.2 (0x0000003c63a00000)</tt><tt><br>
    </tt><tt>    libmpi.so.1 => <b>/opt/openmpi173-GCC48-node/lib/libmpi.so.1</b>
      (0x00007f5ac7115000)</tt><tt><br>
    </tt><tt>    libgcc_s.so.1 => /opt/GCC48c/lib64/libgcc_s.so.1
      (0x00007f5ac6eff000)</tt><tt><br>
    </tt><tt>    libpthread.so.0 => /lib64/libpthread.so.0
      (0x0000003c63600000)</tt><tt><br>
    </tt><tt>    libc.so.6 => /lib64/libc.so.6 (0x0000003c62e00000)</tt><tt><br>
    </tt><tt>    libf77blas.so.3 => /usr/lib64/atlas/libf77blas.so.3
      (0x00007f5ac6cdf000)</tt><tt><br>
    </tt><tt>    libcblas.so.3 => /usr/lib64/atlas/libcblas.so.3
      (0x00007f5ac6abe000)</tt><tt><br>
    </tt><tt>    libxcb.so.1 => /usr/lib64/libxcb.so.1
      (0x0000003c66200000)</tt><tt><br>
    </tt><tt>    libnuma.so.1 => /usr/lib64/libnuma.so.1
      (0x0000003ddee00000)</tt><tt><br>
    </tt><tt>    libpci.so.3 => /lib64/libpci.so.3
      (0x0000003ddea00000)</tt><tt><br>
    </tt><tt>    libxml2.so.2 => /usr/lib64/libxml2.so.2
      (0x0000003c6e200000)</tt><tt><br>
    </tt><tt>    libgssapi_krb5.so.2 => /lib64/libgssapi_krb5.so.2
      (0x0000003c70a00000)</tt><tt><br>
    </tt><tt>    libkrb5.so.3 => /lib64/libkrb5.so.3
      (0x0000003c70e00000)</tt><tt><br>
    </tt><tt>    libcom_err.so.2 => /lib64/libcom_err.so.2
      (0x0000003c69e00000)</tt><tt><br>
    </tt><tt>    libk5crypto.so.3 => /lib64/libk5crypto.so.3
      (0x0000003c6f200000)</tt><tt><br>
    </tt><tt>    libz.so.1 => /lib64/libz.so.1 (0x0000003c64200000)</tt><tt><br>
    </tt><tt>    libopen-rte.so.6 => <b>/opt/openmpi173-GCC48-node/lib/libopen-rte.so.6</b>
      (0x00007f5ac684a000)</tt><tt><br>
    </tt><tt>    libopen-pal.so.6 => <b>/opt/openmpi173-GCC48-node/lib/libopen-pal.so.6</b>
      (0x00007f5ac6575000)</tt><tt><br>
    </tt><tt>    librt.so.1 => /lib64/librt.so.1 (0x0000003c63e00000)</tt><tt><br>
    </tt><tt>    libnsl.so.1 => /lib64/libnsl.so.1
      (0x0000003c71a00000)</tt><tt><br>
    </tt><tt>    libutil.so.1 => /lib64/libutil.so.1
      (0x0000003c6ba00000)</tt><tt><br>
    </tt><tt>    /lib64/ld-linux-x86-64.so.2 (0x0000003c62a00000)</tt><tt><br>
    </tt><tt>    libatlas.so.3 => /usr/lib64/atlas/libatlas.so.3
      (0x00007f5ac5f18000)</tt><tt><br>
    </tt><tt>    libXau.so.6 => /usr/lib64/libXau.so.6
      (0x0000003c66a00000)</tt><tt><br>
    </tt><tt>    libresolv.so.2 => /lib64/libresolv.so.2
      (0x0000003c64e00000)</tt><tt><br>
    </tt><tt>    libkrb5support.so.0 => /lib64/libkrb5support.so.0
      (0x0000003c6fe00000)</tt><tt><br>
    </tt><tt>    libkeyutils.so.1 => /lib64/libkeyutils.so.1
      (0x0000003c6f600000)</tt><tt><br>
    </tt><tt>    libselinux.so.1 => /lib64/libselinux.so.1
      (0x0000003c64600000)</tt><tt><br>
    </tt><tt><br>
    </tt><br>
    Matthew Knepley a écrit :
    <blockquote type="cite">
      <div dir="ltr">
        <div class="gmail_extra">
          <div class="gmail_quote">On Fri, Jan 6, 2017 at 2:39 AM,
            Patrick Begou <span dir="ltr"><<a
                href="mailto:Patrick.Begou@legi.grenoble-inp.fr"
                target="_blank"><a class="moz-txt-link-abbreviated" href="mailto:Patrick.Begou@legi.grenoble-inp.fr">Patrick.Begou@legi.grenoble-inp.fr</a></a>></span>
            wrote:<br>
            <blockquote class="gmail_quote" style="margin:0 0 0
              .8ex;border-left:1px #ccc solid;padding-left:1ex">
              <div text="#000000" bgcolor="#FFFFFF"> <tt>Hi Matthew,<br>
                  <br>
                  Launching manualy ex19 shows only one process
                  consuming cpu time, after 952mn I've killed the job
                  this morning.<br>
                  <br>
                  [begou@kareline tutorials]$ make ex19<br>
                  mpicc -o ex19.o -c -Wall -Wwrite-strings
                  -Wno-strict-aliasing -Wno-unknown-pragmas
                  -fvisibility=hidden -g3  
                  -I/kareline/data/begou/YALES2_<wbr>1.0.0/PREREQUIS/petsc-git/<wbr>include

                  -I/kareline/data/begou/YALES2_<wbr>1.0.0/PREREQUIS/petsc-git/<wbr>GCC48/include   

                  `pwd`/ex19.c<br>
                  mpicc -Wall -Wwrite-strings -Wno-strict-aliasing
                  -Wno-unknown-pragmas -fvisibility=hidden -g3  -o ex19 
                  ex19.o -L/kareline/data/begou/YALES2_<wbr>1.0.0/PREREQUIS/petsc-git/<wbr>GCC48/lib 

                  -lpetsc -llapack -lblas -lX11 -lhwloc -lssl -lcrypto
                  -L/opt/openmpi173-GCC48-node/<wbr>lib
                  -L/opt/GCC48c/lib/gcc/x86_64-<wbr>unknown-linux-gnu/4.8.1

                  -L/opt/GCC48c/lib64 -L/opt/GCC48c/lib -lmpi_usempi
                  -lmpi_mpifh -lgfortran -lm -lgfortran -lm -lquadmath
                  -lm -lmpi_cxx -lstdc++ -L/opt/openmpi173-GCC48-node/<wbr>lib

                  -L/opt/GCC48c/lib/gcc/x86_64-<wbr>unknown-linux-gnu/4.8.1

                  -L/opt/GCC48c/lib64 -L/opt/GCC48c/lib -ldl -lmpi
                  -lgcc_s -lpthread -ldl <br>
                  /bin/rm -f ex19.o<br>
                  [begou@kareline tutorials]$ mpiexec -n 2 ./ex19
                  -snes_monitor<br>
                  <br>
                  top command shows:<br>
                  <br>
                    PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM   
                  TIME+  COMMAND                <br>
                  32184 begou     20   0  249m 7152 5132 R 99.8  0.0
                  952:15.97 ex19                   <br>
                  32183 begou     20   0 71676 3508 2264 S  0.0  0.0  
                  0:00.04 mpiexec                <br>
                  32185 begou     20   0  185m 7132 5124 S  0.0  0.0  
                  0:00.04 ex19                   <br>
                  <br>
                  looks like the first process waiting for something
                  that never occur in MPI communication....<br>
                </tt></div>
            </blockquote>
            <div><br>
            </div>
            <div>1000s of people run this every day, so I am skeptical
              of that explanation.</div>
            <div><br>
            </div>
            <div>However, this could happen if the 'mpiexec' in your
              path does not match the MPI libraries that PETSc is linked
              to.</div>
            <div><br>
            </div>
            <div>   Matt</div>
            <div> </div>
            <blockquote class="gmail_quote" style="margin:0 0 0
              .8ex;border-left:1px #ccc solid;padding-left:1ex">
              <div text="#000000" bgcolor="#FFFFFF"><tt> Patrick<br>
                </tt><tt><br>
                  Matthew Knepley a écrit :
                  <blockquote type="cite">
                    <div dir="ltr">
                      <div class="gmail_extra">
                        <div class="gmail_quote">On Thu, Jan 5, 2017 at
                          6:31 AM, Patrick Begou <span dir="ltr"><<a
class="m_5136712065869368947moz-txt-link-abbreviated"
                              href="mailto:Patrick.Begou@legi.grenoble-inp.fr"
                              target="_blank"><a class="moz-txt-link-abbreviated" href="mailto:Patrick.Begou@legi.grenoble">Patrick.Begou@legi.grenoble</a>-<wbr>inp.fr</a>></span>
                          wrote:<br>
                          <blockquote class="gmail_quote"
                            style="margin:0 0 0 .8ex;border-left:1px
                            #ccc solid;padding-left:1ex">I am unable to
                            run any test on petsc. It looks like if the
                            ex19 run freeze on the server as it do not
                            use any cpu time and pstree shows<br>
                            <br>
                            sshd---bash-+-gedit<br>
                                        `-make---sh-+-gmake---sh---gma<wbr>ke---sh---mpiexec---ex19<br>
                                                    `-tee<br>
                            I've tested petsc-3.7.5.tar.gz and  the
                            latest sources on the Git repository.<br>
                          </blockquote>
                          <div><br>
                          </div>
                          <div>All make is doing is running ex19, which
                            you can do by hand. What do you get for</div>
                          <div><br>
                          </div>
                          <div>  cd $PETSC_DIR</div>
                          <div>  cd src/snes/examples/tutorials</div>
                          <div>  make ex19</div>
                          <div>  mpiexec -n 2 ./ex19 -snes_monitor</div>
                          <div><br>
                          </div>
                          <div>  Thanks,</div>
                          <div><br>
                          </div>
                          <div>    Matt</div>
                          <div> </div>
                          <blockquote class="gmail_quote"
                            style="margin:0 0 0 .8ex;border-left:1px
                            #ccc solid;padding-left:1ex"> Setup from the
                            Git repo:<br>
                            ./configure --prefix=/kareline/data/begou/<wbr>YALES2_1.0.0/PREREQUIS/petsc-g<wbr>it-binaries


                            \<br>
                            --PETSC_ARCH=GCC48 \<br>
                            --PETSC_DIR=/kareline/data/beg<wbr>ou/YALES2_1.0.0/PREREQUIS/pets<wbr>c-git


                            \<br>
                            --with-shared-libraries=0 \<br>
                            --with-fortran-interfaces=1 \<br>
                            --with-fortran-kernels=1 \<br>
                            --with-cc=mpicc \<br>
                            --with-fc=mpif90 \<br>
                            --with-cxx=mpicxx<br>
                            <br>
                             make PETSC_DIR=/kareline/data/begou<wbr>/YALES2_1.0.0/PREREQUIS/petsc-<wbr>git


                            PETSC_ARCH=GCC48 all<br>
                            <br>
                             make PETSC_DIR=/kareline/data/begou<wbr>/YALES2_1.0.0/PREREQUIS/petsc-<wbr>git


                            PETSC_ARCH=GCC48 install<br>
                            <br>
                             make PETSC_DIR=/kareline/data/begou<wbr>/YALES2_1.0.0/PREREQUIS/petsc-<wbr>git-binaries


                            PETSC_ARCH="" test<br>
                            <br>
                            <br>
                            In the log file I've just:<br>
                            <br>
                            Running test examples to verify correct
                            installation<br>
                            Using PETSC_DIR=/kareline/data/begou<wbr>/YALES2_1.0.0/PREREQUIS/petsc-<wbr>git-binaries


                            and PETSC_ARCH=<br>
                            <br>
                            I'm using:<br>
                            gcc version 4.8.1<br>
                            Open MPI: 1.7.3 (build with gcc 4.8.1)<br>
                            (This environment is in production for a
                            while for many local software and works
                            fine)<br>
                            <br>
                            Any suggestion is welcome<span
                              class="m_5136712065869368947HOEnZb"><font
                                color="#888888"><br>
                                <br>
                                Patrick<br>
                                <br>
                                -- <br>
                                ==============================<wbr>==============================<wbr>=======<br>
                                |  Equipe M.O.S.T.         |           
                                                          |<br>
                                |  Patrick BEGOU           | mailto:<a
                                  href="mailto:Patrick.Begou@grenoble-inp.fr"
                                  target="_blank">Patrick.Begou@grenoble-<wbr>inp.fr</a>
                                |<br>
                                |  LEGI                    |           
                                                          |<br>
                                |  BP 53 X                 | Tel 04 76
                                82 51 35                   |<br>
                                |  38041 GRENOBLE CEDEX    | Fax 04 76
                                82 52 71                   |<br>
                                ==============================<wbr>==============================<wbr>=======<br>
                                <br>
                              </font></span></blockquote>
                        </div>
                        <br>
                        <br clear="all">
                        <span class="HOEnZb"><font color="#888888">
                            <div><br>
                            </div>
                            -- <br>
                            <div
                              class="m_5136712065869368947gmail_signature"
                              data-smartmail="gmail_signature">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</div>
                          </font></span></div>
                      <span class="HOEnZb"><font color="#888888"> </font></span></div>
                    <span class="HOEnZb"><font color="#888888"> </font></span></blockquote>
                  <span class="HOEnZb"><font color="#888888"> <br>
                      <br>
                    </font></span></tt><span class="HOEnZb"><font
                    color="#888888">
                    <pre class="m_5136712065869368947moz-signature" cols="80">-- 
==============================<wbr>==============================<wbr>=======
|  Equipe M.O.S.T.         |                                      |
|  Patrick BEGOU           | <a class="m_5136712065869368947moz-txt-link-freetext" href="mailto:Patrick.Begou@grenoble-inp.fr" target="_blank">mailto:Patrick.Begou@grenoble-<wbr>inp.fr</a> |
|  LEGI                    |                                      |
|  BP 53 X                 | Tel 04 76 82 51 35                   |
|  38041 GRENOBLE CEDEX    | Fax 04 76 82 52 71                   |
==============================<wbr>==============================<wbr>=======
</pre>
  </font></span></div>

</blockquote></div>

<div>
</div>-- 
<div class="gmail_signature" data-smartmail="gmail_signature">What most experimenters take for granted before they begin their experiments is infinitely more interesting than any results to which their experiments lead.
-- Norbert Wiener</div>
</div></div>



</blockquote>

<pre class="moz-signature" cols="80">-- 
===================================================================
|  Equipe M.O.S.T.         |                                      |
|  Patrick BEGOU           | <a class="moz-txt-link-freetext" href="mailto:Patrick.Begou@grenoble-inp.fr">mailto:Patrick.Begou@grenoble-inp.fr</a> |
|  LEGI                    |                                      |
|  BP 53 X                 | Tel 04 76 82 51 35                   |
|  38041 GRENOBLE CEDEX    | Fax 04 76 82 52 71                   |
===================================================================
</pre></body></html>