Storm学习记录(五、容错机制)

1.集群各节点挂掉的影响 

nimbus挂掉对集群影响较小

当某个supervisor 挂掉后,zookeeper会向nimbus返回信息,当发现nimbus也挂掉后,会要求nimbus重启。

当有新任务上传时,也会要求nimbus重启。

其余的时候nimbus是闲置状态。

supervisor挂掉后,zookeeper会在通知nimbus,nimbus会将分配到该supervisor的任务重新分配到其他supervisor,当挂掉的supervisor重启后,会向zookeeper询问其任务是否被分配,如果是,则杀死其存储的worker。

Worker挂掉时,Supervisor会重新启动这个进程。如果启动过程中仍然一直失败,并且无法向Nimbus发送心跳,Nimbus会将该Worker重新分配到其他服务器上。

2.消息的完整性

从Spout中发出的Tuple,以及基于他所产生Tuple(例如上个例子当中Spout发出的句子,以及句子当中单词的tuple等)

由这些消息就构成了一棵tuple树,当这棵tuple树发送完成,并且树当中每一条消息都被正确处理,就表明spout发送消息被“完整处理”,即消息的完整性。发送tuple同时会传一个16位2进制的tid,通过比对tid的异或运算确认消息是否发送成功。

注意:ack无法保证数据不被重复计算,但是可以保证数据至少被正确处理一次。

3.样例 

public class MySpout implements IRichSpout{

	private static final long serialVersionUID = 1L;

	int index = 0;
	
	FileInputStream fis;
	InputStreamReader isr;
	BufferedReader br;			
	SpoutOutputCollector collector = null;
	String str = null;

	@Override
	public void nextTuple() {
		try {
			if ((str = this.br.readLine()) != null) {
				// 过滤动作
				index++;
				collector.emit(new Values(str), index);
//				collector.emit(new Values(str));
			}
		} catch (Exception e) {
		}
		
		
	}
	@Override
	public void close() {
		try {
			br.close();
			isr.close();
			fis.close();
		} catch (Exception e) {
			e.printStackTrace();
		}
	}
	@Override
	public void open(Map conf, TopologyContext context,
			SpoutOutputCollector collector) {
		try {
			this.collector = collector;
			this.fis = new FileInputStream("track.log");
			this.isr = new InputStreamReader(fis, "UTF-8");
			this.br = new BufferedReader(isr);
		} catch (Exception e) {
			e.printStackTrace();
		}
	}

	@Override
	public void declareOutputFields(OutputFieldsDeclarer declarer) {
		declarer.declare(new Fields("log"));
	}

	@Override
	public Map<String, Object> getComponentConfiguration() {
		return null;
	}
	
	@Override
	public void ack(Object msgId) {
		System.err.println(" [" + Thread.currentThread().getName() + "] "+ " spout ack:"+msgId.toString());
	}

	@Override
	public void activate() {
		
	}

	@Override
	public void deactivate() {
		
	}

	@Override
	public void fail(Object msgId) {
		System.err.println(" [" + Thread.currentThread().getName() + "] "+ " spout fail:"+msgId.toString());
	}

}
public class MyBolt implements IRichBolt {

	private static final long serialVersionUID = 1L;

	OutputCollector collector = null;
	@Override
	public void cleanup() {

	}
	int num = 0;
	String valueString = null;
	@Override
	public void execute(Tuple input) {
		try {
			valueString = input.getStringByField("log") ;
			
			if(valueString != null) {
				num ++ ;
				System.err.println(Thread.currentThread().getName()+"   lines  :"+num +"   session_id:"+valueString.split("\t")[1]);
			}
			collector.emit(input, new Values(valueString));
//			collector.emit(new Values(valueString));
			collector.ack(input);
			Thread.sleep(2000);
		} catch (Exception e) {
			collector.fail(input);
			e.printStackTrace();
		}
		
	}

	@Override
	public void prepare(Map stormConf, TopologyContext context,
			OutputCollector collector) {
		this.collector = collector ;
	}

	@Override
	public void declareOutputFields(OutputFieldsDeclarer declarer) {
		declarer.declare(new Fields("session_id")) ;
	}

	@Override
	public Map<String, Object> getComponentConfiguration() {
		return null;
	}

}
public class Main {

	public static void main(String[] args) {

		TopologyBuilder builder = new TopologyBuilder();

		builder.setSpout("spout", new MySpout(), 1);
		builder.setBolt("bolt", new MyBolt(), 2).shuffleGrouping("spout");
		
//		Map conf = new HashMap();
//		conf.put(Config.TOPOLOGY_WORKERS, 4);
		
		Config conf = new Config() ;
		conf.setDebug(true);
		conf.setMessageTimeoutSecs(conf, 100);
		conf.setNumAckers(4);
		
		if (args.length > 0) {
			try {
				StormSubmitter.submitTopology(args[0], conf, builder.createTopology());
			} catch (AlreadyAliveException e) {
				e.printStackTrace();
			} catch (InvalidTopologyException e) {
				e.printStackTrace();
			}
		}else {
			LocalCluster localCluster = new LocalCluster();
			localCluster.submitTopology("mytopology", conf, builder.createTopology());
		}
	}
}
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值