<html><head></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; color: rgb(0, 0, 0); font-size: 14px; font-family: Calibri, sans-serif; "><div>Done.</div><div><br></div><div>Thanks,</div><div>Carson</div><div><br></div><div><br></div><span id="OLK_SRC_BODY_SECTION"><div style="font-family:Calibri; font-size:11pt; text-align:left; color:black; BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt"><span style="font-weight:bold">From: </span> "Krishnakumar, Vivek" <<a href="mailto:vkrishnakumar@jcvi.org">vkrishnakumar@jcvi.org</a>><br><span style="font-weight:bold">Date: </span> Thursday, 24 January, 2013 4:29 PM<br><span style="font-weight:bold">To: </span> Carson Holt <<a href="mailto:carsonhh@gmail.com">carsonhh@gmail.com</a>><br><span style="font-weight:bold">Cc: </span> "<a href="mailto:mcampbell@genetics.utah.edu">mcampbell@genetics.utah.edu</a>" <<a href="mailto:mcampbell@genetics.utah.edu">mcampbell@genetics.utah.edu</a>><br><span style="font-weight:bold">Subject: </span> Re: [maker-devel] DBD::SQLite::db do failed: database is locked (GFFDB.pm lines 407, 408, 497) error - Using MAKER 2.27b MPI<br></div><div><br></div><div>
  
    <meta content="text/html; charset=ISO-8859-1" http-equiv="Content-Type">
  
  <div bgcolor="#FFFFFF" text="#000000">
    Ok. Seems to be working fine. I pointed the TMP parameter to our
    scratch area.<br>
    <br>
    Thanks for your suggestion.<br>
    Should I forward this conversation thread to the maker-devel list ?<br>
    <br>
    Vivek<br>
    <br>
    <br>
    <div class="moz-cite-prefix">On 01/24/2013 04:08 PM, Carson Holt
      wrote:<br>
    </div>
    <blockquote cite="mid:CD270651.17FB2%25carsonhh@gmail.com" type="cite">
      <div>SQLlite will choke almost immediately if it's not set up with
        the correct flags, so you will know right away.</div>
      <div><br>
      </div>
      <div>--Carson</div>
      <div><br>
      </div>
      <span id="OLK_SRC_BODY_SECTION">
        <div style="font-family:Calibri; font-size:11pt;
          text-align:left; color:black; BORDER-BOTTOM: medium none;
          BORDER-LEFT: medium none; PADDING-BOTTOM: 0in; PADDING-LEFT:
          0in; PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt solid;
          BORDER-RIGHT: medium none; PADDING-TOP: 3pt"><span style="font-weight:bold">From: </span> "Krishnakumar,
          Vivek" <<a moz-do-not-send="true" href="mailto:vkrishnakumar@jcvi.org">vkrishnakumar@jcvi.org</a>><br>
          <span style="font-weight:bold">Date: </span> Thursday, 24
          January, 2013 3:35 PM<br>
          <span style="font-weight:bold">To: </span> Carson Holt <<a moz-do-not-send="true" href="mailto:carsonhh@gmail.com">carsonhh@gmail.com</a>><br>
          <span style="font-weight:bold">Cc: </span> "<a moz-do-not-send="true" href="mailto:mcampbell@genetics.utah.edu">mcampbell@genetics.utah.edu</a>"
          <<a moz-do-not-send="true" href="mailto:mcampbell@genetics.utah.edu">mcampbell@genetics.utah.edu</a>><br>
          <span style="font-weight:bold">Subject: </span> Re:
          [maker-devel] DBD::SQLite::db do failed: database is locked
          (GFFDB.pm lines 407, 408, 497) error - Using MAKER 2.27b MPI<br>
        </div>
        <div><br>
        </div>
        <div>
          <meta content="text/html; charset=ISO-8859-1" http-equiv="Content-Type">
          <div bgcolor="#FFFFFF" text="#000000"> Okay. I will try doing
            that. But how can I find out that my NFS mounted area has
            -nolock set? The filesystem areas that we work on are
            managed by our IT department.<br>
            <br>
            Thanks.<br>
            Vivek<br>
            <br>
            <div class="moz-cite-prefix">On 01/24/2013 03:27 PM, Carson
              Holt wrote:<br>
            </div>
            <blockquote cite="mid:CD2702CC.17F93%25carsonhh@gmail.com" type="cite">
              <div>If you don't have locally mounted space big enough,
                then just use an NFS mounted /TMP.  You will take a
                slight performance hit, but it may still work if it is
                mounted with -nolock set so SQLite stops complaining.</div>
              <div><br>
              </div>
              <div>--Carson </div>
              <div><br>
              </div>
              <span id="OLK_SRC_BODY_SECTION">
                <div style="font-family:Calibri; font-size:11pt;
                  text-align:left; color:black; BORDER-BOTTOM: medium
                  none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0in;
                  PADDING-LEFT: 0in; PADDING-RIGHT: 0in; BORDER-TOP:
                  #b5c4df 1pt solid; BORDER-RIGHT: medium none;
                  PADDING-TOP: 3pt"><span style="font-weight:bold">From:
                  </span> "Krishnakumar, Vivek" <<a moz-do-not-send="true" href="mailto:vkrishnakumar@jcvi.org">vkrishnakumar@jcvi.org</a>><br>
                  <span style="font-weight:bold">Date: </span>
                  Thursday, 24 January, 2013 3:24 PM<br>
                  <span style="font-weight:bold">To: </span> Carson
                  Holt <<a moz-do-not-send="true" href="mailto:carsonhh@gmail.com">carsonhh@gmail.com</a>><br>
                  <span style="font-weight:bold">Cc: </span> "<a moz-do-not-send="true" href="mailto:mcampbell@genetics.utah.edu">mcampbell@genetics.utah.edu</a>"
                  <<a moz-do-not-send="true" href="mailto:mcampbell@genetics.utah.edu">mcampbell@genetics.utah.edu</a>><br>
                  <span style="font-weight:bold">Subject: </span> Re:
                  [maker-devel] DBD::SQLite::db do failed: database is
                  locked (GFFDB.pm lines 407, 408, 497) error - Using
                  MAKER 2.27b MPI<br>
                </div>
                <div><br>
                </div>
                <div>
                  <meta content="text/html; charset=ISO-8859-1" http-equiv="Content-Type">
                  <div bgcolor="#FFFFFF" text="#000000"> I no longer see
                    that warning.<br>
                    <br>
                    The only thing I see is that MAKER now complains
                    that the DB is corrupt, and that is because I only
                    have 5GB of /tmp space on the server I currently
                    have access to and that is not enough for the whole
                    genome FASTA + est2genome + protein2genome +
                    model_gff (EVM).<br>
                    <br>
                    Thanks,<br>
                    Vivek<br>
                    <br>
                    <br>
                    <div class="moz-cite-prefix">On 01/24/2013 03:20 PM,
                      Carson Holt wrote:<br>
                    </div>
                    <blockquote cite="mid:CD27013E.17F86%25carsonhh@gmail.com" type="cite">
                      <div>You should be able to run on an NFS drive.
                         After creating the initial SQLite database,
                        maker now gives each node it's own copy that is
                        stored in the TMP directory (should be local to
                        each node).  Could you send me the GFF3 file
                        that gives the warning, so I can look at it.</div>
                      <div><br>
                      </div>
                      <div>Thanks,</div>
                      <div>Carson</div>
                      <div><br>
                      </div>
                      <div><br>
                      </div>
                      <span id="OLK_SRC_BODY_SECTION">
                        <div style="font-family:Calibri; font-size:11pt;
                          text-align:left; color:black; BORDER-BOTTOM:
                          medium none; BORDER-LEFT: medium none;
                          PADDING-BOTTOM: 0in; PADDING-LEFT: 0in;
                          PADDING-RIGHT: 0in; BORDER-TOP: #b5c4df 1pt
                          solid; BORDER-RIGHT: medium none; PADDING-TOP:
                          3pt"><span style="font-weight:bold">From: </span>
                          "Krishnakumar, Vivek" <<a moz-do-not-send="true" href="mailto:vkrishnakumar@jcvi.org">vkrishnakumar@jcvi.org</a>><br>
                          <span style="font-weight:bold">Date: </span>
                          Thursday, 24 January, 2013 3:17 PM<br>
                          <span style="font-weight:bold">To: </span>
                          Carson Holt <<a moz-do-not-send="true" href="mailto:carsonhh@gmail.com">carsonhh@gmail.com</a>><br>
                          <span style="font-weight:bold">Cc: </span> "<a moz-do-not-send="true" href="mailto:mcampbell@genetics.utah.edu">mcampbell@genetics.utah.edu</a>"
                          <<a moz-do-not-send="true" href="mailto:mcampbell@genetics.utah.edu">mcampbell@genetics.utah.edu</a>><br>
                          <span style="font-weight:bold">Subject: </span>
                          Re: [maker-devel] DBD::SQLite::db do failed:
                          database is locked (GFFDB.pm lines 407, 408,
                          497) error - Using MAKER 2.27b MPI<br>
                        </div>
                        <div><br>
                        </div>
                        <div>
                          <meta content="text/html; charset=ISO-8859-1" http-equiv="Content-Type">
                          <div bgcolor="#FFFFFF" text="#000000"> Hi
                            Carson,<br>
                            <br>
                            I did make sure that every time I reran the
                            pipeline, I deleted the maker.output
                            directory and the temporary files in /tmp.
                            That way I know that the DBs get created
                            fresh.<br>
                            <br>
                            When it was openMPI based, I noticed that
                            irrespective of whether I enabled or
                            disabled the `-TMP` parameter, I would get
                            the database locked error. At this point,
                            mpicc was openMPI based and mpiexec (hydra)
                            was MPICH2 based.<br>
                            <br>
                            On installing MPICH2 as a non-root user and
                            rebuilding `maker`, this issue vanished.
                            Just an observation.<br>
                            <br>
                            The issue I have now with scaling up is that
                            I do not currently have access to a server
                            that has a large amount of non-NFS mounted
                            storage space (we do have such a server but
                            it has been reserved for a different project
                            by a group in our institute). Thus, the
                            creation of the SQLite db is halted mid way
                            and now MAKER complains that the database is
                            corrupt. I can either run the chromosomes
                            individually (serially) and then the
                            scaffolds as one group or wait to get a time
                            slot of that server.<br>
                            <br>
                            Will let you know if I have any more issues.<br>
                            <br>
                            Thank you.<br>
                            Vivek<br>
                            <br>
                            <br>
                            <div class="moz-cite-prefix">On 01/24/2013
                              03:02 PM, Carson Holt wrote:<br>
                            </div>
                            <blockquote cite="mid:CD26FB98.17F6C%25carsonhh@gmail.com" type="cite">
                              <div>If you still see issue, let me know.
                                 I  have a hard time seeing how just
                                moving to MPICH2 would have solved the
                                issue. It may also have been an initial
                                problem with the first GFF# db being
                                created and running on a new job
                                recreated it correctly.  Either way, let
                                me know if it comes up on your bigger
                                run.</div>
                              <div><br>
                              </div>
                              <div>Thanks,</div>
                              <div>Carson</div>
                              <div><br>
                              </div>
                              <div><br>
                              </div>
                              <span id="OLK_SRC_BODY_SECTION">
                                <div style="font-family:Calibri;
                                  font-size:11pt; text-align:left;
                                  color:black; BORDER-BOTTOM: medium
                                  none; BORDER-LEFT: medium none;
                                  PADDING-BOTTOM: 0in; PADDING-LEFT:
                                  0in; PADDING-RIGHT: 0in; BORDER-TOP:
                                  #b5c4df 1pt solid; BORDER-RIGHT:
                                  medium none; PADDING-TOP: 3pt"><span style="font-weight:bold">From: </span>
                                  "Krishnakumar, Vivek" <<a moz-do-not-send="true" href="mailto:vkrishnakumar@jcvi.org">vkrishnakumar@jcvi.org</a>><br>
                                  <span style="font-weight:bold">Date: </span>
                                  Thursday, 24 January, 2013 2:49 PM<br>
                                  <span style="font-weight:bold">To: </span>
                                  <<a moz-do-not-send="true" href="mailto:maker-devel@yandell-lab.org">maker-devel@yandell-lab.org</a>>,



                                  <<a moz-do-not-send="true" href="mailto:mcampbell@genetics.utah.edu">mcampbell@genetics.utah.edu</a>><br>
                                  <span style="font-weight:bold">Subject:
                                  </span> Re: [maker-devel]
                                  DBD::SQLite::db do failed: database is
                                  locked (GFFDB.pm lines 407, 408, 497)
                                  error - Using MAKER 2.27b MPI<br>
                                </div>
                                <div><br>
                                </div>
                                <div>
                                  <meta content="text/html;
                                    charset=ISO-8859-1" http-equiv="Content-Type">
                                  <div bgcolor="#FFFFFF" text="#000000">
                                    Hi Everybody,<br>
                                    <br>
                                    Sorry for jumping the gun and
                                    writing to the mailing list before
                                    completely troubleshooting this
                                    issue. I was able to figure out the
                                    problem and get MAKER 2.27b working
                                    fine in MPI mode. <br>
                                    <br>
                                    The issue was with the incorrect
                                    version of mpicc (it was openMPI
                                    based and not mpich2). I was able to
                                    successfully run through the steps
                                    described below on one chromosome in
                                    under a few hours (~2). Now I'm
                                    scaling it to the whole genome (8
                                    chromosomes and ~2200 unanchored
                                    scaffolds).<br>
                                    <br>
                                    Thank you.<br>
                                    Vivek<br>
                                    <br>
                                    <div class="moz-cite-prefix">On
                                      01/23/2013 02:42 PM, Krishnakumar,
                                      Vivek wrote:<br>
                                    </div>
                                    <blockquote cite="mid:51003D32.4080908@jcvi.org" type="cite">
                                      <meta http-equiv="content-type" content="text/html;
                                        charset=ISO-8859-1">
                                      Hello,<br>
                                      <br>
                                      I'm trying to use the MAKER
                                      pipeline to calculate AED scores
                                      for the gene models coming out of
                                      our annotation pipeline (EVidence
                                      Modeler). We then want to compare
                                      these computed AED scores with
                                      those computed for the gene models
                                      coming out of the MAKER2 pipeline.<br>
                                      <br>
                                      For this purpose, I configured the
                                      maker_opts.ctl file like so:<br>
                                      1) Specify gff file coming from
                                      our annotation pipeline using
                                      `model_gff` param <br>
                                      2) Specify gff file containing
                                      est2genome results (computed using
                                      MAKER2) using the `est_gff` param<br>
                                      3) Specify gff file containing
                                      protein2genome results (computed
                                      using MAKER2) using the
                                      `protein_gff` param<br>
                                      4) Specify the genome fasta
                                      sequence file using the `genome`
                                      param<br>
                                      <br>
                                      I recently configured MAKER 2.27b
                                      to work using MPI and I tried
                                      launching one contig (chromosome)
                                      on a compute node spawning 12
                                      processes like so:<br>
                                      <big><tt>mpiexec -n 12 maker -R
                                          -TMP /tmp/maker maker_opts.ctl
                                          maker_bopts.ctl maker_exe.ctl
                                          2>&1 | tee maker.err</tt></big><br>
                                      <br>
                                      Side note: Without MPI, it works
                                      fine but takes very very long to
                                      complete running on one chromosome
                                      (since it is going through the
                                      chunks serially).<br>
                                      <br>
                                      On inspecting the `maker.err`
                                      file, I see server error messages
                                      that say `DBD::SQLite::db do
                                      failed: database is locked`. On
                                      referring to a previous email
                                      thread on the `maker-devel`
                                      mailing list, I see that Carson
                                      had replied to the person with a
                                      similar issue asking them to check
                                      if the working directory and TMP
                                      directory are being mounted on an
                                      NFS filesystem because this is a
                                      potential problem with SQLite
                                      databases.<br>
                                      <br>
                                      Based on that suggestion, I
                                      switched the TMP directory to use
                                      a local filesytem but the working
                                      directory is still NFS mounted.
                                      Despite that, I still get this
                                      error. The previous email thread
                                      had no resolution for this issue.
                                      My question is, should the working
                                      directory be local as well? I see
                                      that despite setting the `TMP`
                                      variable, the SQLite DB is getting
                                      created in the working directory.<br>
                                      <br>
                                      What might be the issue?<br>
                                      <br>
                                      Also, apart from this one error, I
                                      notice another error:<br>
                                      <big><tt>ERROR: Non-unique top
                                          level ID for chr5:hit:1454:4_0<br>
                                          While this is technically
                                          legal in GFF3, it usually<br>
                                          indicates a poorly fomatted
                                          GFF3 file (perhaps you<br>
                                          tried to merge two GFF3 files
                                          without accounting for<br>
                                          unique IDs).  MAKER will not
                                          handle these correctly.</tt></big><br>
                                      <br>
                                      On inspecting the input GFF file
                                      (protein2genome), I see that this
                                      ID is never repeated. So I have no
                                      idea whats going on here either.
                                      Any suggestions?<br>
                                      <br>
                                      Below is my `maker.err` file from
                                      the run. (Dropbox link)<br>
                                      <br>
                                      Thank you very much.<br>
                                      Best,<br>
                                      <br>
                                      Vivek<br>
                                      <br>
                                      --<br>
                                      Vivek Krishnakumar<br>
                                      <a moz-do-not-send="true" class="moz-txt-link-abbreviated" href="mailto:vkrishnakumar@jcvi.org">vkrishnakumar@jcvi.org</a><br>
                                      <br>
                                      <div style="padding: 15px;
                                        background-color: rgb(217, 237,
                                        255);">
                                        <div style="margin-bottom:
                                          15px;">I've linked 1 file to
                                          this email:</div>
                                        <div style="background-color:
                                          rgb(255, 255, 255); padding:
                                          15px;">
                                          <div style="border: 1px solid
                                            rgb(205, 205, 205);
                                            border-radius: 5px 5px 5px
                                            5px; margin-top: 10px;
                                            margin-bottom: 10px;
                                            padding: 15px;" class="cloudAttachmentItem"><img src="cid:part17.07020406.00030802@jcvi.org" style="margin-right: 5px;
                                              float: left; width: 24px;
                                              height: 24px;"><a moz-do-not-send="true" style="color: rgb(15, 126,
                                              219) ! important;" href="http://db.tt/2tYyKCD7">maker.err</a><span style="margin-left: 5px;
                                              font-size: small; color:
                                              grey;">(97.7 MB)</span><span style="float: right;"><img style="margin-right:
                                                5px;" src="cid:part19.00010506.01070403@jcvi.org"><a moz-do-not-send="true" style="color: rgb(15,
                                                126, 219) ! important;" href="https://www.dropbox.com/">Dropbox</a></span><a moz-do-not-send="true" style="color: rgb(15, 126,
                                              219) ! important;
                                              font-size: small; display:
                                              block;" href="http://db.tt/2tYyKCD7">http://db.tt/2tYyKCD7</a></div>
                                        </div>
                                        <div style="color: rgb(68, 68,
                                          68); font-size: small;
                                          margin-top: 15px;"><a moz-do-not-send="true" style="color: rgb(15, 126,
                                            219) ! important;" href="http://www.getthunderbird.com">Mozilla


                                            Thunderbird</a> makes it
                                          easy to share large files over
                                          email.</div>
                                      </div>
                                      <br>
                                    </blockquote>
                                    <br>
                                  </div>
                                </div>
                                _______________________________________________

                                maker-devel mailing list <a moz-do-not-send="true" href="mailto:maker-devel@box290.bluehost.com">maker-devel@box290.bluehost.com</a>
                                <a moz-do-not-send="true" href="http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org">http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org</a>
                              </span> </blockquote>
                            <br>
                          </div>
                        </div>
                      </span> </blockquote>
                    <br>
                  </div>
                </div>
              </span> </blockquote>
            <br>
          </div>
        </div>
      </span>
    </blockquote>
    <br>
  </div></div></span></body></html>