As you already know, you can run a job scheduler multiple times a day. However, for Pega Platform 8.1 through 8.4, currently, the only options are run After every N minute(s) or After every N hour(s), where N is an integer that specifies the delay before starting the next run -- i.e., run After every N second(s) is not currently supported.
Thanks for the reply BelleIsle, The reason for asking this question is, Pega suggests to upgrade to latest version. In recent times we upgraded our systems from 7.3.1 to 8.2.5 and Pega suggested us to replace the Advance agents with Job scheduler.
We are integrating with middle ware system to legacy system to create work object in Pega.
Once the case get resolve in Pega we need to again integrate with legacy system once case is completed, as mentioned we used Advance agent to process the records for every 30 seconds in the queue.
What will be the best approach do we need to continue with same Advance agent in latest versions too?
Even in 8.4, you can continue to leverage Agents, but to echo what others at Pegasystems said, I recommend that you change your implementation sooner than later.
In Pega Infinity (8.x), as you already know, Job Scheduler and Queue Processor replace Agents. However, the changes in rule types also require taking a different approach. To help me help you, would you kindly elaborate on the inbound and outbound integrations? In other words:
For the data coming into Pega from the middleware system or legacy, how is the integration implemented? Are you leveraging a Service Rest? A file listener? Reading from a database table?
When you send data back to the middleware system or legacy, how is the integration implemented? Are you leveraging a Connect Rest? Writing to a batch file? Writing to a database table?
I need a little bit more details to suggest an approach for removing the Agent Queues.
Posted: 9 months ago
Updated: 9 months ago
Posted: 24 Jun 2020 11:36 EDT Updated: 24 Jun 2020 11:51 EDT
For inbound, assuming you were using Agent Queues, you can leverage Queue Processor instead. A suggested approach is something like Service Rest > Activity with the Queue-For-Processing method > Queue Processor with Activity that creates the work object in Pega.
Queue Processor will not work for the outbound integration as it processes a single message. Instead, you could leverage a staging table to hold the outbound messages. Using a Job Scheduler that runs every 1 minute, you can leverage an Activity to retrieve up to X messages (e.g., 1 000) from the staging table using a report definition or Obj method, and write them to the file.