Healthz

Overview

GET /healthz returns the current bridge service health and runtime state.

It is not only a static liveness endpoint. It is intended as a runtime-oriented inspection endpoint.

Response Shape

A typical response includes:

• ok
• service
• status
• timestamp
• uptimeSec
• runtime
• config
• bridge
• sessions

Runtime Section

runtime describes the current process and environment, for example:

• nodeVersion
• platform
• arch
• pid
• nodeEnv

Use cases:

• confirm runtime version
• distinguish deployment environments
• investigate environment mismatches

Config Section

config exposes the currently effective config summary, for example:

• host
• port
• logLevel
• ffmpegPath
• rtspUrlTemplate
• streamRestartDelayMs
• streamMaxRestarts
• streamIdleTimeoutMs
• streamSweepIntervalMs

Use cases:

• confirm .env is applied
• verify fallback/default behavior
• diagnose configuration issues

Bridge Section

bridge describes the current manager runtime state, for example:

• activeSessionCount
• idleTimeoutMs
• sweepIntervalMs
• lastSweepAt

Use cases:

• check whether active sessions exist
• check whether the sweep loop is running
• verify idle detection is progressing

Sessions Section

sessions is the list of current managed session snapshots.

Each snapshot typically includes:

• streamId
• sessionId
• state
• pid
• clientCount
• restartCount
• lastRestartAt
• lastStartedAt
• lastStoppedAt
• lastDataAt
• lastErrorAt
• lastExitCode
• lastExitSignal

How to Use It in Debugging

Recommended usage scenarios:

1. Startup verification

Confirm the service is up and config has been loaded.

2. After opening a live connection

Confirm:

• activeSessionCount increases
• a new session appears
• clientCount looks correct

3. After disconnecting

Confirm:

• the session is cleaned up
• activeSessionCount drops
• idle session removal has happened

4. Recovery validation

Inspect:

• restartCount
• lastRestartAt
• lastDataAt
• lastExitCode
• lastExitSignal

Known Limits

The current /healthz endpoint is still a Phase 1 implementation. It prioritizes usefulness over completeness.

It does not yet include:

• Prometheus metrics
• deeper performance statistics
• historical session aggregation
• alerting integration

Its role is:

• current runtime inspection
• local debugging
• deployment verification support