Submission checklist
gmestdac edited this page Jan 12, 2022
·
5 revisions
These were written with UL in mind, might not all apply for pre-UL
- in 2016 3 triggers should be added, unless someone objects
-
- 2016 Single Electron trigger set: add HLT_Ele27_eta2p1_WPLoose_Gsf
-
- 2016 Singe Muon trigger set: add HLT_IsoMu22_eta2p1 and HLT_IsoTkMu22_eta2p1
- prefiring corrections are now available for both electrons and muons, and are relevant for all 3 years (2018 no longer unaffected) https://twiki.cern.ch/twiki/bin/view/CMS/L1PrefiringWeightRecipe
- there are now 46 psWeights available (index 0 through 45) (currently storing 14). The ordering has changed so that we need indices 4,5, 26, and 27 --> we're currently not storing the last 2. In principle the order depends on the pythia8 version, so we need to be sure it's consistent for all UL samples.
- are the GlobalTags updated? Check on https://twiki.cern.ch/twiki/bin/view/CMS/PdmVRun2LegacyAnalysis
- storeAllTauID should be set for anything Liam uses
Confirmed. It should be used in general if you want access to all deepTau information instead of the set standard working point. - storeLheParticles should be set for TTGamma samples only (so that events with photon _lhePt < 10 can be removed)
- storeJecSources should be set as long as there as precision measurements going on in the group, can always be needed
- the sample list name should contain the year and ideally some name that makes the batch of samples recognizable
- even if all samples seem to have been submitted successfully some might not have been submitted, one way to check:
run
for folder in */*/*; do crab status "$folder" >> crabStatusLog.txt ; done
in your crab folder, then check this log the folders for failing jobs in the crab directory should be deleted before resubmitting - is the skim you're using loose enough for everyone?
- recommendations for what exact name to use for primary datasets / eras has changed in the past. Check if this happened again
- when processing the TTW EWK samples Kirill privately produced a change needs to be made in multilep.py:
lheEventInfo = cms.InputTag("externalLHEProducer") --> lheEventInfo = cms.InputTag("source")
Otherwise psweights and other lhe info will not be available. Don't forget to put it back
A reasonably complete list of required samples for UL can be obtained by taking everything Liam submitted for UL:
/pnfs/iihe/cms/store/user/lwezenbe/heavyNeutrino/WZG_TuneCP5_13TeV-amcatnlo-pythia8
(check for samples with UL in the name)
and adding everything Gianny submitted (these were submitted exactly because they were missing / incomplete on Liam's side) :
/pnfs/iihe/cms/store/user/gmestdac/heavyNeutrinoUL
(already only contains UL samples)