Zookeeper和CuratorFramework实践系列之：配置管理

看过Zookeeper相关文档后都知道它可以实现分布式集群的配置管理，本文以一个简单的实例来演示它是如何实现的并工作的。

情景需要，简单理解为下图：

一个web集群，需要通过zk来控制集群的日志输出级别，比如管理员需要在生产环境下查看一下DEBUG日志，他可以临时将集群的日志输出级别改为DEBUG，获取他想要的信息后还要将级别调回到INFO或者ERROR级别。

今天的主角是Curator-Framework，它的存在使得我们操作ZK变得简单有趣起来！

环境，框架，工具

Zookeeper集群
Zookeeper API
Curator-Framework
Spring-MVC
Logback
Maven + IDEA

POM

依赖关系和jetty插件

<dependencies>
    <dependency>
        <groupId>org.springframework</groupId>
        <artifactId>spring-context</artifactId>
        <version>3.2.3.RELEASE</version>
    </dependency>
    <dependency>
        <groupId>org.springframework</groupId>
        <artifactId>spring-web</artifactId>
        <version>3.2.3.RELEASE</version>
    </dependency>
    <dependency>
        <groupId>org.springframework</groupId>
        <artifactId>spring-webmvc</artifactId>
        <version>3.2.3.RELEASE</version>
    </dependency>
    <dependency>
        <groupId>org.apache.zookeeper</groupId>
        <artifactId>zookeeper</artifactId>
        <version>3.4.5</version>
        <exclusions>
            <exclusion>
                <groupId>log4j</groupId>
                <artifactId>log4j</artifactId>
            </exclusion>
            <exclusion>
                <groupId>org.slf4j</groupId>
                <artifactId>slf4j-log4j12</artifactId>
            </exclusion>
        </exclusions>
    </dependency>
    <dependency>
        <groupId>org.apache.curator</groupId>
        <artifactId>curator-framework</artifactId>
        <version>2.0.1-incubating</version>
    </dependency>
    <dependency>
        <groupId>org.apache.curator</groupId>
        <artifactId>curator-recipes</artifactId>
        <version>2.0.1-incubating</version>
    </dependency>
    <dependency>
        <groupId>ch.qos.logback</groupId>
        <artifactId>logback-classic</artifactId>
        <version>1.0.13</version>
    </dependency>
    <dependency>
        <version>1.7.5</version>
        <groupId>org.slf4j</groupId>
        <artifactId>slf4j-api</artifactId>
    </dependency>
    <dependency>
        <groupId>org.testng</groupId>
        <artifactId>testng</artifactId>
        <version>6.8.5</version>
        <scope>test</scope>
    </dependency>
</dependencies>
<build>
    <finalName>zookeeper-configuration</finalName>
    <plugins>
        <plugin>
            <groupId>org.mortbay.jetty</groupId>
            <artifactId>jetty-maven-plugin</artifactId>
            <version>8.1.7.v20120910</version>
            <configuration>
                <stopKey>pbase</stopKey>
                <stopPort>8080</stopPort>
            </configuration>
        </plugin>
    </plugins>
</build>

WEB.XML

这个不用解释了，地球人都知道

<web-app>
  <display-name>Archetype Created Web Application</display-name>
    <context-param>
        <param-name>contextConfigLocation</param-name>
        <param-value>
            classpath*:/applicationContext.xml
        </param-value>
    </context-param>
    <listener>
        <listener-class>org.springframework.web.context.ContextLoaderListener</listener-class>
    </listener>
    <servlet>
        <servlet-name>springServlet</servlet-name>
        <servlet-class>org.springframework.web.servlet.DispatcherServlet</servlet-class>
        <init-param>
        <param-name>contextConfigLocation</param-name>
        <param-value>classpath*:/spring-mvc.xml</param-value>
    </init-param>
        <load-on-startup>1</load-on-startup>
    </servlet>
    <servlet-mapping>
        <servlet-name>springServlet</servlet-name>
        <url-pattern>/</url-pattern>
    </servlet-mapping>
</web-app>

Spring 配置

applicationContext.xml

<context:component-scan base-package="cn.bg">
    <context:exclude-filter type="annotation" expression="org.springframework.stereotype.Controller"/>
</context:component-scan>
<!-- Curator的FactoryBean，Spring启动时创建Curator实例。 -->
<bean id="zookeeperFactoryBean" class="cn.bg.zk.core.ZookeeperFactoryBean" lazy-init="false">
    <property name="zkConnectionString" value="hadoopmaster:2181"/>
    <!-- 设置zookeeper的事件监听者，本例是一个logback日志级别znode监听器 -->
    <property name="listeners">
        <list>
            <bean class="cn.bg.zk.configuration.LogbackLevelListener">
                <constructor-arg value="/zk_test/logbacklevel"/>
            </bean>
        </list>
    </property>
</bean>

spring-mvc.xml

<context:component-scan base-package="cn.bg">
    <context:include-filter type="annotation" expression="org.springframework.stereotype.Controller"/>
</context:component-scan>
<mvc:annotation-driven></mvc:annotation-driven>

Zookeeper配置管理实现相关类

ZookeeperFactoryBean.java

在Spring Context加载过程中创建Zookeeper链接对像并设置触发监听，通过Curator.

package cn.bg.zk.core;

public class ZookeeperFactoryBean implements FactoryBean<CuratorFramework>, InitializingBean, DisposableBean {

    private Logger logger = LoggerFactory.getLogger(this.getClass());
    private CuratorFramework zkClient;

    //设置Zookeeper启动后需要调用的监听或者，或者需要做的初始化工作。
    public void setListeners(List<IZKListener> listeners) {
        this.listeners = listeners;
    }

    private List<IZKListener> listeners;

    //设置ZK链接串
    public void setZkConnectionString(String zkConnectionString) {
        this.zkConnectionString = zkConnectionString;
    }

    private String zkConnectionString;

    @Override
    public CuratorFramework getObject() {
        return zkClient;
    }

    @Override
    public Class<?> getObjectType() {
        return CuratorFramework.class;
    }

    @Override
    public boolean isSingleton() {
        return true;
    }

    @Override
    public void destroy() throws Exception {
        zkClient.close();
    }

    //创建ZK链接
    @Override
    public void afterPropertiesSet(){
        //1000 是重试间隔时间基数，3 是重试次数
        RetryPolicy retryPolicy = new ExponentialBackoffRetry(1000, 3);
        zkClient = createWithOptions(zkConnectionString, retryPolicy, 2000, 10000);
        registerListeners(zkClient);
        zkClient.start();
    }


    /**
     * 通过自定义参数创建
     */
    public CuratorFramework  createWithOptions(String connectionString, RetryPolicy retryPolicy, int connectionTimeoutMs, int sessionTimeoutMs)
    {
        return CuratorFrameworkFactory.builder()
                .connectString(connectionString)
                .retryPolicy(retryPolicy)
                .connectionTimeoutMs(connectionTimeoutMs)
                .sessionTimeoutMs(sessionTimeoutMs)
                .build();
    }        

    //注册需要监听的监听者对像. 
    private void registerListeners(CuratorFramework client){
        client.getConnectionStateListenable().addListener(new ConnectionStateListener() {
            @Override
            public void stateChanged(CuratorFramework client, ConnectionState newState) {
                logger.info("CuratorFramework state changed: {}", newState);
                if(newState == ConnectionState.CONNECTED || newState == ConnectionState.RECONNECTED){
                    for(IZKListener listener : listeners){
                        listener.executor(client);
                        logger.info("Listener {} executed!", listener.getClass().getName());
                    }
                }
            }
        });

        client.getUnhandledErrorListenable().addListener(new UnhandledErrorListener() {
            @Override
            public void unhandledError(String message, Throwable e) {
                logger.info("CuratorFramework unhandledError: {}", message);
            }
        });
    }
}

监听事件接口

所有需要在ZK客户端链接成功后需要做的事件，需要实现这个接口，由上面的ZookeeperFactoryBean统一调度。

package cn.bg.zk.core;

import org.apache.curator.framework.CuratorFramework;

public interface IZKListener {
    void executor(CuratorFramework client);
}

Logback监听实现

这里主要用到Curator的NodeCache类，它的主要功能是用来监听znode本身的变化，并可以获取当前值，而且会自动重复监听，简化了原生API开发的繁琐过程。

package cn.bg.zk.configuration;

public class LogbackLevelListener implements IZKListener {

    //获取logback实例
    Logger log = (Logger) LoggerFactory.getLogger(this.getClass());

    private String path;

    //Logback日志级别ZNode
    public LogbackLevelListener(String path) {
        this.path = path;
    }

    @Override
    public void executor(CuratorFramework client) {

        //使用Curator的NodeCache来做ZNode的监听，不用我们自己实现重复监听
        final NodeCache cache = new NodeCache(client, path);
        cache.getListenable().addListener(new NodeCacheListener() {
            @Override
            public void nodeChanged() throws Exception {

                byte[] data = cache.getCurrentData().getData();

                //设置日志级别
                if (data != null) {
                    String level = new String(data);
                    Logger logger = (Logger) LoggerFactory.getLogger("root");
                    Level newLevel = Level.fromLocationAwareLoggerInteger(Integer.parseInt(level));
                    logger.setLevel(newLevel);
                    System.out.println("Setting logback new level to :" + newLevel.levelStr);
                }
            }
        });
        try {
            cache.start(true);
        } catch (Exception e) {
            log.error("Start NodeCache error for path: {}, error info: {}", path, e.getMessage());
        }
    }
}

通过WEB端查看当前日志级别

写一个controller来读取logback的日志级别

MainController.java

package cn.bg.controller;

@Controller
public class MainController {

    @RequestMapping("/")
    @ResponseBody
    public String logbackLevel() throws Exception {
        Logger logger = (Logger) LoggerFactory.getLogger("root");
        String levelStr = logger.getLevel().levelStr;
        return levelStr;
    }

}

运行

在ZK集群端启动zkCli创建/zk_test/logbacklevel的znode，设置值为10，或20在控制台来查看logback的日志输出情况，logback日志级别数据表示如下：

TRACE_INT = 0
DEBUG_INT = 10
INFO_INT = 20
WARN_INT = 30
ERROR_INT = 40

到此就实现了一个简单的ZK配置管理情境，有了Curator-Framework后一切变的简单起来，我们主要精力只要解决业务相关的的需要，而ZK相关的实现由Curator来解决。
下一篇介绍Curator分布式队列实现。