flowchart TD
    A[Docker Health Check] --> B{Docker Failures > Threshold?}
    B -->|yes| C[Return Error: Too Many Failures]
    B -->|no| D[Container Inspect API Call]
    D --> E{Inspect Successful?}
    E -->|no| F[Increment Failure Count]
    E -->|yes| G[Parse Container State]

    G --> H{Container Status}
    H -->|dead/exited/paused/restarting/removing| I[Unhealthy: Container State]
    H -->|created| J[Unhealthy: Not Started]
    H -->|running| K{Health Check Configured?}

    K -->|no| L[Return Error: No Health Check]
    K -->|yes| M[Check Health Status]
    M --> N{Health Status}
    N -->|healthy| O[Healthy]
    N -->|unhealthy| P[Unhealthy: Last Log Output]

    I --> Q[Reset Failure Count]
    J --> Q
    O --> Q
    P --> Q

Key Features:

Intercepts Docker API responses to extract container state
Tracks failure count with configurable threshold (3 failures)
Supports containers with and without health check configurations
Returns detailed error information from Docker health check logs

HTTP Health Check

Performs HTTP/HTTPS health checks using fasthttp for optimal performance.

Flow:

flowchart TD
    A[HTTP Health Check] --> B[Create FastHTTP Request]
    B --> C[Set Headers and Method]
    C --> D[Execute Request with Timeout]
    D --> E{Request Successful?}

    E -->|no| F{Error Type}
    F -->|TLS Error| G[Healthy: TLS Error Ignored]
    F -->|Other Error| H[Unhealthy: Error Details]

    E -->|yes| I{Status Code}
    I -->|5xx| J[Unhealthy: Server Error]
    I -->|Other| K[Healthy]

    G --> L[Return Result with Latency]
    H --> L
    J --> L
    K --> L

Key Features:

Uses fasthttp for high-performance HTTP requests
Supports both GET and HEAD methods
Configurable timeout and path
Handles TLS certificate verification errors gracefully
Returns latency measurements

H2C Health Check

Performs HTTP/2 cleartext (h2c) health checks for services that support HTTP/2 without TLS.

Flow:

flowchart TD
    A[H2C Health Check] --> B[Create HTTP/2 Transport]
    B --> C[Set AllowHTTP: true]
    C --> D[Create HTTP Request]
    D --> E[Set Headers and Method]
    E --> F[Execute Request with Timeout]
    F --> G{Request Successful?}

    G -->|no| H[Unhealthy: Error Details]
    G -->|yes| I[Check Status Code]
    I --> J{Status Code}
    J -->|5xx| K[Unhealthy: Server Error]
    J -->|Other| L[Healthy]

    H --> M[Return Result with Latency]
    K --> M
    L --> M

Key Features:

Uses HTTP/2 transport with cleartext support
Supports both GET and HEAD methods
Configurable timeout and path
Returns latency measurements

FileServer Health Check

Checks if a file server root directory exists and is accessible.

Flow:

flowchart TD
    A[FileServer Health Check] --> B[Start Timer]
    B --> C[Stat Directory Path]
    C --> D{Directory Exists?}

    D -->|no| E[Unhealthy: Path Not Found]
    D -->|yes| F[Healthy: Directory Accessible]
    D -->|error| G[Return Error]

    E --> H[Return Result with Latency]
    F --> H
    G --> I[Return Error]

Key Features:

Simple directory existence check
Measures latency of filesystem operation
Distinguishes between "not found" and other errors
Returns detailed error information

Stream Health Check

Checks stream endpoint connectivity by attempting to establish a network connection.

Flow:

flowchart TD
    A[Stream Health Check] --> B[Create Dialer]
    B --> C[Set Timeout and Fallback Delay]
    C --> D[Start Timer]
    D --> E[Dial Network Connection]
    E --> F{Connection Successful?}

    F -->|no| G{Error Type}
    G -->|Connection Errors| H[Unhealthy: Connection Failed]
    G -->|Other Error| I[Return Error]

    F -->|yes| J[Close Connection]
    J --> K[Healthy: Connection Established]

    H --> L[Return Result with Latency]
    K --> L

Key Features:

Generic network connection check
Supports any stream protocol (TCP, UDP, etc.)
Handles common connection errors gracefully
Measures connection establishment latency
Automatically closes connections

Common Features

Error Handling

All health checks implement consistent error handling:

Temporary Errors: Network timeouts, connection failures
Permanent Errors: Invalid configurations, missing resources
Graceful Degradation: Returns health status even when errors occur

Performance Monitoring

Latency Measurement: All checks measure execution time
Timeout Support: Configurable timeouts prevent hanging
Resource Cleanup: Proper cleanup of connections and resources

Integration

These health checks are used by the monitor package to implement route-specific health monitoring:

HTTP/HTTPS routes use HTTP health checks
File server routes use FileServer health checks
Stream routes use Stream health checks
Docker containers use Docker health checks with fallbacks