How to debug an outbound campaign failure in production?

General
,
1

How to debug an outbound campaign failure in production?

2

best way to debug an outbound campaign failure in production is to trace the failed run using its exact run time and run ID.
Open the campaign run details first and note the status, error message, and the time the failure happened.
Then check the production logs for that same time window, especially the background processing or agent logs that executed the campaign.
Find the first real error in the stack, not the later follow-up errors, because that usually shows the true cause.
Match that error to the campaign step that failed, such as audience selection, treatment generation, or delivery.
Once you identify that failing step, fix that specific issue and rerun the campaign with the same setup to confirm it works.

3

Diagnostic Approach for Outbound Campaign Failures:

  1. Access Campaign Status:
    Navigate to: Customer Decision Hub → Campaigns → [Your Campaign]
    Check Status section for error details
    Click Run details link at bottom of Status section
    Pegasystems Documentation
    Pegasystems Documentation

  2. Review Data Flow Execution:
    Navigate to dataflow landing page > Open the PR- run
    Identify which shape/step failed from the component statistics and validate the failed records.

  3. Common Outbound Failure Patterns from PDC/PegaRules/PegaAlert logs:
    A. Cassandra timeout exceptions
    B. Long execution times
    C. Strategy compilation failures
    D. Audience/segment data source issues

  4. Capture and analyze Decision Profiler for latencies/CPU spikes.

  5. Check Resource Constraints:
    A. Monitor Batch tier pod scaling
    B. Review JVM heap and GC metrics
    C. Check Cassandra query performance

4

Hi @AnilP1004 , you can follow the below troubleshooting approach:

Scenario 1: The system does not purge the backend dataflow

Check the backend dataflow execution, sort the components with errors, and analyze the error details to determine the root cause.

Scenario 2: The system successfully purges the backend dataflow

  1. Review the system logs for any warnings or errors related to the dataflow.
  2. Check the PDC (Pega Predictive Diagnostic Cloud) for:
    • Dataflow-related issues
    • Campaign execution issues
    • Any associated alerts or exceptions
  3. If no obvious issues are found, execute the NBA_TopLevel strategy in standalone mode with the relevant persona to validate whether the Next-Best-Action processing is functioning correctly