Continuous Error Tracking (CET)
The Harness Continuous Error Tracking (CET) module provides real-time application error monitoring and diagnosis capabilities, helping development teams quickly discover, locate, and fix problems in production environments.Core Features
Real-Time Monitoring
- Real-Time Error Collection: Immediately capture exceptions and errors in applications
- Performance Impact Analysis: Understand the impact of errors on user experience
- Error Trends: Track changes in error quantity and frequency
- Multi-Language Support: Java, Python, Node.js, Go, and other major languages
Root Cause Analysis
- Stack Tracing: Complete error stack information
- Variable State: Variable values when errors occurred
- Log Correlation: Correlate related logs for easier debugging
- Distributed Tracing: Cross-service error tracking
Team Collaboration
- Alert Notifications: Alert through Slack, Email, PagerDuty, and other channels
- Issue Assignment: Assign errors to corresponding developers
- Status Management: Track error handling progress
- Team Filtering: Filter errors by team or service
Smart Aggregation
- Fingerprinting: Intelligently identify and aggregate similar errors
- Version Correlation: Correlate errors with code versions
- Duplicate Detection: Avoid duplicate alerts
Use Cases
| Scenario | CET Features |
|---|---|
| Production Issue Discovery | Real-time alerting |
| Fast Debugging | Complete stack and variable information |
| Regression Detection | New version introduced error detection |
| SLA Monitoring | Error rate trend tracking |
Getting Started
1. Install Agent
Install the corresponding CET Agent based on your tech stack:2. Configure Agent
Set up data collection and upload configuration:3. Deploy Application
Deploy the application with the CET Agent.4. View Errors
View collected errors through the Harness console.5. Configure Alerts
Set alert rules and notification channels.Error Information
CET collects the following error information:| Category | Information |
|---|---|
| Exception Type | Exception name and class |
| Stack Trace | Complete call stack |
| Request Info | HTTP request details |
| User Info | User who triggered the error |
| Environment Context | Server info, version, etc. |
Best Practices
- Full Coverage: Deploy Agents in all critical services
- Filter Noise: Configure appropriate ignore rules to filter unimportant errors
- Quick Response: Set reasonable alert thresholds to avoid alert fatigue
- Continuous Improvement: Regularly review error trends to identify systemic issues