<html>
  <head>

    <meta http-equiv="content-type" content="text/html; charset=UTF-8">
  </head>
  <body>
    <p>Friends,</p>
    <p>Please find the minutes of today's meeting <a
        moz-do-not-send="true"
href="https://halldweb.jlab.org/wiki/index.php/GlueX_Software_Meeting,_July_7,_2020#Minutes">here</a>
      and below.</p>
    <p>  -- Mark</p>
    <p>
    </p>
    <div id="globalWrapper">
      <div id="column-content">
        <div id="content" class="mw-body" role="main">
          <h2 id="firstHeading" class="firstHeading" lang="en"><span
              dir="auto">GlueX Software Meeting, July 7, 2020, </span><span
              class="mw-headline" id="Minutes">Minutes</span></h2>
          <div id="bodyContent" class="mw-body-content">
            <div id="mw-content-text" dir="ltr" class="mw-content-ltr"
              lang="en">
              <p>Present: Alex Austregesilo, Thomas Britton, Sean Dobbs,
                Mark Ito (chair), Igal Jaegle, Richard Jones, Naomi
                Jarvis, David Lawrence, Keigo Mizutani, Susan Schadmand,
                Simon Taylor, Nilanga Wickramaarachchi, Beni Zihlmann
              </p>
              <p>There is <a rel="nofollow" class="external text"
                  href="https://bluejeans.com/s/MR_xjcUQ0sR/">a
                  recording of his meeting</a> on the BlueJeans site.
                Use your JLab credentials to authenticate.
              </p>
              <h3><span class="mw-headline" id="Announcements">Announcements</span></h3>
              <ol>
                <li> <a rel="nofollow" class="external text"
href="https://mailman.jlab.org/pipermail/halld-offline/2020-June/004104.html">New
                    version set with upgrade to Geant4:
                    version_4.21.3.xml</a> Ready for testers of an
                  updated version of Geant4.</li>
                <li> <a rel="nofollow" class="external text"
href="https://mailman.jlab.org/pipermail/halld-offline/2020-July/004107.html">multi-threaded
                    hdgeant4 now working</a> Richard implemented patches
                  to Geant4 that fixed issues that prevented
                  multi-threaded running from giving sensible results.
                  <ul>
                    <li> To address thread safety in the HDGeant4 code
                      proper, he made a change to provide each thread
                      with its own copy of the magnetic field map at the
                      cost of 300 MB of memory per additional thread.
                      This was done out of an abundance of caution; if
                      not necessary he will back out the change to
                      recover the memory used.</li>
                    <li> Richard proposed changing the CCDB to so that
                      the magnetic field is no longer suppressed inside
                      selected volumes, in particular in the BCAL. Other
                      volumes that would get changed are the FCAL, TOF,
                      DIRC, and CCAL where the effect was not manifest
                      because the field is so much lower there. He
                      discovered that the suppression resulted field
                      discontinuities that caused crashes in particle
                      propagation reported by Naomi. The suppression had
                      originally been introduced to speed up shower
                      development in GEANT, but Geant4 seems not to be
                      affected by having the field on. We endorse the
                      proposal.</li>
                  </ul>
                </li>
                <li> <a rel="nofollow" class="external text"
href="https://mailman.jlab.org/pipermail/halld-offline/2020-July/004108.html">New
                    release of Build Scripts: version 1.58</a> Although
                  this new release will accommodate future versions
                  Geant4, it currently has a problem with our default
                  version, 10.02.p02. Mark addressed this presently.</li>
                <li> <a rel="nofollow" class="external text"
href="https://mailman.jlab.org/pipermail/halld-offline/2020-July/004109.html">New
                    version set: version_4.22.0.xml</a>. This version
                  set incorporates the fix to multi-threaded running of
                  HDGeant4 mentioned above.</li>
                <li> <a rel="nofollow" class="external text"
                    href="https://halldweb.jlab.org/talks/2020/crc_md5.pdf">Checksum
                    changes for files written to tape</a> The Computer
                  Center is dropping MD5 checksum and will rely on CRC32
                  sums for tape data validation.</li>
              </ol>
              <h3><span class="mw-headline"
                  id="Review_of_Minutes_from_the_Last_Software_Meeting">Review
                  of Minutes from the Last Software Meeting</span></h3>
              <p>We went over <a
href="https://halldweb.jlab.org/wiki/index.php/GlueX_Software_Meeting,_June_9,_2020#Minutes"
                  title="GlueX Software Meeting, June 9, 2020">the
                  minutes from June 9</a>.
              </p>
              <h4><span class="mw-headline"
                  id="New_Release_of_JANA:_version_0.8.2">New Release of
                  JANA: version 0.8.2</span></h4>
              <p>David has addressed the request for suppressing of
                warnings when geometry paths are requested but not
                present. This is often a normal situation, e.g., when
                probing the geometry to see how the data should be
                analyzed. His fix is in a new release of JANA, version
                0.8.2.
              </p>
              <h4><span class="mw-headline" id="Jana_2.0">Jana 2.0</span></h4>
              <p>Nathan Brei continues work on porting halld_recon to
                use JANA 2.0. We will get a report at a future meeting
                on this new major release.
              </p>
              <h3><span class="mw-headline"
                  id="Report_from_Recent_HDGeant4_Meetings">Report from
                  Recent HDGeant4 Meetings</span></h3>
              <p>We went over <a
href="https://halldweb.jlab.org/wiki/index.php/HDGeant4_Meeting,_June_16,_2020#Minutes"
                  title="HDGeant4 Meeting, June 16, 2020">the minutes
                  from the June 16 meeting</a> and <a
href="https://halldweb.jlab.org/wiki/index.php/HDGeant4_Meeting,_June_30,_2020#Minutes"
                  title="HDGeant4 Meeting, June 30, 2020">those from the
                  June 30 meeting</a> without much comment.
              </p>
              <h3><span class="mw-headline"
                  id="Report_from_the_SciComp_Meeting">Report from the
                  SciComp Meeting</span></h3>
              <p>Please see <a rel="nofollow" class="external text"
href="https://docs.google.com/presentation/d/1bmy-IYamSmEtv4HsOeXHgceD66vVPstLZXymLNf4tE0/edit?usp=sharing">Mark's
                  slides</a> for Scientific Computing news from the
                Computer Center.
              </p>
              <h3><span class="mw-headline" id="NERSC_status">NERSC
                  status</span></h3>
              <p>David brought us up-to-date on processing at NERSC.
                Please see <a rel="nofollow" class="external text"
href="https://docs.google.com/presentation/d/1BGYUvKztfzSGDGwcfTrs6DoS_EUUvaeTHReeYFAEKvc/edit?usp=sharing">his
                  slides</a> for all of the details. Some broad points:
              </p>
              <ul>
                <li> We have moved to two-hour jobs rather than the 6 to
                  8 hour jobs run in past campaigns. This gives us much
                  better access to the backfill mechanism at NERSC.</li>
                <li> The move involved significant development to our
                  workflow to (a) process only selected parts of our 20
                  GB raw data files in a single job and (b) recombine
                  the resulting output files to correspond to the
                  original 20 GB file.</li>
                <li> One significant challenge has been to run
                  monitoring launches, with their 57 plugins, to give
                  complete ROOT output files for each selection of the
                  raw data file.</li>
                <li> David pointed out that the current version of the
                  Oasis image of our software does not support
                  development (i.e., building new versions of software),
                  only running. He has put up a Docker container that
                  remedies this. Richard has also run into this issue;
                  he has put the missing pieces in an undisclosed
                  location on Oasis.
                  <ul>
                    <li> Mark remarked that having a developer-friendly
                      version of Oasis would involve very little work,
                      only real estate on Oasis. He will look into this.
                    </li>
                  </ul>
                </li>
                <li> Alex suggested that one way forward is to abandon
                  processing of monitoring launches at NERSC and
                  concentrate on REST file production which uses a
                  smaller set of plugins, plugins that have had better
                  past records of success.</li>
              </ul>
              <h3><span class="mw-headline" id="OSG_Jobs_and_mcsmear">OSG
                  Jobs and mcsmear</span></h3>
              <p>Today, Thomas noticed that many jobs were crashing in
                mcsmear when accessing the SQLite form of the CCDB on
                the OSG. Two issues:
              </p>
              <ol>
                <li> What is wrong with the SQLite file?</li>
                <li> Why is it that mcsmear exits with status code = 0
                  (i.e., success) after bombing?</li>
              </ol>
              <p>We were only able to address the first issue. David
                noticed that today's SQLite file was a bit smaller than
                usual, making it suspect. Mark promised to recreate the
                SQLite file and ship it out via Oasis.
                [Added in press: (a) Mark <a rel="nofollow"
                  class="external text"
href="https://mailman.jlab.org/pipermail/halld-offline/2020-July/004115.html">made
                  good on his promise</a> and regenerated the SQLite
                file. (b) David had reported this issue via email to
                Mark early this morning. Suffice it to say that Mark is
                behind on his email.]
              </p>
            </div>
            <div class="printfooter">
              Retrieved from "<a dir="ltr"
href="https://halldweb.jlab.org/wiki/index.php?title=GlueX_Software_Meeting,_July_7,_2020&oldid=100104">https://halldweb.jlab.org/wiki/index.php?title=GlueX_Software_Meeting,_July_7,_2020&oldid=100104</a>"</div>
          </div>
        </div>
      </div>
      <div id="footer" role="contentinfo">
        <ul id="f-list">
          <li id="lastmod"> This page was last modified on 7 July 2020,
            at 19:25.</li>
        </ul>
      </div>
    </div>
  </body>
</html>