DFOS installation: 4. Finish the installation

[not maintained anymore]

11. Install createJob and processAB; find your favorite data reduction system

• install createJob, processAB
• install ngasClient for the data and header downloads, hotfly for header updates
• createJob writes the AB processing tasks into $DFO_JOB_DIR/execAB_<mode><date> which are ready to execute
• a job file is filled for off-line processing ('JOBS_NIGHT')
• the supported data reduction systems are: CPL (calling esorex), CON (parallel processing with esorex).
  esorex is the recipe interface and comes as part of the standard DFS installation; CONDOR is a batch queue system and is also DFS-installed.
• installation of your pipeline needs to be done by your pipeline developer or by DFS.
• once your pipeline runs and is properly controlled by processAB, install getStatusAB to have a GUI as AB monitor

12. Install processQC, certifyProducts, scoreQC

• processQC is just a wrapper calling QC scripts. The QC scripts are not part of dfos and need to be written by the QC scientist.
• the same is true for certifyProducts: the certification script is user-provided
• for both tools an example MIDAS procedure is part of the package.
• the calls of processQC are written by createJob into $DFO_JOB_DIR/execQC_<mode><date>
• install scoreQC ; it is called by the QC procedures; its configuration is based on experience and specifications and will require fine-tuning
• if you have no pipeline support: skip these workflow steps

13. moveProducts, ingestProducts

• moveProducts is part of the POST workflow part. It is used in any case, even if no pipeline support exists
• install renameProducts, updateAB which are all called by moveProducts
• install ingestProducts which is a wrapper around the DFS tool cdbIngest to insert product files into NGAS

14. Install updateDP, finishNight, extractStat, histoMonitor

• updateDP: ABs are read for packing information
• this tool has a graphical monitor
• this tool is needed even if no pipeline support exists
• finishNight: cleans up (almost) everything; extractStat and histoMonitor provide storing relevant information

Now the suite of DFOS tools for the daily workflow is complete.