Readiness Checks

A started process is not a healthy service. fuku polls until your service actually accepts requests.

How it works

Poll until healthy

When a service has a readiness check configured, fuku waits for it to pass before marking the service as ready. The next tier only begins after all services in the current tier are ready.

fuku run default

│ tier foundation
│   ○ auth        checking http localhost:8081/health...
│   ○ gateway     checking http localhost:8082/health...
│   ✓ gateway     ready    1.8s
│   ✓ auth        ready    2.1s
│
│ tier platform
│   ○ api         checking http localhost:8080/health...
│   ✓ api         ready    4.5s

Three check types

Pick the one that matches your service

HTTP

Wait for an HTTP endpoint to respond with a 2xx status code:

fuku.yaml

services:
  api:
    dir: ./api
    readiness:
      type: http
      url: http://localhost:8080/health
      timeout: 30s
      interval: 1s

TCP

Wait for a TCP port to accept connections:

fuku.yaml

services:
  gateway:
    dir: ./gateway
    readiness:
      type: tcp
      address: localhost:9000
      timeout: 10s
      interval: 1s

Log

Wait for a specific pattern to appear in the service's output:

fuku.yaml

services:
  grpc-server:
    dir: ./grpc
    readiness:
      type: log
      pattern: "gRPC server started"
      timeout: 30s

Configuration

Per-service, in fuku.yaml

Field	Description
`type`	`http`, `tcp`, or `log`
`url`	HTTP endpoint to check (http type only)
`address`	Host:port to connect to (tcp type only)
`pattern`	String to match in log output (log type only)
`timeout`	Maximum time to wait before marking as failed
`interval`	Time between check attempts (http/tcp only)

Retries

Automatic retry on failure

If a readiness check times out, fuku can retry the service start based on your retry configuration:

fuku.yaml

retry:
  attempts: 3 # Max retry attempts (default: 3)
  backoff: 500ms # Initial backoff duration (default: 500ms)

Ready to try it out?

Get Started