iQ MuleSoft Study Group (09/18): When the Mule Runs Amuck, Part 2: Retries, Resilience & Duplicates

In this Integration Quest MuleSoft Study Group, Diane Kesler returns for When the Mule Runs Amuck, Part 2: Retries, Resilience & the Attack of the Duplicates, continuing the conversation about designing an error-handling strategy before adding Mule configuration. Diane opens with a real-world Salesforce outage and the recovery work that followed, including a MuleSoft Platform Event listener that needed to reconnect after service was restored. The discussion highlights the importance of monitoring, useful logs, and alerts that help teams recognize when an integration has stopped processing. Attendee questions also spark a discussion about Change Data Capture versus Platform Events and when each might be appropriate. The group then works through the Error Handling Strategy Discovery Challenge, discussing business versus technical errors, transient failures, and who can actually resolve a problem. Diane connects those decisions to local and global error handling, On Error Continue versus On Error Propagate, and the responses an API consumer should receive. Retries bring another important question: what if the downstream system completed the request, but the response never arrived? Diane discusses the risk of duplicate processing when retrying writes, why idempotency matters, and what teams should consider when recovery attempts fail. The conversation also covers correlation IDs, protecting sensitive information in logs, and deciding which errors to address before going live. The session closes with a walkthrough of the flight application solution, including APIKit-generated error handlers, validation and custom error mapping, global error handlers, and Try scopes inside Scatter-Gather routes. Diane explains how the solution clears failed-route payloads so available flight results can still be returned, and challenges attendees to test that behavior and decide whether they would make the same choices. The challenge continues in Part 3: The Final Roundup, with more discussion of common error handlers, application-specific failures, and the reasoning behind attendees’ strategies. Whether you’re a MuleSoft developer, architect, consultant, or preparing for certification, this session offers practical questions to help you plan for failures, communicate useful responses, and recover with fewer surprises. Items from the chat: https://status.salesforce.com/product... https://drive.google.com/drive/folder... diane@integration.quest Catch up on Part 1:    • Free iQ MuleSoft Study Group (09/11): When...