[Halld-offline] Software Meeting Minutes, December 20, 2021
marki at jlab.org
Mon Dec 20 13:50:40 EST 2021
Please find the minutes here
GlueX Software Meeting, December 20, 2021, Minutes
Present: Alex Austregesilo, Edmundo Barriga, Mark Ito (chair), Igal
Jaegle, Naomi Jarvis
There is a recording of this meeting
<https://bluejeans.com/s/qxTLU3FPpVx>. Log into the BlueJeans site
<https://jlab.bluejeans.com> first to gain access (use your JLab
1. New version set: 5.2.0
New production release from December 8.
2. new recon-launch-compatible version sets
These correspond to version set 5.2.0.
3. group disk cleaned up, old releases deleted
500 GB deleted of 3.0 TB total.
Review of Minutes from the Last Software Meeting
We went over the minutes from the meeting on December 6
* Alex has loaded all of the 2017-01 REST files onto the cache disk.
Naomi is copying them to CMU and Richard Jones is copying them to UConn.
* Igal has uploaded PrimEx calibration data, but is having problems
accessing them. He will post a description of the problem to the
Software Help List.
Beni Zihlmann called a meeting of the Documentation Task Force on
December 9. Naomi reported that a wiki page
<https://halldweb.jlab.org/wiki/index.php/Software> has been created to
give an overview of GlueX software at a beginner's level. It is linked
from the main wiki page
<https://halldweb.jlab.org/wiki/index.php/Main_Page> under "Overview" in
the "Software" section of the column on the right. Writing assignments
for subsidiary pages were made and many have been fulfilled.
FAQ of the Fortnight: What do I need to specify to get the correct
constants from the CCDB?
Mark went over the details of what is needed to fully specify
the set of calibration constants to be used.
Last Week's Computing Review
Mark reported that the Review went well for us. He thanked all those
that contributed. See his email from last Friday
Review of recent issues and pull requests
We went over gluex_root_analysis Issue #156
Crashes with minimal DSelector upon writing output trees, probably
Naomi told us that Paul Mattione had reported that when cloning ROOT
forgets branch addresses when it uses PROOF or when it operates on a
TChain[?] and so he had to write his own routine, which does not seem to
work on CentOS 8. To debug the problem further she needs to go the main
branch of ROOT to make sure that all recent changes have been
incorporated. Mark said that he would look into adding that capability
to the ROOT makefile in Build Scripts.
For now Naomi is using a two-part workaround:
1. process one file at a time
2. do not use PROOF
Meeting Time in January
After the Holidays, we will go back to meeting on Tuesdays at 3:00 pm
for both the Software and HDGeant4 meetings.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Halld-offline