Managing agents
Use Admin Studio to manage agents to debug and trace any problems with agent activity. You can also start, stop, restart, or trace agents.
You can view a list of all the available agents across the whole cluster, on the current node, or on a remote node. You can check the status of an agent, see the number of running or stopped agents on a specific node, check the nodes on which the specific agent is running, or see the exception details for agents. You can also enable sending notifications on agent failure and specify the email address to which the notification is sent.
To access the Agent Management landing page and view the list of agents and the configuration of email notifications on agent failure, you must have the PegaRULES:SysOpsObserver role. To perform all available actions on agents, for example, to start or stop an agent, and configure email notifications on agent failure, you must also have the PegaRULES:SysOpsAdministrator role.
- Implementing agents
An agent is an internal background process operating on the server that runs activities on a periodic basis. In a multinode cluster, an agent can run on multiple nodes.
- Starting, stopping, and restarting agents
Normally, all agents start each time that you start your Pega Platform system. Agents running on a specific node start when that node starts.
- Setting up agent notifications
You can enable notification on agent failure to receive emails on each occurrence of the PEGA0010 alert that is triggered when an exception occurs during agent processing.
- Understanding agent queues
An agent queue is a list of work to be done by an agent. Use Admin Studio to view work items in the agent queues by their status and take action on them. This information is useful for diagnosing agent processing problems.
Previous topic Changing a queue processor rule state Next topic Implementing agents