Debugging Failed Bolt Schedules
Last updated
Was this helpful?
Last updated
Was this helpful?
When your Bolt schedules encounter issues and fail to execute successfully, it's crucial to have the right tools and methods to quickly identify and resolve the problems. This documentation covers the various ways you can debug failed Bolt runs and get your data pipelines back on track.
The easiest way to detect failed Bolt runs is to set up notifications. You can configure your Bolt schedules to send alerts whenever a run fails. These notifications provide a summary of the issue(s), and provide links to the Individual Run Details interface, providing a seamless starting point for debugging.
Alternatively, you can go to the Bolt home screen and identify failed runs via the "Status" column of the Bolt Schedule List. The "Status" is representative of the most recent execution of a given schedule.
Bolt provides several tools to investigate and resolve failed runs, including:
Viewing Model Execution Timeline
Dino AI Summary
The run logs provide a detailed breakdown of each scheduled run, offering insights into execution status, performance, and any encountered issues. The logs are divided into three tabs:
{% tabs %} {% tab title="Summary" %} Overview: Displays high-level execution details for commands, showing the overall completion status, duration, and success metrics.
Warnings and Errors: Lists any warnings or errors encountered during the run, such as deprecated configurations or unused paths.
Suggested Actions: Recommendations to address identified warnings and errors, including updates for alignment with best practices.
Use Case: Quickly assess if the run was successful, spot any configuration issues, and take corrective actions. {% endtab %}
{% tab title="Console Logs" %} Overview: Shows a line-by-line output of the command execution, starting with initialization and progressing through each execution step, including model or task processing times and completion summaries.
Use Case: Ideal for reviewing the detailed flow of the run, identifying stages where issues may have occurred, and monitoring performance. {% endtab %}
{% tab title="Debug Logs" %} Overview: Provides in-depth technical details, including system interactions, resource allocation, thread usage, and connection status to support troubleshooting.
Use Case: Best for advanced diagnostics and analyzing system performance, especially when investigating anomalies or unexpected behaviors. {% endtab %} {% endtabs %}
Each tab gives you a targeted view of the run details, providing a complete understanding of your pipeline's performance.
{% @arcade/embed flowId="TgVn754OtMX4OBm6Z9rV" url="https://app.arcade.software/share/TgVn754OtMX4OBm6Z9rV" %}
The Bolt UI makes it easy to explore and navigate the run logs. You can click through the different tabs and use the arrows to move between individual run IDs, allowing you to quickly and efficiently investigate failed executions.
Paradime's Dino AI feature analyzes the run logs and provides a summary of any issues or areas for improvement. This can help you quickly identify the root cause of a failure and suggest next steps for resolution.
If a specific model or task within your Bolt schedule failed to execute successfully, you can view the SQL that was run for that component. This can help you debug any issues with the model logic or data quality.
By leveraging these various debugging tools and methods, you can quickly identify the cause of failed Bolt runs and take the necessary steps to get your data pipelines back on track.
