从 Google Kubernetes Engine 集群连接到 Redis 实例

您只能从与 Redis 实例使用相同授权网络的 Google Kubernetes Engine 集群连接到该 Redis 实例。

设置

如果您已安装 Google Cloud CLI 且已创建 Redis 实例，则可以跳过这些步骤。

  1. 安装 gcloud CLI 并初始化：

    gcloud init

  2. 按照快速入门指南创建一个 Redis 实例。记下该 Redis 实例的地区、IP 地址和端口。

准备 GKE 集群

  1. 如果您尚未创建 GKE 集群，请使用以下命令对 Google Cloud CLI 创建一个：

    1. 在 gcloud 中为此示例应用指定项目。

      gcloud config set project PROJECT_ID

    2. 在 gcloud 中设置 Compute Engine 地区配置变量。 

      gcloud config set compute/zone ZONE

    3. 创建一个名为 visitcount-cluster 的 GKE 集群。

      gcloud container clusters create visitcount-cluster --num-nodes=3 --enable-ip-alias

  2. 如果您未使用 gcloud 创建集群，请使用以下命令检索集群凭据：

    gcloud container clusters get-credentials CLUSTER_NAME --zone CLUSTER_ZONE --project PROJECT_ID
    1. CLUSTER_NAME 是 GKE 集群的名称。
    2. CLUSTER_ZONE 是集群所在的地区。
    3. PROJECT_ID 是集群和 Redis 实例所在的项目。

  3. 如果您的集群是版本 1.8 或更高版本且启用了 IP 别名，请跳过此步骤。如果您的集群是版本 1.7 或更低版本，或者集群为版本 1.8 或更高版本但未启用 IP 别名，请在尝试连接到实例之前，按照以下解决方法步骤操作：

    1. 运行以下命令，将 RESERVED_IP_RANGE 替换为实例的预留 IP 范围：

      git clone https://github.com/bowei/k8s-custom-iptables.git
cd k8s-custom-iptables/
TARGETS="RESERVED_IP_RANGE" ./install.sh
cd ..

    2. 如果您不知道实例的预留 IP 范围，可通过控制台（高级选项）或以下命令找到：

      gcloud redis instances describe INSTANCE_ID --region=REGION

    如需详细了解 IP 别名（包括如何创建启用了此设置的集群），请参阅 IP 别名文档。

示例应用

此示例 HTTP 服务器应用通过 Google Kubernetes Engine 集群建立与 Redis 实例的连接。

克隆所选编程语言的代码库，并转到包含示例代码的文件夹：

Go

git clone https://github.com/GoogleCloudPlatform/golang-samples
cd golang-samples/memorystore/redis

Java

git clone https://github.com/GoogleCloudPlatform/java-docs-samples
cd java-docs-samples/memorystore/redis

Node.js

git clone https://github.com/GoogleCloudPlatform/nodejs-docs-samples
cd nodejs-docs-samples/memorystore/redis

Python

git clone https://github.com/GoogleCloudPlatform/python-docs-samples
cd python-docs-samples/memorystore/redis

每次访问 / 端点时，此示例应用都会增加一个 Redis 计数器。

Go

此应用使用 github.com/gomodule/redigo/redis 客户端。通过运行以下命令进行安装：

go get github.com/gomodule/redigo/redis@latest

应用示例：


// Command redis is a basic app that connects to a managed Redis instance.
package main

import (
	"fmt"
	"log"
	"net/http"
	"os"

	"github.com/gomodule/redigo/redis"
)

var redisPool *redis.Pool

func incrementHandler(w http.ResponseWriter, r *http.Request) {
	conn := redisPool.Get()
	defer conn.Close()

	counter, err := redis.Int(conn.Do("INCR", "visits"))
	if err != nil {
		http.Error(w, "Error incrementing visitor counter", http.StatusInternalServerError)
		return
	}
	fmt.Fprintf(w, "Visitor number: %d", counter)
}

func main() {
	redisHost := os.Getenv("REDISHOST")
	redisPort := os.Getenv("REDISPORT")
	redisAddr := fmt.Sprintf("%s:%s", redisHost, redisPort)

	const maxConnections = 10
	redisPool = &redis.Pool{
		MaxIdle: maxConnections,
		Dial:    func() (redis.Conn, error) { return redis.Dial("tcp", redisAddr) },
	}

	http.HandleFunc("/", incrementHandler)

	port := os.Getenv("PORT")
	if port == "" {
		port = "8080"
	}
	log.Printf("Listening on port %s", port)
	if err := http.ListenAndServe(":"+port, nil); err != nil {
		log.Fatal(err)
	}
}

Java

此应用基于 Jetty 3.1 servlet。

它使用 Jedis 库：

<dependency>
  <groupId>redis.clients</groupId>
  <artifactId>jedis</artifactId>
  <version>5.1.0</version>
</dependency>

AppServletContextListener 类用于创建长效 Redis 连接池：


package com.example.redis;

import java.io.IOException;
import java.util.Properties;
import javax.servlet.ServletContextEvent;
import javax.servlet.ServletContextListener;
import javax.servlet.annotation.WebListener;
import redis.clients.jedis.JedisPool;
import redis.clients.jedis.JedisPoolConfig;

@WebListener
public class AppServletContextListener implements ServletContextListener {

  private Properties config = new Properties();

  private JedisPool createJedisPool() throws IOException {
    String host;
    Integer port;
    config.load(
        Thread.currentThread()
            .getContextClassLoader()
            .getResourceAsStream("application.properties"));
    host = config.getProperty("redis.host");
    port = Integer.valueOf(config.getProperty("redis.port", "6379"));

    JedisPoolConfig poolConfig = new JedisPoolConfig();
    // Default : 8, consider how many concurrent connections into Redis you will need under load
    poolConfig.setMaxTotal(128);

    return new JedisPool(poolConfig, host, port);
  }

  @Override
  public void contextDestroyed(ServletContextEvent event) {
    JedisPool jedisPool = (JedisPool) event.getServletContext().getAttribute("jedisPool");
    if (jedisPool != null) {
      jedisPool.destroy();
      event.getServletContext().setAttribute("jedisPool", null);
    }
  }

  // Run this before web application is started
  @Override
  public void contextInitialized(ServletContextEvent event) {
    JedisPool jedisPool = (JedisPool) event.getServletContext().getAttribute("jedisPool");
    if (jedisPool == null) {
      try {
        jedisPool = createJedisPool();
        event.getServletContext().setAttribute("jedisPool", jedisPool);
      } catch (IOException e) {
        // handle exception
      }
    }
  }
}

VisitCounterServlet 类是一个网络 Servlet，用于增加 Redis 计数器。


package com.example.redis;

import java.io.IOException;
import java.net.SocketException;
import javax.servlet.annotation.WebServlet;
import javax.servlet.http.HttpServlet;
import javax.servlet.http.HttpServletRequest;
import javax.servlet.http.HttpServletResponse;
import redis.clients.jedis.Jedis;
import redis.clients.jedis.JedisPool;

@WebServlet(name = "Track visits", value = "")
public class VisitCounterServlet extends HttpServlet {

  @Override
  public void doGet(HttpServletRequest req, HttpServletResponse resp) throws IOException {
    try {
      JedisPool jedisPool = (JedisPool) req.getServletContext().getAttribute("jedisPool");

      if (jedisPool == null) {
        throw new SocketException("Error connecting to Jedis pool");
      }
      Long visits;

      try (Jedis jedis = jedisPool.getResource()) {
        visits = jedis.incr("visits");
      }

      resp.setStatus(HttpServletResponse.SC_OK);
      resp.getWriter().println("Visitor counter: " + String.valueOf(visits));
    } catch (Exception e) {
      resp.sendError(HttpServletResponse.SC_INTERNAL_SERVER_ERROR, e.getMessage());
    }
  }
}

Node.js

此应用使用 redis 模块。

{
  "name": "memorystore-redis",
  "description": "An example of using Memorystore(Redis) with Node.js",
  "version": "0.0.1",
  "private": true,
  "license": "Apache Version 2.0",
  "author": "Google Inc.",
  "engines": {
    "node": ">=16.0.0"
  },
  "dependencies": {
    "redis": "^4.0.0"
  }
}


'use strict';
const http = require('http');
const redis = require('redis');

const REDISHOST = process.env.REDISHOST || 'localhost';
const REDISPORT = process.env.REDISPORT || 6379;

const client = redis.createClient(REDISPORT, REDISHOST);
client.on('error', err => console.error('ERR:REDIS:', err));

// create a server
http
  .createServer((req, res) => {
    // increment the visit counter
    client.incr('visits', (err, reply) => {
      if (err) {
        console.log(err);
        res.status(500).send(err.message);
        return;
      }
      res.writeHead(200, {'Content-Type': 'text/plain'});
      res.end(`Visitor number: ${reply}\n`);
    });
  })
  .listen(8080);

Python

此应用使用 Flask 进行网络服务，并使用 redis-py 软件包与 Redis 实例进行通信。

Flask==3.0.3
gunicorn==23.0.0
redis==6.0.0
Werkzeug==3.0.3
 
import logging
import os

from flask import Flask
import redis

app = Flask(__name__)

redis_host = os.environ.get("REDISHOST", "localhost")
redis_port = int(os.environ.get("REDISPORT", 6379))
redis_client = redis.StrictRedis(host=redis_host, port=redis_port)


@app.route("/")
def index():
    value = redis_client.incr("counter", 1)
    return f"Visitor number: {value}"


@app.errorhandler(500)
def server_error(e):
    logging.exception("An error occurred during a request.")
    return (
        """
    An internal error occurred: <pre>{}</pre>
    See logs for full stacktrace.
    """.format(
            e
        ),
        500,
    )


if __name__ == "__main__":
    # This is used when running locally. Gunicorn is used to run the
    # application on Google App Engine and Cloud Run.
    # See entrypoint in app.yaml or Dockerfile.
    app.run(host="127.0.0.1", port=8080, debug=True)

构建容器映像

构建容器映像并将其推送到 Container Registry：

    gcloud artifacts repositories create --location REPO_REGION --repository-format=docker REPO_ID
    cp gke_deployment/Dockerfile .
    docker build -t REPO_REGION-docker.pkg.dev/PROJECT_ID/REPO_ID/visit-counter:v1 .
    gcloud docker -- push REPO_REGION-docker.pkg.dev/PROJECT_ID/REPO_ID/visit-counter:v1

将应用部署到 Google Kubernetes Engine

为避免对 Redis 实例 IP 进行硬编码，您可以创建 redishost ConfigMap：

    export REDISHOST_IP=REDISHOST_IP
    kubectl create configmap redishost --from-literal=REDISHOST=${REDISHOST_IP}

使用以下命令验证配置：

kubectl get configmaps redishost -o yaml

更新 gke_deployment/visit-counter.yaml，将 <REPO_REGION>、<PROJECT_ID> 和 <REPO_ID> 替换为在构建容器映像中创建的容器映像中的值。此文件包含部署和服务的配置。

apiVersion: apps/v1
kind: Deployment
metadata:
  name: visit-counter
  labels:
    app: visit-counter
spec:
  replicas: 1
  selector:
    matchLabels:
      app: visit-counter
  template:
    metadata:
      labels:
        app: visit-counter
    spec:
      containers:
      - name: visit-counter
        image: "<REPO_REGION>-docker.pkg.dev/<PROJECT_ID>/<REPO_ID>/visit-counter:v1"
        env:
        - name: REDISHOST
          valueFrom:
            configMapKeyRef:
              name: redishost
              key: REDISHOST
        ports:
        - name: http
          containerPort: 8080
---
apiVersion: v1
kind: Service
metadata:
  name: visit-counter
spec:
  type: LoadBalancer
  selector:
    app: visit-counter
  ports:
  - port: 80
    targetPort: 8080
    protocol: TCP

将配置应用到您的集群：

kubectl apply -f gke_deployment/visit-counter.yaml

通过运行以下命令，确定此示例应用的外部 IP 地址：

kubectl get service visit-counter

使用浏览器验证您的应用是否托管在外部 IP 地址处，或者使用 curl 或浏览器发送 GET 请求：

curl http://EXTERNAL_IP

删除 Redis 实例的 IP 表条目

如果您按照本演示中准备 GKE 集群部分的步骤 3 进行了操作，则您已将 Redis 实例的预留 IP 范围安装到了 GKE 实例的 IP 表中。如果要从 GKE 实例的 IP 表中移除此 Redis IP 范围条目，请从 k8s-custom-iptables/ 目录运行以下命令：

./uninstall.sh