elasticsearch-hadoop提供ElasticSearch与Apache Storm的集成支持。从ElasticSearch读取的数据是以Storm里Tuple的形式进行操作处理。
依赖版本信息:
<dependency>
<groupId>org.apache.storm</groupId>
<artifactId>storm-core</artifactId>
<version>1.0.1</version>
</dependency>
<dependency>
<groupId>org.apache.storm</groupId>
<artifactId>storm-starter</artifactId>
<version>1.0.1</version>
</dependency>
<dependency>
<groupId>org.apache.storm</groupId>
<artifactId>storm-hdfs</artifactId>
<version>1.0.1</version>
</dependency>
<dependency>
<groupId>org.apache.storm</groupId>
<artifactId>storm-kafka</artifactId>
<version>1.0.1</version>
</dependency>
<dependency>
<groupId>org.apache.kafka</groupId>
<artifactId>kafka_2.10</artifactId>
<version>0.10.0.0</version>
</dependency>
<dependency>
<groupId>org.elasticsearch</groupId>
<artifactId>elasticsearch-hadoop</artifactId>
<version>2.3.2</version>
</dependency>
Strom的extlib目录下jar包
import java.util.Map;
import org.apache.storm.task.OutputCollector;
import org.apache.storm.task.TopologyContext;
import org.apache.storm.topology.OutputFieldsDeclarer;
import org.apache.storm.topology.base.BaseRichBolt;
import org.apache.storm.tuple.Fields;
import org.apache.storm.tuple.Tuple;
import org.apache.storm.tuple.Values;
public class HandleBolt extends BaseRichBolt {
private static final long serialVersionUID = 1L;
private OutputCollector collector = null;
@SuppressWarnings("rawtypes")
@Override
public void prepare(Map stormConf, TopologyContext context,
OutputCollector collector) {
this.collector = collector;
}
@Override
public void execute(Tuple input) {
String name = "NA";
if (input.contains("name")) {
name = input.getStringByField("name");
}
String phone = "NA";
if (input.contains("phone")) {
phone = input.getStringByField("phone");
}
String rcall = "NA";
if (input.contains("rcall")) {
rcall = input.getStringByField("rcall");
rcall = null == rcall || "null".equals(rcall) ? "NA" : rcall;
}
String address = "NA";
if (input.contains("address")) {
address = input.getStringByField("address");
address = null == address || "null".equals(address) ? "NA" : address;
}
String email = "NA";
if (input.contains("email")) {
email = input.getStringByField("email");
email = null == email || "null".equals(email) ? "NA" : email;
}
String idCard = "NA";
if (input.contains("idCard")) {
idCard = input.getStringByField("idCard");
idCard = null == idCard || "null".equals(idCard) ? "NA" : idCard;
}
this.collector.emit(new Values(name, phone, rcall, address, email, idCard));
this.collector.ack(input);
}
@Override
public void declareOutputFields(OutputFieldsDeclarer declarer) {
declarer.declare(new Fields("name", "phone", "rcal", "address", "email", "idCard"));
}
}
import java.util.HashMap;
import java.util.Map;
import org.apache.storm.Config;
import org.apache.storm.LocalCluster;
import org.apache.storm.StormSubmitter;
import org.apache.storm.hdfs.bolt.HdfsBolt;
import org.apache.storm.hdfs.bolt.format.DefaultFileNameFormat;
import org.apache.storm.hdfs.bolt.format.DelimitedRecordFormat;
import org.apache.storm.hdfs.bolt.format.FileNameFormat;
import org.apache.storm.hdfs.bolt.format.RecordFormat;
import org.apache.storm.hdfs.bolt.rotation.FileRotationPolicy;
import org.apache.storm.hdfs.bolt.rotation.TimedRotationPolicy;
import org.apache.storm.hdfs.bolt.rotation.TimedRotationPolicy.TimeUnit;
import org.apache.storm.hdfs.bolt.sync.CountSyncPolicy;
import org.apache.storm.hdfs.bolt.sync.SyncPolicy;
import org.apache.storm.starter.bolt.PrinterBolt;
import org.apache.storm.topology.TopologyBuilder;
import org.apache.storm.utils.Utils;
public class ES2StormTopology {
private static final String TOPOLOGY_NAME = "es-storm-topology";
public static void main(String[] args) {
if (args.length != 1) {
System.exit(0);
}
boolean isCluster = Boolean.parseBoolean(args[0]);
TopologyBuilder builder = new TopologyBuilder();
String target = "operator/telecom";
String query = "?q=*";
Map<Object, Object> configuration = new HashMap<Object, Object>();
configuration.put("es.nodes", "192.168.10.20:9200");
configuration.put("es.read.field.include", "name,phone,rcall,email,idCard,zipCode,address");
configuration.put("es.storm.spout.fields", "name,phone,rcall,email,idCard,zipCode,address");
builder.setSpout("es-storm-spout", new ESSpout(target, query, configuration), 1);
builder.setBolt("storm-print-bolt", new PrinterBolt()).shuffleGrouping("es-storm-spout");
builder.setBolt("storm-handle-bolt", new HandleBolt()).shuffleGrouping("es-storm-spout");
RecordFormat recordFormat = new DelimitedRecordFormat().withFieldDelimiter(":");
SyncPolicy syncPolicy = new CountSyncPolicy(10);
FileRotationPolicy fileRotationPolicy = new TimedRotationPolicy(1.0f, TimeUnit.MINUTES);
FileNameFormat fileNameFormat = new DefaultFileNameFormat().withPath("/storm/")
.withPrefix("es_").withExtension(".log");
HdfsBolt hdfsBolt = new HdfsBolt().withFsUrl("hdfs://centos.host1:9000")
.withFileNameFormat(fileNameFormat).withRecordFormat(recordFormat)
.withRotationPolicy(fileRotationPolicy).withSyncPolicy(syncPolicy);
builder.setBolt("storm-hdfs-bolt", hdfsBolt).globalGrouping("storm-handle-bolt");
Config config = new Config();
config.setDebug(true);
if (isCluster) {
try {
config.setNumWorkers(3);
StormSubmitter.submitTopologyWithProgressBar(
TOPOLOGY_NAME, config, builder.createTopology());
} catch (Exception e) {
e.printStackTrace();
}
} else {
LocalCluster cluster = new LocalCluster();
cluster.submitTopology(TOPOLOGY_NAME, config, builder.createTopology());
Utils.sleep(100000);
cluster.killTopology(TOPOLOGY_NAME);
cluster.shutdown();
}
}
}
注意:elasticsearch-hadoop里的EsSpout类用到的Storm版本过低,所以重写了一个ESSpout替换旧版本Storm的API。
$bin/storm jar /home/hadoop/Documents/esstorm-0.0.1-SNAPSHOT.jar org.platform.storm.elasticsearch.ES2StormTopology false
import java.util.HashMap;
import java.util.Map;
import org.apache.storm.Config;
import org.apache.storm.LocalCluster;
import org.apache.storm.StormSubmitter;
import org.apache.storm.starter.bolt.PrinterBolt;
import org.apache.storm.topology.TopologyBuilder;
import org.apache.storm.utils.Utils;
import org.platform.storm.elasticsearch.bolt.ESBolt;
import org.platform.storm.elasticsearch.spout.ESSpout;
public class Storm2ESTopology {
private static final String TOPOLOGY_NAME = "storm-es-topology";
public static void main(String[] args) {
if (args.length != 1) {
System.exit(0);
}
boolean isCluster = Boolean.parseBoolean(args[0]);
TopologyBuilder builder = new TopologyBuilder();
String target = "operator/telecom";
String query = "?q=*";
Map<Object, Object> spoutConf = new HashMap<Object, Object>();
spoutConf.put("es.nodes", "192.168.10.20:9200");
spoutConf.put("es.read.field.include", "name,phone,rcall,email,idCard,zipCode,address");
spoutConf.put("es.storm.spout.fields", "name,phone,rcall,email,idCard,zipCode,address");
builder.setSpout("es-storm-spout", new ESSpout(target, query, spoutConf), 1);
builder.setBolt("storm-print-bolt", new PrinterBolt()).shuffleGrouping("es-storm-spout");
Map<Object, Object> boltConf = new HashMap<Object, Object>();
boltConf.put("es.nodes", "192.168.10.20:9200");
boltConf.put("es.index.auto.create", "true");
boltConf.put("es.ser.writer.bytes.class", "org.platform.storm.elasticsearch.bolt.StormTupleBytesConverter");
//boltConf.put("es.input.json", "true");
builder.setBolt("storm-es-bolt", new ESBolt("data/telecom", boltConf))
.globalGrouping("es-storm-spout");
Config config = new Config();
config.setDebug(true);
if (isCluster) {
try {
config.setNumWorkers(3);
StormSubmitter.submitTopologyWithProgressBar(
TOPOLOGY_NAME, config, builder.createTopology());
} catch (Exception e) {
e.printStackTrace();
}
} else {
LocalCluster cluster = new LocalCluster();
cluster.submitTopology(TOPOLOGY_NAME, config, builder.createTopology());
Utils.sleep(100000);
cluster.killTopology(TOPOLOGY_NAME);
cluster.shutdown();
}
}
}
注意:elasticsearch-hadoop里的EsBolt、StormTupleBytesConverter类用到的Storm版本过低,所以重写了一个ESBolt、StormTupleBytesConverter替换旧版本Storm的API。
$bin/storm jar /home/hadoop/Documents/esstorm-0.0.1-SNAPSHOT.jar org.platform.storm.elasticsearch.Storm2ESTopology false