Re: [Brahms-dev-l] Request: pp ltr for run02

To anybody who wants to analyze ltr/gtr/dst in the official way, here is 
the steps.  (By the way, there was no pp in run01)  (NOTE:  Please 
change production version.  See the stop no 10. )

For ltr.
1.  logon to a rcas machine as bramreco
2.  change directory to DataProduction directory by "cd DataProduction"
3.  create an appropriate directory.  For example, for pp runs in RUN 02 
at sqrt(s_NN) =200 GeV, you would do "mkdir -p run02/pp/200/ltr"
 (NOTE:  make sure to follow the format.  The first one is run period.  
The second one is specie.  And, the third one is energy.  And the last 
one is type of data that you want to create.
4.  change to that directory
5.  copy the updateScript.sh from other production runs (for example, 
from run03/pp/200/ltr/updateScript.sh)
6.  Copy/create the bratmain script for your (ltr) job.  (NOTE:  Your 
old script may/may not work.)
7.  Change the content of the script (updateScript.sh) accordingly.  Pay 
particular attention to "set" statement.  For example, your homedir 
should be "/brahms/u/bramreco/DataProduction/run02/pp/200/ltr" in the 
above example.  "configscript" should be what you just made in step 6.  
The "datadir" is output directory.  You can find the use of our data 
disk by DataFileUsage.perl.  "basedir" should be 
"/brahms/u/bramreco/DataProduction/scripts/ltr".
8.  run updateScript.sh   This will copy all necessary files from the 
repository and change the content accordingly to your updateScript.sh
9.  If you are remaking the production runs after substantial change in 
brat, it is highly adviserable (and sometimes required) to make a new 
production version. You can view and create production version  at 
http://brahms-web.brahms.bnl.gov/daq-cgi-bin/EnterNewProduction.pl
10.  The change the production version number in submitLtr.perl script 
according to new production version number.
11.  Now, you are ready to submit your job.  Log on the rcrsuser4 
machine instead of rcas.  Then, you submit jobs by "./submitLtr.perl -u 
yourusername firstRunNo lastRunNo" at the directory you just created.
12.  You can monitor the progress by crs_status and crs_panel

Hiro


Bjorn H Samset wrote:

>Dear BRAHMS analysts,
>
>I see that for the 2001 p+p data there are some runs/settings that have a
>very varying FS efficiency. I'd like to study this using Pawels eff.
>package, but it seems the local track files are not there anymore.
>According to Kris, they may have been deleted and not put into HPSS - in
>any case they are not in the file catalogue as far as I can see.
>
>So a request from me: Could someone who's used to doing local tracking the
>'proper' way re-run the ltr's for runs 6250-6757? It's about 500 sequences
>all in all, so no humongous job luckily.  (I can of course also specify
>just the runs that I need, but since it's not really a big set...) The ltr
>script that was used, located here:
>~bramreco/bhs/reco/run02/pp/200/ltr/ppLocalTracking.C
>may need some modification to fit the new file catalogue system, but
>nothing major I believe.
>
>Or, of course, if someone tells me that the files are in HPSS after all
>I'd be even more happy since then I can use the files that are the actual
>basis for the analysis...
>
>Hoping for some help, best regards
>--Bjorn
>
>
>
>
>--
>Bjorn H. Samset                           Phone: 22856465/92051998
>PhD student, heavy ion physics            Adr:   P.b. 1048 Blindern
>Inst. of Physics, University of Oslo             0316 Oslo
>                              \|/
>----------------------------> -*- <-----------------------------
>                              /|\
>_______________________________________________
>Brahms-dev-l mailing list
>Brahms-dev-l@lists.bnl.gov
>http://lists.bnl.gov/mailman/listinfo/brahms-dev-l
>  
>

_______________________________________________
Brahms-dev-l mailing list
Brahms-dev-l@lists.bnl.gov
http://lists.bnl.gov/mailman/listinfo/brahms-dev-l