phisym

dataset

/MinBias/Summer09-STARTUP3X_V8K_900GeV-v1/GEN-SIM-RECO

gives weird et spectra: it is bugged !

Launch on

/MinBias/Summer09-STARTUP3X_V8P_900GeV-v1/GEN-SIM-RECO

crab_0_100217_163431 : miscalibrated by SM

crab_0_100217_163754 : startup (ideal?) miscalib


18/02

sample :
datasetpath=/MinBias/Summer09-V16D_900GeV-v1/GEN-SIM-RECO
crab_0_100218_100729 : no miscalib
crab_0_100218_101000: miscalib
ABORTED !

19/02 relaunch

datasetpath=/MinBias/Summer09-V16D_900GeV-v1/GEN-SIM-RECO

crab_0_100219_163133 miscalibrated
crab_0_100219_163327 not miscalibrated

22/02

launch with less miscalibration on

/MinBias/Summer09-STARTUP3X_V8P_900GeV-v1/GEN-SIM-RECO

sm scale miscalibrated by 0.03

crab_0_100222_160913

25/02

launch with 0.03 sm scale miscalibration on data:

/MinimumBias/BeamCommissioning09-BSCNOBEAMHALO-Dec19thSkim_336p3_v1/RAW-RECO

crab_0_100225_14522
Problem with conf, revert to previous code


1/03

rerun as above
crab_0_100301_222437

old crab dirs moved to /data/argiro/crab-archive

2/03

miscal 3% on data

crab_0_100301_222437 crashed too... try again
crab_0_100303_114710 crashed too... try again
try again in
crab_0_100303_170238


in meanwhile run
0.015 SM miscale on real data
crab_0_100303_132521

shit, wrong miscal file name, relaunch

crab -cfg crab-smscale-realdata-ms0.015.cfg -create

new dir crab_0_100303_155654

4/3
run on full stat of

/MinBias/Summer09-STARTUP3X_V8P_900GeV-v1/GEN-SIM-RECO

crab_0_100304_161618 -> wrong config used.

run on 1MEvt of

/MinBias/Summer09-STARTUP3X_V8P_900GeV-v1/GEN-SIM-RECO

no Sm miscalibration

crab_0_100309_134225/ seems stuck
also in
crab_0_100309_153657




simulation full stat with no thresholds (same dataset as above)
crab_0_100308_222456

simulation 500kevts with no thresholds
crab_0_100308_225733

Phi sym miscalibration and crab

in
scratch0/phisym-simsmscale/CMSSW_3_3_6/src/Calibration/EcalCalibAlgos/scripts

miscalibandphisym.py.tmpl

phisym-submit template to miscalibrate, requires head of CalibCalorimetry/CaloMiscalibTools/src/CaloMiscalibTools.cc

miscalibandphisym.crab.py

same but for crab

crab.cfg

to run phisymmetry with crab and above pset.

source /afs/cern.ch/cms/LCG/LCG-2/UI/cms_ui_env.csh
cmsenv
source /afs/cern.ch/cms/ccs/wm/scripts/Crab/crab.csh

crab -create
crab -submit


To run normal calibration workfow with crab, modify crab.cfg to use
phisym-step1.crab.py

update 17/02 : bug found !
miscalibandphisym.crab.py did not miscalibrate !

directory

crab_0_100212_104838

has ideal calibration.

directory
crab_0_100217_105851

has 10% miscalib by SM

Phi Sym loop

Under test in

scratch0/w09/calibrate/CMSSW_3_2_5/src/Calibration/EcalCalibAlgos/scripts

Command

./phisym-analysis.sh phisym-loop-test.py.tmpl w09-alcareco.flst pccmsto03 /data/argiro/developement/looptest 3 cmscaf1nh

Work on package integration

on lxbuild151

still have to sort out tags in the queue

Kerberos @Cern

Problem: login worked, but logging in from lxplus, with a valid krb tokens, did not, and the server kept asking for a password.

Solution:

install

ncm-afscl
ncm-krb4clt

1. in /etc/sysconfig/network
have the fully qualified name (e.g. pccmsto03.cern.ch)

2. in /etc/hosts
have both qualified and unqualified, qualified first, i.e.

137.138.99.57 pccmsto03.cern.ch pccmsto03

work

1. phi sym plots on first collisions /data/argiro/collisions/CMSSW_3_3_4/src/DQMOffline/CalibCalo/
2. try to setup iteration script scratch0/w09/calibrate
   

EcalCondDB.py

latest news: with 3_3_X there is no need to download additional packages.
For example, in

CMSSW_3_3_X_2009-09-28-0100

and above

do

$CMSSW_RELEASE_BASE/src/CondTools/Ecal/python/EcalCondDB.py --listtags

Older things:

with

cmsrel CMSSW_3_1_0_pre7
addpkg CondCore/EcalPlugins sa6
addpkg CondTools/Ecal sa5
addpkg CondCore/IOVService zx29052009
addpkg CondCore/Utilities zx29052009
addpkg CondFormats/Common zx29052009


we have

1. histograms and comparison for the endcap as well
2. no need to specify till, only since

Example

./EcalCondDB.py --compare=comp.png -t EcalIntercalibConstants_mc --since=1 -t EcalIntercalibConstantsMC_EBg50_EEwithB --since=1


Note : the tag
addpkg CondCore/EcalPlugins sa6
fixes the bug by which the same constants were shown on the plus and minus side in EB