Monitoring Configuration¶

Overview¶

The laboratory system uses Sentry for comprehensive monitoring, including error tracking, performance monitoring, and log aggregation. This provides visibility into system health and performance issues.

Current Implementation¶

✅ Sentry Integration (Active)¶

Configuration Location: src/config/settings.py:54-71

Features Enabled: - Error Tracking: All exceptions and errors automatically sent to Sentry - Performance Monitoring: Request traces with 10% sample rate in development, 100% in production - Profiling: Sampled profiling data collection (disabled in dev, enabled in production) - PII Tracking: Personally identifiable information sent for better debugging - Environment Tagging: Development/production environment separation

Environment Variables:

SENTRY_DSN=your-sentry-dsn-here
SENTRY_ENVIRONMENT=development  # or production
SENTRY_TRACES_SAMPLE_RATE=0.1   # 10% in dev, 1.0 in prod
SENTRY_PROFILES_SAMPLE_RATE=0.0 # Disabled in dev, 1.0 in prod

Accessing Sentry¶

Debug Endpoint: /sentry-debug/ - Triggers an error for testing
Dashboard: Access your Sentry project dashboard for:
Error overview and trends
Performance traces
User sessions with errors
Release tracking

What Sentry Currently Captures¶

Errors & Exceptions¶

All unhandled exceptions
Custom error reports
JavaScript errors (if configured in frontend)
404 and 5xx server errors

Performance Data¶

Request/response times
Database query performance
External API calls
Celery task execution

Context Information¶

User ID and email (when authenticated)
Request headers and parameters
Release version
Environment (dev/prod)

Missing Components¶

❌ Custom Metrics¶

While Sentry captures performance traces, custom business metrics are not yet implemented:

Potential Metrics to Add: - Protocol submission rates - Report generation times - Email delivery success rates - User login patterns - Sample processing throughput

Implementation Options:

# Using Sentry custom metrics
from sentry_sdk import set_measurement

# Example: Track protocol submission time
with sentry_sdk.configure_scope() as scope:
    scope.set_tag("protocol_type", "histopathology")
    set_measurement("protocol_submission_time", duration, "second")

❌ Dashboard Integration¶

No custom dashboards within the application for displaying metrics. Consider adding: - Real-time metrics sidebar - Health check page - Admin panel with key metrics

Recommended Next Steps¶

1. Set Up Alerts¶

# In Sentry dashboard, configure alerts for:
- Error rate spikes
- Performance degradation
- New issue patterns

2. Add Custom Metrics¶

# Track important business KPIs
def track_protocol_submission(protocol):
    sentry_sdk.add_breadcrumb(
        category='protocol',
        message=f'Protocol {protocol.temporary_code} submitted',
        level='info',
        data={
            'protocol_type': protocol.get_protocol_type(),
            'veterinarian_id': protocol.veterinarian_id,
        }
    )

3. Performance Optimization¶

Use Sentry profiling data to identify bottlenecks
Monitor database query performance
Track Celery task execution times

4. Log Integration¶

Ensure all important actions create breadcrumbs
Add context for better debugging
Track user flows through the system

Security Considerations¶

PII is enabled - Ensure Sentry DSN is properly secured
Review what data is sent to Sentry
Consider data retention policies
Use source maps for frontend errors

Monitoring Best Practices¶

Review Sentry regularly - Check for new errors weekly
Set up meaningful alerts - Avoid alert fatigue
Add context to errors - Include relevant business context
Track releases - Mark deployments to correlate errors
Monitor performance trends - Watch for degradation over time

Conclusion¶

Sentry provides excellent coverage for error tracking and performance monitoring. The system is well-monitored for operational issues. The main gap is custom business metrics, which can be added as needed based on business priorities.

Status: ✅ Core monitoring functional with Sentry