ssh -Y lxplus.cern.ch -o ServerAliveInterval=240
bash -l
export SCRAM_ARCH=slc7_amd64_gcc700
cmsrel CMSSW_10_6_4
cd CMSSW_10_6_4/src
cmsenv
The instructions to generate an SSH key can be found here. Once you have the SSH key it has to be associated with your GitHub account. If the key is not permanently added, the following should be done in every new login.
eval "$(ssh-agent -s)"
ssh-add ~/.ssh/id_rsa
ssh-add -l
cd $CMSSW_BASE/src/
git clone --branch 13TeV [email protected]:latinos/setup.git LatinosSetup
source LatinosSetup/SetupShapeOnly.sh
cd $CMSSW_BASE/src/LatinoAnalysis/Tools/python
cp userConfig_TEMPLATE.py userConfig.py
emacs -nw userConfig.py
Replace the following line
baseDir = '/afs/cern.ch/user/x/xjanssen/cms/HWW2015/'
with your own directory,
baseDir = '/afs/cern.ch/user/[YourInitial]/[YourUsername]/cms/HWW2015/'
And add the following line
batchType = 'condor'
cd $CMSSW_BASE/src/
git clone [email protected]:latinos/PlotsConfigurations.git
cd $CMSSW_BASE/src/
cmsenv
scram b -j 10
The complete set of instructions can be found here. In order to obtain a VOMS certificate you must first get a personal certificate (Steps 1 and 2 of the previous link), which consists of a private and a public keys (PEM files). You can follow the personal certificates TWiki to install the certificate in your browser. Then you will need to export the certificate into a pair of PEM files, following these how to instructions. Once you have your personal certificate you can proceed with Step 3. Go to the VOMS registration page and submit a registration. Then you need to wait for an admin to confirm your registration (you will receive a notification email). Here you can find the VOMS FAQ.
To verify the expiration date of your certificate,
openssl x509 -subject -dates -noout -in $HOME/.globus/usercert.pem
To check if you can generate proxies,
grid-proxy-init -debug -verify
To check if you are a member of the CMS VO,
voms-proxy-init -voms cms
voms-proxy-init -voms cms -rfc --valid 168:0
cmsenv
Run 2 analyses are split in 2016, 2017 and 2018. For each year there is a set of data and MC trees, together with different lepton and b-tag working points. Besides, some systematic uncertainties might change. Please go to each year to produce the corresponding histograms.
cd $CMSSW_BASE/src/PlotsConfigurations/Configurations/VH2j/Full2016_v6
cd $CMSSW_BASE/src/PlotsConfigurations/Configurations/VH2j/Full2017_v6
cd $CMSSW_BASE/src/PlotsConfigurations/Configurations/VH2j/Full2018_v6
mkShapesMulti.py \
--inputDir=/eos/cms/store/group/phys_higgs/cmshww/amassiro/HWWNano/ \
--doBatch=True \
--batchQueue=workday \
--batchSplit=Samples,Files
condor_q
If needed, you can check the (automatically produced) configuration files of all the submitted jobs, and even submit manually some of them for debugging or test purposes.
ls $HOME/cms/HWW2015/jobs
If you need to cancel all the submitted jobs,
condor_rm [YourUsername]
cd $HOME/cms/HWW2015/jobs/mkShapes__VH2j_2017
for i in *jid; do condor_submit ${i/jid/jds}; done
mkShapesMulti.py \
--inputDir=/eos/cms/store/group/phys_higgs/cmshww/amassiro/HWWNano/ \
--doHadd=True \
--batchSplit=Samples,Files \
--nThreads=8 \
--doNotCleanup
In the unexpected event that you get errors drawing the distributions, it could be due to a mismatch between plot.py
and samples.py
. To fix them, check and remove from plot.py
the proccesses that are not in samples.py
.
mkPlot.py --inputFile=rootFile/plots_VH2j_2016.root --minLogC=0.01 --minLogCratio=0.01 --maxLogC=1000 --maxLogCratio=1000 --showIntegralLegend=1
mkPlot.py --inputFile=rootFile/plots_VH2j_2017.root --minLogC=0.01 --minLogCratio=0.01 --maxLogC=1000 --maxLogCratio=1000 --showIntegralLegend=1
mkPlot.py --inputFile=rootFile/plots_VH2j_2018.root --minLogC=0.01 --minLogCratio=0.01 --maxLogC=1000 --maxLogCratio=1000 --showIntegralLegend=1
To produce blinded distributions (without data) in the signal region, open plot.py
and set the variable isBlind
to 1
for DATA
.
mkPlot.py --onlyCut=VH_2j_emu --inputFile=rootFile/plots_VH2j_2016.root --minLogC=0.01 --minLogCratio=0.01 --maxLogC=1000 --maxLogCratio=1000 --showIntegralLegend=1
mkPlot.py --onlyCut=VH_2j_emu --inputFile=rootFile/plots_VH2j_2017.root --minLogC=0.01 --minLogCratio=0.01 --maxLogC=1000 --maxLogCratio=1000 --showIntegralLegend=1
mkPlot.py --onlyCut=VH_2j_emu --inputFile=rootFile/plots_VH2j_2018.root --minLogC=0.01 --minLogCratio=0.01 --maxLogC=1000 --maxLogCratio=1000 --showIntegralLegend=1
Once the blinded distributions have been produced, don't forget to set back to 0
the variable isBlind
for DATA
.
Follow these instructions. Once the website has been created you can manage it here. In order to allow access to your website, add an .htaccess
file to your www directory.
wget https://raw.githubusercontent.com/piedraj/AnalysisCMS/master/test/.htaccess
Check that it contains Options+Indexes
. To improve the usability of the website add an index.php
file to every directory and subdirectory.
wget https://raw.githubusercontent.com/piedraj/AnalysisCMS/master/index.php
If you have followed these steps and the website is unavailable (error 403) please open a ticket with the CERN Service Portal.
cp -r plotVH2j_2017 /afs/cern.ch/user/p/piedra/www/.
cd /afs/cern.ch/user/p/piedra/www/plotVH2j_2017
cp ../index.php .
Go to the web.
https://piedra.web.cern.ch/piedra/