<html>
  <head>
    <meta content="text/html; charset=windows-1252"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">SIGSEGV errors can be not easy to
      solve, because they can have many possible causes:<br>
      <br>
      <a class="moz-txt-link-freetext" href="http://www.democritos.it/pipermail/pw_forum/2005-March/002272.html">http://www.democritos.it/pipermail/pw_forum/2005-March/002272.html</a><br>
<a class="moz-txt-link-freetext" href="http://software.intel.com/en-us/articles/determining-root-cause-of-sigsegv-or-sigbus-errors/">http://software.intel.com/en-us/articles/determining-root-cause-of-sigsegv-or-sigbus-errors/</a><br>
      <br>
      However, maybe the cause is the same as a similar case that
      occurred before where the error occurred when -lmkl_blacs_lp64 was
      used when -lmkl_blacs_openmpi_lp64 should have been used instead
      for Open MPI [
<a class="moz-txt-link-freetext" href="http://www.mail-archive.com/wien%40zeus.theochem.tuwien.ac.at/msg12746.html">http://www.mail-archive.com/wien%40zeus.theochem.tuwien.ac.at/msg12746.html</a>
      ].<br>
      <br>
      What blacs did you use in your parallel compiler options (for
      RP_LIB) when you compiled WIEN2k using Open MPI?<br>
      <br>
      On 6/22/2016 7:30 AM, Md. Fhokrul Islam wrote:<br>
    </div>
    <blockquote
cite="mid:HK2PR04MB0868B8001161E0B925843967D32C0@HK2PR04MB0868.apcprd04.prod.outlook.com"
      type="cite">
      <meta http-equiv="Content-Type" content="text/html;
        charset=windows-1252">
      <style type="text/css" style="display:none;"><!-- P {margin-top:0;margin-bottom:0;} --></style>
      <div id="divtagdefaultwrapper"
style="font-size:12pt;color:#000000;background-color:#FFFFFF;font-family:Calibri,Arial,Helvetica,sans-serif;">
        <p>Hi Prof Blaha,</p>
        <p><br>
        </p>
        <p>    I have compiled MPI version of Wien2k_14.2 with OpenMPI <span
            style="font-size: 12pt;">and have got all </span><span
            style="font-size: 12pt;">MPI executables. </span></p>
        <p><span style="font-size: 12pt;">But when I run a test
            calculation using 4 </span>
          <span style="font-size: 12pt;">cores</span><span
            style="font-size: 12pt;">, I get the the </span><span
            style="font-size: 12pt;">following error message. Could you</span></p>
        <p><span style="font-size: 12pt;">please let me know what should
            I do to fix this problem. </span></p>
        <p><br>
        </p>
        <p>case.dayfile</p>
        <p><br>
        </p>
        <div>   start       (Tue Jun 21 20:57:01 CEST 2016) with lapw0
          (100/99 to go)</div>
        <div><br>
        </div>
        <div>    cycle 1     (Tue Jun 21 20:57:01 CEST 2016)        
          (100/99 to go)</div>
        <div><br>
        </div>
        <div>>   lapw0 -p    (20:57:01) starting parallel lapw0 at
          Tue Jun 21 20:57:02 CEST 2016</div>
        <div>-------- .machine0 : 4 processors</div>
        <div>4.684u 0.733s 0:04.50 120.2%    0+0k 220432+4688io 78pf+0w</div>
        <div>>   lapw1  -up -p    -c         (20:57:06) starting
          parallel lapw1 at Tue Jun 21 20:57:06 CEST 2016</div>
        <div>->  starting parallel LAPW1 jobs at Tue Jun 21 20:57:06
          CEST 2016</div>
        <div>running LAPW1 in parallel mode (using .machines)</div>
        <div>1 number_of_parallel_jobs</div>
        <div>     au063 au063 au063 au063(47)  Child id           3
          SIGSEGV</div>
        <div> Child id           0 SIGSEGV</div>
        <div> Child id           1 SIGSEGV</div>
        <div> Child id           2 SIGSEGV</div>
        <div>--------------------------------------------------------------------------</div>
        <div>MPI_ABORT was invoked on rank 0 in communicator
          MPI_COMM_WORLD</div>
        <div>with errorcode 1543169252.</div>
        <div><br>
        </div>
        <br>
        <p>Error file:</p>
        <p><br>
        </p>
        <div>LAPW0 END</div>
        <div>w2k_dispatch_signal(): received: Segmentation fault</div>
        <div>w2k_dispatch_signal(): received: Segmentation fault</div>
        <div>w2k_dispatch_signal(): received: Segmentation fault</div>
        <div>w2k_dispatch_signal(): received: Segmentation fault</div>
        <div>[au063:29993] 3 more processes have sent help message
          help-mpi-api.txt / mpi-abort</div>
        <div>[au063:29993] Set MCA parameter "orte_base_help_aggregate"
          to 0 to see all help / error messages</div>
        <div><br>
        </div>
        <br>
        <p>Thanks,</p>
        <p>Fhokrul</p>
        <p><br>
        </p>
        <p><br>
        </p>
        <br>
        <br>
        <div style="color: rgb(0, 0, 0);">
          <div>
            <hr tabindex="-1" style="display:inline-block; width:98%">
            <div id="x_divRplyFwdMsg" dir="ltr"><font
                style="font-size:11pt" face="Calibri, sans-serif"
                color="#000000"><b>From:</b> Wien
                <a class="moz-txt-link-rfc2396E" href="mailto:wien-bounces@zeus.theochem.tuwien.ac.at"><wien-bounces@zeus.theochem.tuwien.ac.at></a> on
                behalf of Peter Blaha
                <a class="moz-txt-link-rfc2396E" href="mailto:pblaha@theochem.tuwien.ac.at"><pblaha@theochem.tuwien.ac.at></a><br>
                <b>Sent:</b> Monday, June 13, 2016 8:54 AM<br>
                <b>To:</b> A Mailing list for WIEN2k users<br>
                <b>Subject:</b> Re: [Wien] Running parallel job with
                Slurm+intel mpi</font>
              <div> </div>
            </div>
          </div>
          <font size="2"><span style="font-size:10pt;">
              <div class="PlainText">No, the srun command cannot work
                with these options, but need other
                <br>
                switches.<br>
                Most likely in WIEN2k_14.2 you will have to use  
                mpirun, or try:<br>
                <br>
                srun -n_NP_   (i.e. without the blank between n and
                _NP_)<br>
                <br>
                <br>
                In the next version, we will have support for srun, but
                at least for <br>
                mixed k-point+mpi-support it requires also a small
                change in the <br>
                lapw*para scripts.<br>
                <br>
                <br>
                <br>
                On 06/09/2016 04:29 PM, Md. Fhokrul Islam wrote:<br>
                > Dear Wien2k users,<br>
                ><br>
                ><br>
                >       I am trying to do some calculation with a
                large system with mpi<br>
                > version but having problem with<br>
                ><br>
                > running the job. I have compiled Wien2k 14.2 with
                intel libraries and<br>
                > have generated all mpi executables.<br>
                ><br>
                > But our system requires us to use srun instead of
                mpirun. So I have<br>
                > changed parallel options to<br>
                ><br>
                ><br>
                > srun -n _NP_ -machinefile _HOSTS_ _EXEC_.<br>
                ><br>
                ><br>
                > I also have tried other options that I saw in the
                mailing list<br>
                ><br>
                ><br>
                >      srun -n _NP_<br>
                ><br>
                ><br>
                > but that didn't work. No new files are created and
                the dayfile is stuck<br>
                ><br>
                > in lapw0.<br>
                ><br>
                ><br>
                > case.dayfile:<br>
                > ----------------------------------<br>
                > Calculating GaAs in
                /lunarc/nobackup/users/eishfh/WIEN2k/test/GaAs<br>
                > on au054 with PID 189944<br>
                > using WIEN2k_14.2 (Release 15/10/2014) in<br>
                > /lunarc/nobackup/users/eishfh/SRC/Wien2k14.2-mpi<br>
                ><br>
                ><br>
                >      start       (Thu Jun  9 13:14:39 CEST 2016)
                with lapw0 (100/99 to go)<br>
                ><br>
                >      cycle 1     (Thu Jun  9 13:14:39 CEST
                2016)         (100/99 to go)<br>
                ><br>
                >  >   lapw0 -p    (13:14:39) starting parallel
                lapw0 at Thu Jun  9<br>
                > 13:14:39 CEST 2016<br>
                > -------- .machine0 : 4 processors<br>
                > -------------------<br>
                ><br>
                ><br>
                > I understood from the userguide that the -p option
                in runsp_lapw picks up<br>
                ><br>
                > mpi version depending on the form of the .machines
                file. Here is the<br>
                ><br>
                > .machines file that I have used for this test
                calculation.<br>
                ><br>
                ><br>
                > #<br>
                > lapw0:au165 au165 au165 au165<br>
                > 1:au165 au165 au165 au165<br>
                > granularity:1<br>
                > extrafine:1<br>
                > lapw2_vector_split:2<br>
                ><br>
                > So I am wondering if anyone can tell me how can I
                fix the problem.<br>
                ><br>
                > Thanks,<br>
                ><br>
                > Fhokrul<br>
                ><br>
                ><br>
                ><br>
                ><br>
                > _______________________________________________<br>
                > Wien mailing list<br>
                > <a class="moz-txt-link-abbreviated" href="mailto:Wien@zeus.theochem.tuwien.ac.at">Wien@zeus.theochem.tuwien.ac.at</a><br>
                > <a moz-do-not-send="true"
                  href="http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien"
                  id="LPlnk527074">
http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien</a>
                <div id="LPBorder_GT_14666011401310.3932420394752014"
                  style="margin-bottom: 20px; overflow: auto; width:
                  100%; text-indent: 0px;">
                  <table
                    id="LPContainer_14666011401220.3377222606458159"
                    style="width: 90%; position: relative; overflow:
                    auto; padding-top: 20px; padding-bottom: 20px;
                    margin-top: 20px; border-top-width: 1px;
                    border-top-style: dotted; border-top-color: rgb(200,
                    200, 200); border-bottom-width: 1px;
                    border-bottom-style: dotted; border-bottom-color:
                    rgb(200, 200, 200); background-color: rgb(255, 255,
                    255);" cellspacing="0">
                    <tbody>
                      <tr style="border-spacing: 0px;" valign="top">
                        <td
                          id="TextCell_14666011401260.44635978717927016"
                          colspan="2" style="vertical-align: top;
                          position: relative; padding: 0px; display:
                          table-cell;">
                          <div
                            id="LPTitle_14666011401260.07450109101694036"
                            style="top: 0px; color: rgb(0, 120, 215);
                            font-weight: normal; font-size: 21px;
                            font-family: wf_segoe-ui_light, "Segoe
                            UI Light", "Segoe WP Light",
                            "Segoe UI", "Segoe WP",
                            Tahoma, Arial, sans-serif; line-height:
                            21px;">
                            <a moz-do-not-send="true"
                              id="LPUrlAnchor_14666011401280.26770733641546407"
href="http://zeus.theochem.tuwien.ac.at/mailman/listinfo/wien"
                              target="_blank" style="text-decoration:
                              none;">Wien -- A Mailing list for WIEN2k
                              users</a></div>
                          <div
                            id="LPMetadata_14666011401280.27608741457236086"
                            style="margin: 10px 0px 16px; color:
                            rgb(102, 102, 102); font-weight: normal;
                            font-family: wf_segoe-ui_normal, "Segoe
                            UI", "Segoe WP", Tahoma,
                            Arial, sans-serif; font-size: 14px;
                            line-height: 14px;">
                            zeus.theochem.tuwien.ac.at</div>
                          <div
                            id="LPDescription_14666011401290.9744229272922187"
                            style="display: block; color: rgb(102, 102,
                            102); font-weight: normal; font-family:
                            wf_segoe-ui_normal, "Segoe UI",
                            "Segoe WP", Tahoma, Arial,
                            sans-serif; font-size: 14px; line-height:
                            20px; max-height: 100px; overflow: hidden;">
                            A Mailing list for WIEN2k users. Please post
                            questions, suggestions or comments about
                            WIEN2k ONLY in this list. Please follow the
                            following "Nettiquette" (depending ...</div>
                        </td>
                      </tr>
                    </tbody>
                  </table>
                </div>
                <br>
                <br>
                > SEARCH the MAILING-LIST at:  <a
                  moz-do-not-send="true"
href="http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html"><a class="moz-txt-link-freetext" href="http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html">http://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/index.html</a></a><br>
                ><br>
                <br>
                -- <br>
                <br>
                                                       P.Blaha<br>
--------------------------------------------------------------------------<br>
                Peter BLAHA, Inst.f. Materials Chemistry, TU Vienna,
                A-1060 Vienna<br>
                Phone: +43-1-58801-165300             FAX:
                +43-1-58801-165982<br>
                Email: <a class="moz-txt-link-abbreviated" href="mailto:blaha@theochem.tuwien.ac.at">blaha@theochem.tuwien.ac.at</a>    WIEN2k: <a
                  moz-do-not-send="true" href="http://www.wien2k.at"><a class="moz-txt-link-freetext" href="http://www.wien2k.at">http://www.wien2k.at</a></a><br>
                WWW:   <a moz-do-not-send="true"
                  href="http://www.imc.tuwien.ac.at/staff/tc_group_e.php">http://www.imc.tuwien.ac.at/staff/tc_group_e.php</a><br>
--------------------------------------------------------------------------<br>
              </div>
            </span></font></div>
      </div>
    </blockquote>
  </body>
</html>