Production Release p09

Status: I would not use p09 for trigger simulations. It was basically abandoned in favour of p10 when it was clear that it was unlikely to be as stable as it needed to be before p10 stabilized. So, please stop looking at this page right now. It is based on t01.46.00. As of p09.08.00 things are starting to come together. Several thousand events have been run without crashes. However, a floating point exception is still seen in l3fcalcluster on certain files. The fix is known and should be in p09.09.00.

Known Problems/Fixes

- You must check out the p09-br version of tsim_l3 in order to
run. There is a minor bug in the initScriptRunner.sh script which will 
cause you to crash right away without this patch. Since it is only a
script which changes there is no need to recompile anything.

Default Trigger List - same as p08.12.00 so far

p09 Cookbook (new d0tools instructions coming soon!!!!!)

setup n32                      -|
setup D0RunII p09.xx.xx         |
setup d0cvs                     | generic stuff 
newrel -t p09.xx.xx mydir       |
cd mydir                        |
d0setwa                         |
addpkg d0trigsim               -|


cd d0trigsim/bin               -|
ln -s mydata inputfile          | running on disk file
Runme.sh                       -|

cd d0trigsim/bin               -|
edit Runme_sam.sh project name  | running on SAM data
Runme_sam.sh                   -|

Running with RECO NEW

Laurent Duflot is a good man. He has provided us with a script which creates an executable combining d0trigsim and reco_analyze to produce a common ntuple. The script is here. A script to run this executable is also available here. An example syntax to run it on a particular dataset in SAM is

./send --group=dzero --cpu-per-event=30 --defname=glsqpairstanb2-njmet_p05_uta_poisson-1.0

The arguments are those of the sam_submit script so you can choose batch or interactive running. So far it works only on d0mino. You can find some example root output in

/prj_root/843/trigsim_1/ntuples/trigsim+reco

This is not the longterm solution to combining these ntuple blocks. Eventually this will be done in d0_analyze....but not in p09.

Description of Ntuple variables - nothing yet