Monitoring & Diagnostics

AuthProxy provides comprehensive monitoring capabilities through OpenTelemetry integration and built-in health endpoints.

Telemetry Integration

OpenTelemetry

AuthProxy supports OpenTelemetry for distributed tracing and metrics:

{
  "UptraceDsn": "https://TOKEN@api.uptrace.dev/PROJECT_ID",
  "ServiceName": "AuthProxy",
  "ServiceId": 10
}

Uptrace Dashboard

When configured, Uptrace provides:

• Request latency analysis
• Error rate tracking
• Dependency mapping
• Custom metrics

Health Endpoints

System Health

GET /health

Returns overall system status including:

• Database connectivity
• Memory usage
• Active sessions count

Detailed Info

GET /auth/v1/info

Returns:

• Version information
• Available authentication methods
• Server configuration (non-sensitive)

Key Metrics

Performance KPIs

Metric	Description	Target
Latency (p50)	Median response time	< 50ms
Latency (p99)	99th percentile	< 200ms
Error Rate	Failed requests percentage	< 0.1%
Throughput	Requests per second	5000+

Authentication Metrics

Metric	Description
Login Success Rate	Successful authentications / total attempts
Session Duration	Average active session time
Key Usage	Authentication method distribution
Failed Attempts	Blocked/failed login attempts

Resource Metrics

Metric	Description
CPU Usage	Processor utilization
Memory	RAM consumption
Connections	Active database connections
Cache Hit Rate	File cache effectiveness

Logging

Structured Logging

All logs are output in JSON format for easy parsing:

{
  "timestamp": "2024-01-15T10:30:00Z",
  "level": "Information",
  "message": "User authenticated",
  "userId": "638412345678901234",
  "method": "Fido2Key",
  "ip": "203.0.113.10"
}

Log Levels

Level	Usage
Debug	Detailed diagnostic information
Information	Normal operational events
Warning	Potential issues
Error	Failures requiring attention
Critical	System-level failures

Authentication Logging

The authentication log captures user id, login method, client device details, source IP, and location enrichment. Operators normally review this data through the admin panel or connected observability stack rather than querying tables directly.

Alerting

Recommended Alerts

Condition	Severity	Action
Error rate > 1%	Warning	Investigate logs
Latency p99 > 500ms	Warning	Check resources
Failed logins spike	High	Check for attack
Database connection failed	Critical	Immediate response
Memory > 90%	Warning	Scale or optimize

Integration Options

• Email: SMTP notification support
• Webhooks: Custom HTTP callbacks
• Telegram: Bot notifications
• Uptrace: Built-in alerting

Troubleshooting

Common Issues

High Latency

• Check database query performance
• Verify cache hit rates
• Review active session count

Memory Growth

• Monitor file cache size
• Check for session leaks
• Review route configuration

Authentication Failures

• Verify key configuration
• Check FIDO2 origin settings
• Review rate limiting thresholds

Debug Mode

Enable detailed logging in development:

{
  "Logging": {
    "LogLevel": {
      "Default": "Debug",
      "Microsoft.AspNetCore": "Warning"
    }
  }
}