Repository navigation
Conversation
|
REQUEST FOR PRODUCTION RELEASES: This will add The following labels are available |
… o2dpg_sim_workflow.py
| configFilePath='json://${O2DPG_ROOT}/MC/config/QC/json/mft-digits-' + str(flp) + '.json', | ||
| objectsFile='mftDigitsQC.root') | ||
| needs=[getDigiTaskName("MFT")], | ||
| readerCommand='o2-itsmft-digit-reader-workflow --runmft --disable-mc --suppress-triggers-output --mft-digit-infile mftdigits.root' + staggMFT, |
There was a problem hiding this comment.
why do we need suppress-triggers-output?
There was a problem hiding this comment.
in https://github.com/AliceO2Group/AliceO2/blob/c84b32c0a8771584af70cb91a4dbb73265d68bc5/Detectors/ITSMFT/common/workflow/src/DigitReaderSpec.cxx#L294 we define a trigger output without this flag for which a dummy value is sent at the moment. But the QC task only subscribes to the DIGITS and DIGITSROF outputs, so the dummy trigger output is sent to the sink, i.e. it is not needed.
This fixes https://its.cern.ch/jira/browse/O2-7331 observed in recent MC productions with empty TFs.
Two remarks:
--suppress-triggers-outputoption suppresses sending the trigger output, but currently, the output is still defined unconditionally in the workflow spec. This will produce a harmless error message for each TF ([ERROR] Expected Lifetime::Timeframe data MFT/PHYSTRIG/0 was not created for timeslice 0 and might result in dropped timeframes) for a declared output which was not sent. This will be fixed with a follow up PR (declare trigger output in ITS/MFT digit reader only when running with triggers AliceO2#16145) for the digit reader in O2.