Table of Contents

Pega-ProcessEngine agent


Only available versions of this content are shown in the dropdown

The Pega-ProcessEngine agent performs three types of background processing:

  • Case type calculations
  • Incoming email processing
  • Flow dependency processing

Process Case Type Calculations

The System-Queue-CaseType.pzProcessCalculationRequest activity supports calculations defined in case type rules associated with cover flows. If your application includes case type rules, enable the Agent Schedule instance for this agent.

When an input property value included on the Calculations tab is updated, it invokes an associated declare trigger rule, which updates the Property to Calculate value on the cover object. The agent attempts to obtain a lock so that the trigger can update this value. If unsuccessful, the agent re-tries in 30 seconds to obtain a lock for the update.

The queue for this agent is defined by instances of the System-Queue-CaseType class.

Process Email Requests

The Work-ProcessEmail.pyProcessEmailRequest activity supports email conversations that allow email responses to be used for approvals or other actions on work items. For example, you can configure the application to send an approver an email that contains hyperlinks to non-local flow actions. The approver can click the appropriate flow action, then send the response. An email listener validates the response, creates a work item using the service activity pyCreateAndManageWorkFromEmail, then delegates work item processing to the pyProcessEmailRequest agent.

This pyProcessEmailRequest agent is disabled by default in the Pega-ProcessEmail-Agents rule.

Process Flow Dependencies

The System-Queue-FlowDependency.pzProcessFlowDependency activity supports dependent assignment processing. A Declare Trigger rule (defined in the Applies To class that is being waited on), running the Work-.pxCheckFlowDependencies trigger activity, must be able to obtain a lock on the waiting assignment in order to complete it and advance the flow. If it cannot obtain a lock, the work is queued to be completed later. The agent periodically retries until a lock is available.

Have a question? Get answers now.

Visit the Collaboration Center to ask questions, engage in discussions, share ideas, and help others.