π¨ Common Issues#
Quick solutions to common Stream DaQ problems.
Installation Issues#
Problem: pip install streamdaq fails
Solutions:
- Ensure Python >= 3.11: python --version
- Update pip: pip install --upgrade pip
- Use virtual environment: python -m venv venv && source venv/bin/activate
Configuration Problems#
Problem: βNo data in windowsβ
Solutions: - Check time column format matches your data - Verify data source is producing records - Ensure window size matches data frequency
Problem: High memory usage
Solutions:
- Reduce wait_for_late parameter
- Use larger window sizes for high-volume streams
- Check for data leaks in custom functions
Quality Check Issues#
Problem: Assessments always fail/pass
Solutions: - Verify threshold values against actual data - Test with known good/bad data samples - Use debug output to see measure values
Problem: Custom assessment functions are slow
Solutions: - Simplify assessment logic - Use string assessments when possible - Cache expensive calculations
Output Problems#
Problem: Results not appearing
Solutions: - Check sink operation configuration - Verify destination permissions and connectivity - Test with console output first
Problem: Delayed results
Solutions:
- Reduce window size
- Decrease wait_for_late tolerance
- Use sliding windows instead of tumbling
Multi-Source Issues#
Problem: Tasks not running simultaneously
Solutions: - Use single StreamDaQ instance with multiple tasks - Check task criticality settings - Verify all tasks are properly configured
Problem: Critical task failures stop everything
Solutions: - Review which tasks are marked as critical - Use try-catch blocks for error handling - Consider making non-essential tasks non-critical
Getting Help#
Check the π User Guide for detailed configuration
Review π‘ Examples for working examples
Search existing issues on GitHub
Create a new issue with minimal reproduction case