People often refer to the various Rapid Automation (RA) solutions as agents, but this way of speaking can be misleading. Although one does define different types of RA agent in the AE, there actually is only one RA agent program: the RA Core Agent, ucxjcitx. RA solutions. like RA Informatica (PowerCenter), RA FTP and RA Web Services REST run on top of the RA Core, and the same core is used for all RA solutions.
RA solutions are installed by loading them in the Automation Engine database using the ucybdbld program. Once an RA solution has been loaded in to the DB, you can define and configure a new agent that uses that solution. Once it is set up correctly (as garan18 described above), it works like this:
- The RA core starts up and reads its INI file.
- The RA connects to the Automation Engine and identifies itself by name.
- The AE looks up the details of that Agent from its DB, including the type of RA solution that this agent uses, and the agent’s configuration settings.
- The AE loads the appropriate RA solution file (a JAR file) from the AE DB and sends it to the RA Agent, along with the configuration settings.
- The RA agent places the solution file in a temporary location.
- The RA agent starts the solution using the configuration details loaded from the agent definition.
Communication between the RA solution and the Automation Engine is passed via the RA Core.
Andrew, please correct me if I got any of the specifics wrong.