P14 Reprocessing: GridKA Operation Skripts

by Thomas Nunnemann

All files are available in TN.tar.

1. Prestaging of files:
Command: ./get_datasets.sh
Input files: datasets.dat (variable, list of datasets, one per line)
getroot.py (generic)
2. Job submission:
Command: ./submit_datasets.sh (calls do_it2)
Input files: datasets.dat (variable, list of datasets, one per line)
do_it2 (generic)
reproc_XX.macro (generic, only site specific mods)
3. Dataset definition for failed jobs:
Command: ./create_failed_datasets.sh (calls dataset_failed.sh)
Input files: datasets.submitted (variable, list of datasets, and (partially processed) datasets)
dataset_failed.sh (generic) 3 arguments:
original dataset, name suffix for new sets, path to output files
Output files: datasets_failed.dat (list the created datasets for the failed jobs)

These scripts assume a D0 release installation and you will have to modify hard coded paths. Let me know if you would like to adopt those scripts, but you run into problems.


Daniel Wicke, 3-Feb-2004. Last update 3-Feb-2004.
Daniel.Wicke@physik.uni-wuppertal.de

Valid HTML 3.2!