for cms analysis
Test file : catTuple.root can be found in https://www.dropbox.com/s/40tvwebdv6g0x1m/catTuple.root?dl=0
scram p -n cat CMSSW CMSSW_7_2_2_patch1
cd cat/src
cmsenv
git clone git@github.com:vallot/CATTools.git
cd CATTools
git checkout cat72x
cd ..
scram b -j 8
cmsRun $SRT_CMSSW_BASE_SCRAMRTDEL/src/CATTools/CatProducer/prod/runCat.py useMiniAOD=True inputFiles=file:/pnfs/user/jlee/DYJetsToLL_M-50_13TeV-madgraph-pythia8/miniaod.root
or
cmsRun $SRT_CMSSW_BASE_SCRAMRTDEL/src/CATTools/CatProducer/prod/runCat.py useMiniAOD=False inputFiles=/store/mc/Phys14DR/DYJetsToLL_M-50_13TeV-madgraph-pythia8/AODSIM/PU20bx25_PHYS14_25_V1-v1/00000/00CC714A-F86B-E411-B99A-0025904B5FB8.root
- no need to change configurations since jobs are stored all in remote servers
- make sure crab3 is setup first
source /cvmfs/cms.cern.ch/crab3/crab.sh- This script can be used to pass CRAB configuration parameters from the command line depending on the Dataset lists are located on MC/ or RD/ directories for data type.
./submitCrab3.py [Dataset1.txt] [Dataset2.txt]- If we want to submit cattuple ttbar and diboson datasets.
cd $SRT_CMSSW_BASE_SCRAMRTDEL/src/CATTools/CatProducer/prod/crab
./submitCrab3.py MC/ttbar_dilepton.txt MC/diboson.txt- because the jobs only saved in local storage (where the job was done) we need to get the output manually.
- this has the benefit of crab jobs not failing due to transfer errors
- to get the output use "crab out -t " or the script below
python getcrabOut.py