Monitoring¶

Built-in health checks, metrics, and a live dashboard.

Setup¶

from shakti.monitoring import Monitor

monitor = Monitor(title="My App Monitor")
monitor.init_app(app)

Visit http://localhost:8000/monitor/ — auto-refreshes every 10 seconds.

Custom health checks¶

@monitor.health_check("database")
async def check_db() -> str:
    async with db.session() as session:
        await session.execute(text("SELECT 1"))
    return "connected"

@monitor.health_check("ai")
async def check_ai() -> str:
    # ping your AI provider
    return "ok"

Endpoints¶

Route	Description
`GET /monitor/`	Live dashboard
`GET /monitor/health`	Basic health
`GET /monitor/health/live`	Kubernetes liveness probe
`GET /monitor/health/ready`	Kubernetes readiness probe
`GET /monitor/health/full`	Full health + system info
`GET /monitor/metrics`	JSON metrics
`GET /monitor/metrics/endpoints`	Per-endpoint stats
`GET /monitor/metrics/requests`	Recent request log

Dashboard shows¶

Uptime, total requests, avg/p95 response time, error rate
CPU, memory, disk usage (progress bars)
Health check statuses
Per-endpoint request counts and response times
Recent requests log with status codes