DG is designed to pull only specific sources of data (session logs, dashboard, queue snapshot and from processes that use the internal DG VBO) and push it to certain types of destination (file, HTTP, Splunk or database). In simple terms the way it works is:
- The type of data to pull is selected in System> Data Gateways > Settings
- The destination is defined in System> Data Gateways > Configuration
- The BP application server runs DG as a continuous service
- The service watches the DG pipeline table in the BP database
- The BP application writes data into the pipeline table
- The DG service removes the data from the pipeline table and pushes it to the destination specified in the DG Configuration
- It's possible to have multiple configurations to push to different destinations according to the type of data.
DG is an implementation of Logstash and as such is highly customisable. The BP UI helps the user to create a configuration, but it's also possible to handcraft the configuration to make DG go beyond the default options, eg push to a wide variety of outputs. However, be warned that DG is a service and will send output continuously and rapidly. Also bear in mind that DG is designed to pull from only one specific source; even though technically it can use other sources, DG has not been tested to do so and Blue Prism will not take responsibility for the potential impact of bespoke DG input configurations.
