[Halld-offline] Software Meeting Minutes, June 22, 2021

[Mark Ito]

Folks,

Please find the minutes here 
<https://halldweb.jlab.org/wiki/index.php/GlueX_Software_Meeting,_June_22,_2021#Minutes> 
and below.

   -- Mark

      _________________________________________


    GlueX Software Meeting, June 22, 2021, Minutes

Present: Alexander Austregesilo, Thomas Britton, Sean Dobbs, Mark Ito 
(chair), Richard Jones, Simon Taylor, Beni Zihlmann

There is a recording of this meeting 
<https://bluejeans.com/s/HK7PJLxr0uW/>. Log into the BlueJeans site 
<https://jlab.bluejeans.com> first to gain access (use your JLab 
credentials).


      Announcements

 1. Updating Oasis for the GlueX VO
    <https://halldweb.jlab.org/wiki/index.php/Updating_Oasis_for_the_GlueX_VO>
    Mark recently documented the procedure for update the Oasis CVMFS
    share with GlueX software. This is the copy we use on the OSG and at
    the HPC centers.
 2. info on volatile, cache, and Hall D
    <https://mailman.jlab.org/pipermail/halld-offline/2021-May/008552.html>
    Mark emphasized the two points in the email: volatile is
    under-utilized and cache has accumulated a large number of "small"
    files.
 3. HOWTO use hdgeant4 as a beam Monte Carlo generator
    <https://halldweb.jlab.org/wiki/index.php/HOWTO_use_hdgeant4_as_a_beam_Monte_Carlo_generator>
    Richard reviewed his recent HOWTO on using the built-in photon beam
    generator. Featured applications include triplet production from the
    TPOL converter and Bethe-Heitler production from the hydrogen
    target. Simulated beam properties can be studied just before and
    just after the collimator, and before and after the TPOL converter,
    as well as point farther downstream. He has also written a companion
    document, HOWTO generate Bethe-Heitler pairs in the GlueX target
    <https://halldweb.jlab.org/wiki/index.php/HOWTO_generate_Bethe-Heitler_pairs_in_the_GlueX_target>,which
    gives guidance on using the event weighting necessary to study BH
    processes from the target that fall into the GlueX acceptance.
 4. HOWTO Install and Use the CVMFS Client
    <https://halldweb.jlab.org/wiki/index.php/HOWTO_Install_and_Use_the_CVMFS_Client>
    Mark has documented how to import the Oasis share (as well as the
    OSG's singularity share) onto a local Linux machine as the root
    user. This makes all of the essential GlueX software available on
    that machine as long as it is accessed from within our Singularity
    container.
 5. new version sets with bug fix to halld_recon
    <https://mailman.jlab.org/pipermail/halld-offline/2021-June/008562.html>
    This release was necessary to fix some bugs that crept in during
    refurbishing of the tagger energy scheme.


      Review of Minutes from the Last Software Meeting

We went over the minutes from the on May 11th 
<https://halldweb.jlab.org/wiki/index.php/GlueX_Software_Meeting,_May_11,_2021#Minutes>. 


  * Mark passed the list of high-priority files to be recovered from bad
    tapes, generated by Alex, on to Bryan Hess.
  * Beni and Simon pointed out that the change away from TLorentzVector
    et al. would be non-trivial. The new classes are not backward
    compatible and occurrences of the old classes are ubiquitous in our
    code.


      Minutes from the Last HDGeant4 Meeting

We went over the minutes from the meeting on June 15th 
<https://halldweb.jlab.org/wiki/index.php/HDGeant4_Meeting,_June_15,_2021#Minutes> 
without significant comment.


      Report from the June 3rd SciComp Meeting

Mark gave the report. See his slides 
<https://halldweb.jlab.org/doc-public/DocDB/ShowDocument?docid=5177> for 
the details. We discussed the problem with small files (less than one 
megabyte) on the cache disk. Mark remarked that there is no global 
solution to the problem; files come from many different sources for many 
different reasons. More thought is needed here.


      Review of recent issues and pull requests

We went over only one issue: Apparent angle in photon beam direction 
#531 <https://github.com/JeffersonLab/halld_recon/issues/531>, submitted 
by Alex two weeks ago. Alex summarized the discussion for us. Please see 
the issue itself for the details. The plot which opens the issues shows 
a non-zero x-component to the reconstructed, kinematically fitted total 
momentum for ρ events. There was a lot of discussion, back and forth, 
statement and retraction, etc. The progress thus far is documented in 
the issue itself.


      Other Issues

  * Richard re-iterated his request that we move forward in the Geant4
    version line, which will require a more advanced compiler that that
    provided by CentOS 7.
  * Beni noted that the collaborators who maintain ROOTSpy and Hydra,
    namely David Lawrence and Thomas, have both left the Hall D group.
    Thomas assured us that that Hydra remains one of his key job
    responsibilities.


      Action Item Review

 1. Do a build against ROOT 6.24. (Mark I.)
 2. Decide on needed tape volume sets. (Sean, Mark I.)
 3. Review disk usage when re-repatriating recon launch data. (Alex, Mark)
 4. Review outstanding issues with MCwrapper (Thomas)
 5. Take another look at halld_recon issue #512. (Sean)
 6. Think about Jon's list of improvements. (all)
 7. Automatically make CCDB entries for selected RCDB items. (Mark I.)
 8. Modify the documentation on how to copy files from JLab. (Mark I.)


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/halld-offline/attachments/20210622/8358e722/attachment.html>