Deploy and configure staging GeoBatch

GeoBatch only needs to be installed in the staging area. No batch processing is going to be run in the dissemination system.

GeoBatch configuration files

Geobatch uses some directories for its processing needs.

We’ll create all of them under /var/stg_geobatch/:

  • config: GeoBatch main configuration directory
  • input: Root input directory
    • input/ingest: Ingestion flow input directory
    • input/reprocess: Reprocess flow input directory
    • input/publish: Publish flow input directory
  • temp: Root temp directory
    • temp/ingestionFlow: Ingestion flow temp dir, used during flow execution
    • temp/reprocessFlow: Reprocess flow temp dir, used during flow execution
    • temp/publishFlow: Publish flow temp dir, used during flow execution
  • orig: The place where original data from ingested flows will be copied

The config directory is the most important. It contains the flow definitions and the files needed to generate charts from statistics.

Copy the content of WEB-INF/data (included inside of geobatch.war) to /var/stg_geobatch/config/. This is a sample configuration. Edit the ingestionFlow.xml, reprocessFlow.xml, publishFlow.xml, and the groovy scripts under chartscripts to adjust the different paths, users and passwords as needed. You may also change the flow definitions to your needs.

Note

See GeoBatch chapter for details on how to configure the flows.

Remember to change the file permisions as follows:

cd /var/stg_geobatch/
chown -R tomcat6:tomcat6 .

Groovy scripts need to be exeutable:

chmod ug+x config/chartscripts/*.groovy

Environment settings

This is the bin/setenv.sh file:

SERVICE=stg_geobatch

# Application specific environment
GEOBATCH_CONFIG_DIR=/var/$SERVICE/config
GEOBATCH_TEMP_DIR=/var/$SERVICE/temp
UNREDD_OVR=$GEOBATCH_CONFIG_DIR/unredd-ovr.properties

# Java options
JAVA_OPTS="-server -Xms2048m -Xmx2048m -DGEOBATCH_CONFIG_DIR=$GEOBATCH_CONFIG_DIR -DGEOBATCH_TEMP_DIR=$GEOBATCH_TEMP_DIR -Dunredd-ovr=$UNREDD_OVR -Duser.timezone=GMT"