Skip to content

HLTJobOptions.db.json remove filenames to avoid duplicates

Mark Stockton requested to merge mark/athena:HLTJOdb into master

Remove filenames to avoid duplicate DB uploads of HLTJobOptions.db.json, especially when produced for different prescalesets but the same menu ATR-20878

The affected JSON blocks are:

          "HLTConfigSvc": {
              "ConfigSource": "None",
              "InputType": "db",
              "JsonFileName": "None",
              "XMLMenuFile": "None"
          },

Previous value example: "JsonFileName": "HLTMenu_LS2_v1_22.0.37.json",

          "HLTPrescaleCondAlg": {
              "DetStore": "StoreGateSvc/DetectorStore",
              "EvtStore": "StoreGateSvc",
              "ExtraInputs": "[]",
              "ExtraOutputs": "[]",
              "Filename": "None",
              "HLTPrescales": "ConditionStore+HLTPrescales",
              "NeededResources": "[]",
              "PSKFolder": "ConditionStore+/TRIGGER/HLT/PrescaleKey",
              "Source": "COOL",
              "TriggerDB": "JOSVC"

Previous value example: "Filename": "HLTPrescalesSet_LS2_v1_22.0.37.json",

          "LVL1ConfigSvc": {
              "ConfigSource": "XML",
              "InputType": "db",
              "JsonFileName": "None",
              "JsonFileNameBGS": "None",
              "XMLMenuFile": "TriggerMenuMT/LVL1config_LS2_v1_22.0.37.xml"
          },

Previous value examples: "JsonFileName": "./L1Menu_LS2_v1_22.0.37.json", "JsonFileNameBGS": "./BunchGroupSet_LS2_v1_22.0.37.json",

Note XMLMenuFile is needed as is read from the release still currently

Tested running from the DB with release 2021-06-01T2101 athenaHLT.py --imf --perfmon --threads=1 --concurrent-events=1 --nprocs=1 --number-of-events=50 --file=/cvmfs/atlas-nightlies.cern.ch/repo/data/data-art/TrigP1Test/data18_13TeV.00360026.physics_EnhancedBias.merge.RAW._lb0151._SFO-1._0001.1 -b --smk=180 --l1psk=41 --hltpsk=109 --db-server=TRIGGERDBATN

@stelzer is there any implication from these changes for running in a partition?

@astruebi @tamartin maybe its worth making use of the ongoing TR to test this - I can provide a modified SMK if so.

Merge request reports