Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

ItemWhoPre-Meeting NotesNotes and  Action Items
Update status, task review

Bug and issue reviewAll Review of any issues that have come up during the week with faro and metric computation 
  • Several issues around documentation failures when code was removed that documentation referred to.  Is there something that can tell us if the docs are broken? 
    • Run package docs (and scons) locally before commit 
    • KT's jenkins job will help when it is running
  • Pipelines don't see this, why not? Have we missed something in our setup? No, this is probably because pipelines don't remove stuff very often. Adding code with no documentation will not cause a fail but removing code and not updating the documentation will. 
Status of reprocessing 

Problems  running  faro on the w_2021_14 (RC2)  with external calibrations. : 

Jira
serverJIRA
columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
serverId9da94fb6-5771-303d-a785-1b6c5ab0f2d2
keyDM-29914

  • Last weekly that faro was run on with full calibrations was w_2021_06. There was no other RC2 reprocessing in between. Not worth going back to this version to test - too old.  
  • Pipelines should produce external calibrations, this is not a metric/faro task
  • An afternoon of pari coding with Eli to might sort this out?
  • Keith also concerned about applying calibrations in commissioning rapidly
  • CMR for R22. Need to work this out if we want to apply the full calibrations for the CMR. Using SF calibrations will not yeild results as good as previous results, and this will look bad. 

More general question of "how should we have known in advance that these changes to FGCM were made.  What should we monitor to anticipate this?"

  • RHL  - this is a format/data model change not an algorithm change. Should have been RFC'd. This is a format / interface change.  
  •  Robert Luptontalk to Yusra and pipelines and get an answer to this question.  
  • Colin ask why did this not break CI? Is our test coverage too low.
    • KSK: Our test dataset does not have global calibrations, but these would have broken anyway as they would have been in the old format. Updating our  faro a pipeline we run to compute global calibrations would fix this
  •  Leanne Guy Create a story to extend a faro pipeline to compute global   calibrations  

Conclusions and next steps: 

  • Jeffrey Carlin  to proceed with faro on w_2021_14 using SF calibrations 
  • Keith Bechtol and Jeffrey Carlin  to test Keith's solution for a single detector  at the patch / tract level. If this works .... 
  • Run the CNR for R22 using this fix. 
  • If not - rediscuss plan
Development status 

Review progress on the current development epic

Jira
serverJIRA
columnskey,summary,type,created,updated,due,assignee,reporter,priority,status,resolution
serverId9da94fb6-5771-303d-a785-1b6c5ab0f2d2
keyDM-26993

No time 
AOB
Next meeting 2021-05-04

...