[Halld-offline] Software Meeting Minutes, March 16, 2022

Mark Ito marki at jlab.org
Wed Mar 16 18:57:46 EDT 2022


Please find the minutes here 
and below.

   -- Mark


    Minutes, GlueX Software Meeting, March 16, 2022

Present: Alex Austregesilo, Nathan Brei, Thomas Britton, Sean Dobbs, 
Mark Ito (chair), Igal Jaegle, Naomi Jarvis, Richard Jones, Curtis 
Meyer, Susan Schadmand, Simon Taylor, Beni Zihlmann

There is a recording of this meeting 
on the ZoomGov site. (Passcode: JTHgxw%5)


 1. New version set: 5.5.0
    This version set has updates for the new detectors for the CPP
      * Richard asked about his stand-alone HDDM repository. Mark
        replied that that has not been incorporated into the default
        version set yet. One outstanding step is to remove code from
        halld_recon that is provided by the HDDM repo. Richard reminded
        us that the data models themselves are stored in halld_recon's
        HDDM library and should remain so.
 2. New version of Build Scripts: version 2.33
    This version allows set-up of a subset of packages. Packages omitted
    from the XML file will not have their set-up steps performed.

      Review of Minutes from the Last Software Meeting

We went over the minutes from March 2nd 

  * The pull requests from Simon have been merged and incorporated in
    version set 5.5.0.
  * Mark still needs to arrange a meeting for OSG-interested parties.
  * Mark has done some tests showing that a build on the ifarm a
    corresponding build in the CentOS 7 Singularity container are
    equivalent. He is working replacing the two separate builds with one.
  * Richard suggested making JANA2 a standing agenda item for these
  * Richard mentioned that GlueX has been mentioned more than once in
    the opening talks of the OSG All-Hands meeting being held this week.
  * Richard reported that at the All-Hands meeting Miron Livny announced
    funding for an initiative to have HPC Centers accept jobs from OSG
    Submit Hosts. The Pittsburgh Supercomputing Center was mentioned
    specifically as a target for jobs.

      Review of Minutes from the last HDGeant4 meeting

We went over the minutes from the meeting on March 9th 

  * Mark reported, and Thomas confirmed, that the new container
    gluex_prod:v1 is in production us on the OSG via MCwrapper. There is
    also a gluex_devel:latest out on the OSG's Singularity CVMFS share.
  * Mark encouraged us to think about how and when to move to a more
    advanced version of the GCC compiler, one beyond that provided
    natively on CentOS 7 (GCC 4.8.5).
  * Richard advocated putting a CentOS 8 Stream container on CVMFS as
    well as a corresponding build on Oasis. Mark will look into this.
  * Mark highlighted the decision that this would be the last of the
    bi-weekly-scheduled HDGeant4 meetings.

      FAQ of the Fortnight: What is this sbatch error that I am getting?

We went over the FAQ 
first raised by Edmundo Barriga. To run on the JLab farm, you not only 
have to be a member of the halld group, but you also have to be listed 
under the halld Slurm account.


Nathan summarized the status of JANA2 and the conversion of halld_recon 
to use it. See his slides 
<https://halldweb.jlab.org/talks/2022/introducing_the_port.pptx> for the 
details. From the discussion:

  * Alex asked whether Nathan was interested in consulting on GlueX
    software generally now that he has gained familiarity with
    halld_recon. Nathan did not dismiss the idea out of hand.
  * There is an existing branch of halld_recon that has been converted
    to use JANA2. We need to identify a recent version to standardize on
    for comparison.
  * Alex mentioned that Richard has done a lot of work on
    reproducibility of our reconstruction code. We might want to make
    sure that all of those changes are in before flagging a version for
    comparison testing.
  * Mark asked about compiler versions. Nathan responded that JANA2
    compiles with GCC 4.8.5, but that he would welcome the ability to
    access features for more advanced versions in the future.
  * We will invite Nathan back to future meetings to report on further

      Review of recent discussion on the GlueX Software Help List

We reviewed the thread on the Help List on deleting old branches from 
halld_recon <https://groups.google.com/g/gluex-software/c/kbMAF3Uzi2c>. 
Mark has not received any objections.

  * Sean cautioned us that we would not want to inadvertently delete
    Richard's work on reproducibility.
  * Nathan proposed that one way to preserve the old branches would be
    to fork halld_recon, and port all of the current branches of the
    original repo to the fork before deleting them from the original.

      Action Items

 1. Increase Shibboleth time-out beyond its current 12 hour setting. (Bryan)
 2. Call a meeting to discuss OSG issues for the JLab community. (Mark)
 3. Put a CentOS 8 Stream system out on CVMFS (container and software
    stack). (Mark)

Retrieved from 

  * This page was last modified on 16 March 2022, at 18:34.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/halld-offline/attachments/20220316/a96ace88/attachment.html>

More information about the Halld-offline mailing list