Based on the research and GitHub issues, I can explain what's happening here. No, it's not a logic bomb - it's actually a protective mechanism in Prefect, though it can be frustrating when it occurs. Here's what's happening: 1. What the message means: This message appears when a flow run that is currently handling retries cannot be moved back to a PENDING state. It's a safeguard to prevent race conditions and ensure proper flow execution. 2. Why it occurs repeatedly: The repetitive nature occurs because: - A worker is trying to submit/execute the flow run - The server refuses the transition because the flow run is already handling retries - This cycle continues as the worker keeps trying to submit the flow run 3. Common causes: - Task retries occurring when infrastructure changes (like worker restarts) - Multiple workers trying to handle the same flow run - Long-running tasks that might appear stuck to the orchestration layer 4. Impact: When this happens, the flow run can get stuck in this loop and might need manual intervention. Solutions: 1. If this is happening frequently, you might need to: - Cancel the stuck flow run and restart it - Check if your tasks have appropriate timeout settings - Ensure your worker configuration is stable and not frequently restarting 2. For Prefect 3.x specifically, you might want to: - Review your retry configurations - Ensure you're not accidentally running multiple workers for the same work pool - Consider implementing proper error handling in your tasks to manage retries more explicitly Would you like me to help you investigate any specific aspect of this issue? For example, we could look at: - Your worker configuration - Your retry settings - Specific task implementations that might be triggering this