JStorm环境搭建

开始JStorm学习之前需要搭建集群环境，这里演示搭建单机JStorm环境，仅供学习使用，生产环境部署大同小异，但建议参考JStorm社区及相关说明文档。
一、前提
JStorm核心代码均用Java实现，所以依赖Java Runtime，另外，JStorm有脚本采用Python实现，所以还需要Python的支持。
1、JAVA环境

2、Python环境

这里选择Java版本1.6.0_35及Python版本2.6.5，如果默认没有安装可以参考相关文档（www.java.com和www.python.org）。
二、版本选择
zeromq-3.2.4
zookeeper-3.4.5
jstorm-0.7.1
三、JStorm环境搭建
与Storm一样，JStorm的底层消息通信机制依赖zeromq/jzmq，另外，JStorm通过zookeeper实现数据共享和协调服务。
1、安装zeromq
wget http://download.zeromq.org/zeromq-3.2.4.tar.gz
tar zxf zeromq-3.2.4.tar.gz
cd zeromq-3.2.4
./configure
make
sudo make install
sudo ldconfig
2、安装jzmq
wget https://github.com/zeromq/jzmq/tarball/master -O jzmq.tar.gz
tar zxf jzmq.tar.gz
cd jzmq
./autogen.sh
./configure
make
make install
3、安装zookeeper
wget http://apache.dataguru.cn/zookeeper/zookeeper-3.4.5/zookeeper-3.4.5.tar.gz
tar zxf zookeeper-3.4.5.tar.gz
cd zookeeper-3.4.5
./bin/zkServer.sh start
./bin/zkServer.sh stop
4、安装jstorm
wget http://42.121.19.155/jstorm/jstorm-0.7.1.zip
unzip jstorm-0.7.1.zip
编辑配置文件conf/storm.yaml
storm.zookeeper.servers:
- “localhost”
nimbus.host: “localhost”
storm.zookeeper.root: “/jstorm”
storm.local.dir: “/tmp/jstorm”
drpc.servers:
- “localhost”
如果是开发环境本地内存不足情况时启动nimbus可能会抛出异常：
Error occurred during initialization of VM
Could not reserve enough space for object heap
只需要在conf/storm.yaml里配置：
nimbus.childopts: “-Xmx256m”
supervisor.childopts: “-Xmx256m”
worker.childopts: “-Xmx128m”
其中大小可根据实际情况配置
5、UI
前提：tomcat 7.0 或以上版本；
将jstorm-ui-0.7.1.war复制到tomcat的webapps目录下；
6、启动JStorm
启动zookeeper：进入zookeeper目录，执行bin/zkServer.sh start
启动Nimbus：进入JStorm目录，执行bin/jstorm nimbus
启动Supervisor：进入JStorm目录，执行bin/jstorm supervisor
启动Tomcat：进入Tomcat目录，执行bin/startup.sh
四、JStorm HelloWorld
1、编写源码
这个例子取自：github
HelloWorldTopology.java

package storm.cookbook;
import backtype.storm.Config;
import backtype.storm.StormSubmitter;
import backtype.storm.generated.AlreadyAliveException;
import backtype.storm.generated.InvalidTopologyException;
import backtype.storm.topology.TopologyBuilder;
import backtype.storm.utils.Utils;
/**
 * Author: ashrith
 * Date: 8/26/13
 * Time: 12:03 PM
 * Desc: setup the topology and submit it to either a local of remote Storm cluster depending on the arguments
 *       passed to the main method.
 */
public class HelloWorldTopology {
    /*
     * main class in which to define the topology and a LocalCluster object (enables you to test and debug the
     * topology locally). In conjunction with the Config object, LocalCluster allows you to try out different
     * cluster configurations.
     *
     * Create a topology using 'TopologyBuilder' (which will tell storm how the nodes area arranged and how they
     * exchange data)
     * The spout and the bolts are connected using 'ShuffleGroupings'
     *
     * Create a 'Config' object containing the topology configuration, which is merged with the cluster configuration
     * at runtime and sent to all nodes with the prepare method
     *
     * Create and run the topology using 'createTopology' and 'submitTopology'
     */
    public static void main(String[] args) throws AlreadyAliveException, InvalidTopologyException {
        TopologyBuilder builder = new TopologyBuilder();
        builder.setSpout("randomHelloWorld", new HelloWorldSpout(), 10);
        builder.setBolt("HelloWorldBolt", new HelloWorldBolt(), 1).shuffleGrouping("randomHelloWorld");
        Config conf = new Config();
        conf.put(Config.NIMBUS_HOST, "localhost");
        conf.put(Config.NIMBUS_THRIFT_PORT, 6627);
        conf.setDebug(true);
        if(args!=null && args.length > 0) {
            conf.setNumWorkers(3);
            StormSubmitter.submitTopology(args[0], conf, builder.createTopology());
        } else {
        }
    }
}

HelloWorldSpout.java

package storm.cookbook;
import backtype.storm.spout.SpoutOutputCollector;
import backtype.storm.task.TopologyContext;
import backtype.storm.topology.OutputFieldsDeclarer;
import backtype.storm.topology.base.BaseRichSpout;
import backtype.storm.tuple.Fields;
import backtype.storm.tuple.Values;
import java.util.Map;
import java.util.Random;
/**
 * Author: ashrith
 * Date: 8/21/13
 * Time: 8:33 PM
 * Desc: spout essentially emits a stream containing 1 of 2 sentences 'Other Random Word' or 'Hello World' based on
 *       random probability. It works by generating a random number upon construction and then generating subsequent
 *       random numbers to test against the original member variable's value. When it matches "Hello World" is emitted,
 *       during the remaining executions the other sentence is emitted.
 */
public class HelloWorldSpout extends BaseRichSpout{
    private SpoutOutputCollector collector;
    private int referenceRandom;
    private static final int MAX_RANDOM = 10;
    public HelloWorldSpout() {
        final Random rand = new Random();
        referenceRandom = rand.nextInt(MAX_RANDOM);
    }
    /*
     * declareOutputFields() => you need to tell the Storm cluster which fields this Spout emits within the
     *  declareOutputFields method.
     */
    @Override
    public void declareOutputFields(OutputFieldsDeclarer declarer) {
        declarer.declare(new Fields("sentence"));
    }
    /*
     * open() => The first method called in any spout is 'open'
     *           TopologyContext => contains all our topology data
     *           SpoutOutputCollector => enables us to emit the data that will be processed by the bolts
     *           conf => created in the topology definition
     */
    @Override
    public void open(Map conf, TopologyContext topologyContext, SpoutOutputCollector collector) {
        this.collector = collector;
    }
    /*
     * nextTuple() => Storm cluster will repeatedly call the nextTuple method which will do all the work of the spout.
     *  nextTuple() must release the control of the thread when there is no work to do so that the other methods have
     *  a chance to be called.
     */
    @Override
    public void nextTuple() {
        final Random rand = new Random();
        int instanceRandom = rand.nextInt(MAX_RANDOM);
        if(instanceRandom == referenceRandom){
            collector.emit(new Values("Hello World"));
        } else {
            collector.emit(new Values("Other Random Word"));
        }
    }
}

HelloWorldBolt.java

package storm.cookbook;
import backtype.storm.task.OutputCollector;
import backtype.storm.task.TopologyContext;
import backtype.storm.topology.OutputFieldsDeclarer;
import backtype.storm.topology.base.BaseRichBolt;
import backtype.storm.tuple.Tuple;
import java.util.Map;
/**
 * Author: ashrith
 * Date: 8/26/13
 * Time: 11:48 AM
 * Desc: This bolt will consume the produced Tuples from HelloWorldSpout and implement the required counting logic
 */
public class HelloWorldBolt extends BaseRichBolt {
    private int myCount = 0;
    /*
     * prepare() => on create
     */
    @Override
    public void prepare(Map map, TopologyContext topologyContext, OutputCollector outputCollector) {
    }
    /*
     * execute() => most important method in the bolt is execute(Tuple input), which is called once per tuple received
     *  the bolt may emit several tuples for each tuple received
     */
    @Override
    public void execute(Tuple tuple) {
        String test = tuple.getStringByField("sentence");
        if(test == "Hello World"){
            myCount++;
            System.out.println("Found a Hello World! My Count is now: " + Integer.toString(myCount));
        }
    }
    /*
     * declareOutputFields => This bolt emits nothing hence no body for declareOutputFields()
     */
    @Override
    public void declareOutputFields(OutputFieldsDeclarer outputFieldsDeclarer) {
    }
}

2、提交Topology
上述源码编译打包Helloworld.jar后提交到jstorm集群:
bin/jstorm jar Helloworld.jar storm.cookbook.HelloWorldTopology HelloWorld
其中参数[HelloWorld]为TopologyName
3.查看Topology运行状况
通过ui等途径可以查看Topology的执行情况。
五、结语
本节简单介绍了JStorm单机环境的搭建，用供初学者搭建单机JStorm，并能够编写HelloWolrd，生产环境集群搭建仅做参考，详细配置建议查询相关文档。
六、参考文档
[1]https://github.com/alibaba/jstorm/wiki
[2]https://github.com/nathanmarz/storm/wiki

http://hexiaoqiao.sinaapp.com/2014/06/09/jstorm%E7%8E%AF%E5%A2%83%E6%90%AD%E5%BB%BA/