FastAPI Backend Development Bootcamp · 课时

服务发现与健康检查

学习微服务如何动态发现彼此，以及健康检查如何确保流量只发送到健康的实例。

服务发现与健康检查 是 CoddyKit 上的免费 FastAPI Backend Development Bootcamp 课时。这是第 4 节课，共 4 节。你可以在下方免费阅读本课时的完整内容 — 然后在浏览器中使用内置代码编辑器和全天候 AI 导师进行实践。这是 FastAPI Backend Development Bootcamp 学习路径的一部分，你的进度在网页和 CoddyKit 应用中同步。 FastAPI Backend Development Bootcamp 课程共包含 4 节课。

本课时的部分内容尚未翻译，以英文显示。

The Discovery Problem

In a microservices system, instances start, stop, and move across hosts constantly. Hardcoding IP addresses breaks immediately.

Service discovery lets a service find healthy instances of another service by name at runtime.

Client-Side vs Server-Side

Two patterns exist:

Client-side: the caller queries a registry and picks an instance
Server-side: a load balancer or gateway resolves the name and forwards the request

Service Registries

A registry stores live instances. Popular options include Consul, etcd, and Kubernetes DNS.

Services register on startup and deregister on shutdown.

Registering a Service

On boot, a service announces its address to the registry.

import httpx

async def register():
    await httpx.AsyncClient().put(
        "http://consul:8500/v1/agent/service/register",
        json={"Name": "orders", "Address": "10.0.0.5", "Port": 8000})

Why Health Checks Matter

A registered instance might still be broken. Health checks ensure the registry only routes to instances that are actually able to serve requests.

A FastAPI Health Endpoint

Expose a lightweight endpoint that returns quickly when the service is healthy.

from fastapi import FastAPI

app = FastAPI()

@app.get("/health")
def health():
    return {"status": "ok"}

Liveness vs Readiness

Two distinct checks:

Liveness: is the process alive? If not, restart it.
Readiness: can it serve traffic now with DB connected and caches warm? If not, hold traffic.

A Readiness Check

Readiness verifies downstream dependencies before accepting traffic.

from fastapi import Response, status

@app.get("/ready")
async def ready(response: Response):
    if not await db_is_up():
        response.status_code = status.HTTP_503_SERVICE_UNAVAILABLE
        return {"ready": False}
    return {"ready": True}

Kubernetes Probes

Kubernetes uses these endpoints as probes to manage pods automatically.

livenessProbe:
  httpGet:
    path: /health
    port: 8000
readinessProbe:
  httpGet:
    path: /ready
    port: 8000

Graceful Deregistration

On shutdown, deregister so callers stop routing to a dying instance, and finish in-flight requests first.

@app.on_event("shutdown")
async def shutdown():
    await deregister_from_registry()

TTL and Stale Entries

Registry entries usually carry a TTL. If a service stops sending heartbeats, its entry expires automatically so callers stop routing to a dead instance.

Quick Check

Test your discovery knowledge.

Recap

You learned how services find and trust each other:

Service discovery resolves services by name via a registry
Health checks ensure only healthy instances get traffic
Liveness restarts, readiness gates traffic

Together they keep a dynamic microservices system resilient.

常见问题解答

「服务发现与健康检查」课时是免费的吗？

是的 — 「服务发现与健康检查」的完整文本可在网页上免费阅读。要进行交互式练习（内置代码编辑器和全天候 AI 导师）并解锁 FastAPI Backend Development Bootcamp 课程的其余内容，请升级到 CoddyKit PRO。 FastAPI Backend Development Bootcamp 课程共包含 4 节课。

「服务发现与健康检查」这节课中我会学到什么？

学习微服务如何动态发现彼此，以及健康检查如何确保流量只发送到健康的实例。你通过在浏览器中直接运行的动手代码来练习 FastAPI Backend Development Bootcamp，全天候 AI 导师会在你学习这节课的过程中回答你的问题。

学习 FastAPI Backend Development Bootcamp 需要有经验吗？

无需任何先前经验。CoddyKit 上的 FastAPI Backend Development Bootcamp 课程适合初学者到高级学习者，你可以从这里开始或从头开始，按照自己的节奏学习。这是第 4 节课，共 4 节。

「服务发现与健康检查」课时需要多长时间？

大多数 CoddyKit 课程大约需要 5–10 分钟。每节课都很精短且互动，所以你能稳步进步，并在网页和应用中从离开的地方继续。

我能在这节 FastAPI Backend Development Bootcamp 课中编写并运行代码吗？

能。每节 FastAPI Backend Development Bootcamp 课都包含内置代码编辑器，你可以在浏览器中直接编写并运行真实代码，并获得即时 AI 反馈 — 无需本地设置。

此课程中的所有课时

← 返回 FastAPI Backend Development Bootcamp