> ## Documentation Index > Fetch the complete documentation index at: https://docs.markifact.com/llms.txt > Use this file to discover all available pages before exploring further. # Error Handling > Configure how your workflows handle node failures with automatic retries and continuation settings Control how your workflows respond to failures by configuring error handling settings for any node. *** ## How to Access 1. **Click the three dots (⋯)** on any node 2. **Select "Execution Settings"** 3. **Configure your error handling options** Execution Settings Dialog Execution Settings Dialog *** ## Use Cases **BigQuery Reporting Pipeline:** If your Meta Ads report fails, retry 3 times with 10-second delays, but continue the workflow to still generate reports from other data sources. **API Data Collection:** When Meta Ads Get Report hits rate limits, retry 2 times with 15-second delays to wait for the API to recover. *** ## Settings ### Continue on Fail Keep the workflow running even if this node fails. Useful for optional steps like notifications or secondary data sources. ### Retry on Fail Automatically retry the node if it fails. Good for API calls that might hit rate limits or temporary network issues. ### Max Retries How many times to retry (1-5 attempts). More retries = higher chance of success but slower execution. ### Retry Delay Wait time between retries (1-30 seconds). Longer delays help with rate limits and service recovery. *** ## Common Configurations **For API calls:** * Retry: 2-3 times * Delay: 5-10 seconds * Continue on Fail: ✓ (if optional) **For critical operations:** * Retry: 4-5 times * Delay: 15-30 seconds * Continue on Fail: ✗ **For notifications:** * Retry: 1-2 times * Delay: 5 seconds * Continue on Fail: ✓ *** ## Tips * **Start simple**: 2 retries, 5-second delay * **Monitor logs**: Adjust based on what actually fails * **Enable "Continue on Fail"** for non-critical steps * **Use longer delays** for rate-limited APIs