Netty心跳和重连

来源:互联网 发布:摩天手无线鼠标 知乎 编辑:程序博客网 时间:2024/05/16 10:44
转载自:http://blog.csdn.net/z69183787/article/details/52625095

Netty应用心跳和重连的整个过程:
     1)客户端连接服务端;
     2)在客户端的的ChannelPipeline中加入一个比较特殊的IdleStateHandler,设置一下客户端的写空闲时间,例如5s;
     3)当客户端的所有ChannelHandler中4s内没有write事件,则会触发userEventTriggered方法;
     4)在客户端的userEventTriggered中对应的触发事件下发送一个心跳包给服务端,检测服务端是否还存活,防止服务端已经宕机,客户端还不知道;
     5)同样,服务端要对心跳包做出响应,其实给客户端最好的回复就是“不回复”,这样可以降低服务端的压力,假如有10w个空闲Idle的连接,那么服务端光发送心跳回复,则也是费事的事情,那么怎么才能告诉客户端它还活着呢,其实很简单,因为5s服务端都会收到来自客户端的心跳信息,那么如果10秒内收不到,服务端可以认为客户端挂了,可以close链路;
     6)假如服务端因为什么因素导致宕机的话,就会关闭所有的链路链接,所以作为客户端要做的事情就是断线重连。

       IdleStateHandler是Netty提供了对心跳机制的支持,心跳可以检测远程端是否存活,或者活跃。
    public IdleStateHandler(            long readerIdleTime, long writerIdleTime, long allIdleTime,            TimeUnit unit) {
       构造函数中readerIdleTime为读超时时间,writerIdleTime为写超时时间,allIdleTime为所有类型的超时时间。这个类也是一个ChannelHandler,也需要被载入到ChannelPipeline中,加入在服务器端的ChannelInitializer中的代码如下:
            bootstrap.childHandler(new ChannelInitializer<SocketChannel>() {                @Override                public void initChannel(SocketChannel ch) throws Exception {                    ChannelPipeline pipeline = ch.pipeline();                    pipeline.addLast(new IdleStateHandler(5, 0, 0, TimeUnit.SECONDS));                    // 字符串 解码 和 编码                    pipeline.addLast(new StringDecoder());//解码                    pipeline.addLast(new StringEncoder());//编码                    // 自定义Handler                    pipeline.addLast(new HelloServerHandler());                }            });
       这里表示服务器端会每隔5秒来检查一下channelRead方法被调用的情况,如果在5秒内该链上的channelRead方法都没有被触发,就会调用userEventTriggered方法。
       IdleStateHandler内部有定时任务,会根据设置的超时参数的类型和值,检测channelRead和write方法多久没有被调用了,如果这个时间超过了设置的值,那么就会触发对应的事件,read触发read,write触发write,all触发all。如果超时了,则会调用userEventTriggered方法,且会告诉超时的类型;如果没有超时,则会循环定时检测,除非将IdleStateHandler移除Pipeline。
       只有链路建立之后,即channelActive事件触发之后,才会initialize内部的定时任务。

       要实现可靠的心跳重连,需要对ChannelPipeline中的ChannelHandlers的维护,首次连接和重连都需要对ChannelHandlers进行管理;对重连对象的管理,也就是bootstrap对象的管理;还有重连机制。接下来将会以简单的例子来说明。
首先先定义一个接口ChannelHandlerHolder,用来保管ChannelPipeline中的Handlers。
/** * * 客户端的ChannelHandler集合,由子类实现,这样做的好处: * 继承这个接口的所有子类可以很方便地获取ChannelPipeline中的Handlers * 获取到handlers之后方便ChannelPipeline中的handler的初始化和在重连的时候也能很方便 * 地获取所有的handlers */public interface ChannelHandlerHolder {    ChannelHandler[] handlers();}
       服务端需要一个AcceptorIdleStateTrigger,继承ChannelInboundHandlerAdapter,重写userEventTriggered方法,因为客户端是write,那么服务端自然是read,设置的状态就是IdleState.READER_IDLE,代码如下:
@ChannelHandler.Sharablepublic class AcceptorIdleStateTrigger extends ChannelInboundHandlerAdapter {    @Override    public void userEventTriggered(ChannelHandlerContext ctx, Object evt) throws Exception {        if (evt instanceof IdleStateEvent) {            IdleState state = ((IdleStateEvent) evt).state();            if (state == IdleState.READER_IDLE) {                throw new Exception("Idle exception");            }        } else {            super.userEventTriggered(ctx, evt);        }    }}
       然后自定义一个简单的业务处理的Handler:
public class HeartBeatServerHandler extends ChannelInboundHandlerAdapter {    @Override    public void channelRead(ChannelHandlerContext ctx, Object msg) throws Exception {        System.out.println("server channelRead..");        System.out.println(ctx.channel().remoteAddress() + "->Server :" + msg.toString());    }    @Override    public void exceptionCaught(ChannelHandlerContext ctx, Throwable cause) throws Exception {        cause.printStackTrace();        ctx.close();    }}
       再来编写服务端的ServerBootstrap代码:
public class HeartBeatServer {    private final AcceptorIdleStateTrigger idleStateTrigger = new AcceptorIdleStateTrigger();    private int port;    public HeartBeatServer(int port) {        this.port = port;    }    public void start() {        EventLoopGroup bossGroup = new NioEventLoopGroup(1);        EventLoopGroup workerGroup = new NioEventLoopGroup();        try {            ServerBootstrap sbs = new ServerBootstrap().group(bossGroup, workerGroup)                    .channel(NioServerSocketChannel.class).handler(new LoggingHandler(LogLevel.INFO))                    .localAddress(new InetSocketAddress(port)).childHandler(new ChannelInitializer<SocketChannel>() {                        protected void initChannel(SocketChannel ch) throws Exception {                            ch.pipeline().addLast(new IdleStateHandler(5, 0, 0, TimeUnit.SECONDS));                            ch.pipeline().addLast(idleStateTrigger);                            ch.pipeline().addLast("decoder", new StringDecoder());                            ch.pipeline().addLast("encoder", new StringEncoder());                            ch.pipeline().addLast(new HeartBeatServerHandler());                        };                    }).option(ChannelOption.SO_BACKLOG, 128).childOption(ChannelOption.SO_KEEPALIVE, true);            // 绑定端口,开始接收进来的连接            ChannelFuture future = sbs.bind(port).sync();            System.out.println("Server start listen at " + port);            future.channel().closeFuture().sync();        } catch (Exception e) {            bossGroup.shutdownGracefully();            workerGroup.shutdownGracefully();        }    }    public static void main(String[] args) throws Exception {        int port;        if (args.length > 0) {            port = Integer.parseInt(args[0]);        } else {            port = 8080;        }        new HeartBeatServer(port).start();    }}
       服务端就完成了,下面是客户端。首先是ConnectorIdleStateTrigger:
@ChannelHandler.Sharablepublic class ConnectorIdleStateTrigger extends ChannelInboundHandlerAdapter {    private static final ByteBuf HEARTBEAT_SEQUENCE = Unpooled.unreleasableBuffer(Unpooled.copiedBuffer("Heartbeat",            CharsetUtil.UTF_8));    @Override    public void userEventTriggered(ChannelHandlerContext ctx, Object evt) throws Exception {        if (evt instanceof IdleStateEvent) {            IdleState state = ((IdleStateEvent) evt).state();            if (state == IdleState.WRITER_IDLE) {                // write heartbeat to server                ctx.writeAndFlush(HEARTBEAT_SEQUENCE.duplicate());            }        } else {            super.userEventTriggered(ctx, evt);        }    }}
       简单的自定义的Handler:
@ChannelHandler.Sharablepublic class HeartBeatClientHandler extends ChannelInboundHandlerAdapter {    @Override    public void channelActive(ChannelHandlerContext ctx) throws Exception {        System.out.println("激活时间是:" + new Date());        System.out.println("HeartBeatClientHandler channelActive");        ctx.fireChannelActive();    }    @Override    public void channelInactive(ChannelHandlerContext ctx) throws Exception {        System.out.println("停止时间是:" + new Date());        System.out.println("HeartBeatClientHandler channelInactive");    }    @Override    public void channelRead(ChannelHandlerContext ctx, Object msg) throws Exception {        String message = (String) msg;        System.out.println(message);        if (message.equals("Heartbeat")) {            ctx.write("has read message from server");            ctx.flush();        }        ReferenceCountUtil.release(msg);    }}
       接下来就是重点,客户端需要定义一个类,这个类可以去观察链路是否断了,如果断了,进行循环的断线重连操作,ConnectionWatchdog,顾名思义,链路检测狗,先看完整代码:
@ChannelHandler.Sharablepublic abstract class ConnectionWatchdog extends ChannelInboundHandlerAdapter implements TimerTask, ChannelHandlerHolder {    private final Bootstrap bootstrap;    private final Timer timer;    private final int port;    private final String host;    private volatile boolean reconnect = true;    private int attempts;    public ConnectionWatchdog(Bootstrap bootstrap, Timer timer, int port, String host, boolean reconnect) {        this.bootstrap = bootstrap;        this.timer = timer;        this.port = port;        this.host = host;        this.reconnect = reconnect;    }    /**     * channel链路每次active的时候,将其连接的次数重新置0     */    @Override    public void channelActive(ChannelHandlerContext ctx) throws Exception {        System.out.println("当前链路已经激活了,重连尝试次数重新置为0");        attempts = 0;        ctx.fireChannelActive();    }    @Override    public void channelInactive(ChannelHandlerContext ctx) throws Exception {        System.out.println("链接关闭");        if (reconnect) {            System.out.println("链接关闭,将进行重连");            if (attempts < 12) {                attempts++;            }           //重连的间隔时间会越来越长            int timeout = 2 << attempts;            timer.newTimeout(this, timeout, TimeUnit.MILLISECONDS);        }        ctx.fireChannelInactive();    }    @Override    public void run(Timeout timeout) throws Exception {        ChannelFuture future;        //bootstrap已经初始化好了,只需要将handler填入就可以了        synchronized (bootstrap) {            bootstrap.handler(new ChannelInitializer<Channel>() {                @Override                protected void initChannel(Channel ch) throws Exception {                    ch.pipeline().addLast(handlers());                }            });            future = bootstrap.connect(host, port);        }        //future对象        future.addListener(new ChannelFutureListener() {            public void operationComplete(ChannelFuture f) throws Exception {                boolean succeed = f.isSuccess();                //如果重连失败,则调用ChannelInactive方法,再次出发重连事件,一直尝试12次,如果失败则不再重连                if (!succeed) {                    System.out.println("重连失败");                    f.channel().pipeline().fireChannelInactive();                } else {                    System.out.println("重连成功");                }            }        });    }}
       ConnectionWatchdog继承了ChannelInboundHandlerAdapter,说明它也是Handler;
       实现了2个接口,TimeTask,ChannelHandlerHolder,TimeTask就要重写run方法,这应该是一个定时任务,这个定时任务做的事情应该是重连的工作;ChannelHandlerHolder的接口,这个接口维护的所有的Handlers,因为在重连的时候需要获取Handlers;
       bootstrap对象,重连的时候依旧需要这个对象;
       当链路断开的时候会触发channelInactive这个方法,也就说触发重连的导火索是从这边开始的。

       接下来就是客户端的Bootstrap:
public class HeartBeatsClient {    protected final HashedWheelTimer timer = new HashedWheelTimer();    private Bootstrap boot;    private final ConnectorIdleStateTrigger idleStateTrigger = new ConnectorIdleStateTrigger();    public void connect(int port, String host) throws Exception {        EventLoopGroup group = new NioEventLoopGroup();        boot = new Bootstrap();        boot.group(group)                .channel(NioSocketChannel.class)                .handler(new LoggingHandler(LogLevel.INFO));        final ConnectionWatchdog watchdog = new ConnectionWatchdog(boot, timer, port, host, true) {            public ChannelHandler[] handlers() {                return new ChannelHandler[]{                        this,                        new IdleStateHandler(0, 4, 0, TimeUnit.SECONDS),                        idleStateTrigger,                        new StringDecoder(),                        new StringEncoder(),                        new HeartBeatClientHandler()                };            }        };        ChannelFuture future;        try {            synchronized (boot) {//进行连接                boot.handler(new ChannelInitializer<Channel>() {                    //初始化channel                    @Override                    protected void initChannel(Channel ch) throws Exception {                        ch.pipeline().addLast(watchdog.handlers());                    }                });                future = boot.connect(host, port);            }            // 以下代码在synchronized同步块外面是安全的            future.sync();        } catch (Throwable t) {            throw new Exception("connects to  fails", t);        }    }    public static void main(String[] args) throws Exception {        int port = 8000;        new HeartBeatsClient().connect(port, "127.0.0.1");    }}