Writing an Agent

Agents are one of the primary components of OCS. They are long running software servers that make Tasks and Processes (i.e. functions) available to OCS Clients on the network. They can also send data across the network via OCS Feeds. This guide will teach you how to write a basic Agent that contains one Task and one Process and can publish data to a Feed.

Note

Throughout this guide we will run our Agent. We assume you have OCS installed, have followed the Quickstart guide, have at least a crossbar server running, and an OCS site config file that we can add our Agent configuration to.

• Minimum Agent Structure
• Adding a Task
• Adding a Process
• Implement Logging
• Operation Locking
• Using OCS Feeds
• Adding Agent Arguments
• Dockerizing an Agent
• Building Images Automatically
• Next Steps