GlueX Offline Meeting, May 4, 2018

From GlueXWiki
Jump to: navigation, search

GlueX Offline Software Meeting
Friday, May 4, 2018
10:00 am EST
JLab: CEBAF Center A110
BlueJeans: 968 592 007

Agenda

  1. Announcements
    1. CalibBrowser 2.0 alpha CalibBrowser
    2. Software Review, Summer 2018
  2. Review of minutes from the April 6 meeting (all)
  3. Monitoring Dashboard Prototype
  4. Containers
    1. Report from the GlueX Containers meeting on April 27
    2. HOWTO use the GlueX Singularity Container - this is already on the HOWTO page
  5. Review of recent pull requests (all)
  6. Review of recent discussion on the GlueX Software Help List (all)
  7. Action Item Review (all)

Communication Information

Remote Connection

Slides

Talks can be deposited in the directory /group/halld/www/halldweb/html/talks/2018 on the JLab CUE. This directory is accessible from the web at https://halldweb.jlab.org/talks/2018/ .

Minutes

1.1

  • Justin: who owns the "initiate scan" button.
  • Thomas: no one
  • Justin: Any problems with this?
  • Thomas: I don't think so, but maybe. Future development to secure the button

1.2

No one has heard anything more about the Software Review this summer

3.

  • The specific version of firefox did not show the dashboard, for unknown reasons. Sean shared it.
  • Comments? "Great" "fancy"...
  • Next steps include expansion to other tests including single track and reconstruction tests.
  • Thomas would like to roll out the new technology to the plot browser over the summer

At this point Sandy poked her head in, offering to be beaten up for the farm performance. Essentially, the problem falls to the volume of Hall writes and a lack of LTO8 media. She reports the center was starved for tapes and validation of new media took time. The crunch should be over "soon". Eugene inquired about performance of LTO8 vs LTO6; answer: about 2x throughput with LTO8. The priorities were revealed to be: Hall writes, other writes, reads. This caused the bottleneck for jobs. Sandy also mentioned Tuesday downtime corresponding to other shutdowns planned. If all goes well Wednesday afternoon....

4.

The xkcd comic was shown and Sean gave a brief overview of containers at GlueX. Sean also mentioned he rand the containers at FSU. Beni asked for a how-to and volunteered to be the guinea pig to test the documentation. Justin asked if the containers are run on the farm. Thomas noted there were no direct plans/testing. Sean thought it would be a good test. In principle it should not pose any significant issues. Thomas also noted problems with the OSG. I turns out the host certificate had expired. This has been solved.

Alex A. noted that the reconstructed rhos in the reconstruction tests are down a bit again. Simon made some improvements which appear to not have a strong effect on rho's, also a larger spread was noted. Beni asked about how ver01 of offline monitoring compared to the other versions. Answer: different software and calibrations.

While discussing the google questions the topic of changing run comments in the rcdb came up. Sean made a program to make Dmitry's solution more accessible. General agreement more protocols are needed for changing comments but the tech is there.

The question of a new monitoring launch came up. Justin thought it would be a good idea and Alex A agreed to launch special runs over the weekend.

Beni thought that there might need to be some deadline for calibrations (2018). Sean proposed some rolling deadlines as some calibrations have no procedures. Thomas: automated? Sean: No, none. Sean mentioned the FCAL 4ns shifts as one such calibration. A lively debate ensued between Sean and Beni about how to check the 4ns shifts. In the end, Matt tried things, which did not work and everyone was a little surprised.