Spark版本1.5.2，Flume版本:1.6
Flume agent配置文件：spool-8.51.conf
agent.sources = source1 agent.channels = memoryChannel agent.sinks = sink1 agent.sources.source1.type = spooldir agent.sources.source1.spoolDir=/data/apache-flume-1.6.0-bin/spooldir agent.sources.source1.fileHeader = true #agent.sources.source1.deletePolicy =immediate agent.channels.memoryChannel.type = memory agent.channels.memoryChannel.capacity = 1000 agent.channels.memoryChannel.keep-alive = 1000 agent.sinks.sink1.type = avro agent.sinks.sink1.hostname = 192.168.1.11 # 这是spark集群中任意executor 的ip agent.sinks.sink1.port = 23004 agent.sinks.sink1.channel = memoryChannel agent.sources.source1.channels = memoryChannel
maven文件：
<project xmlns="http://maven.apache.org/POM/4.0.0" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd"> <modelVersion>4.0.0</modelVersion> <groupId>cn.test</groupId> <artifactId>sparkTest</artifactId> <version>0.0.1</version> <packaging>jar</packaging> <name>pconliners</name> <url>http://maven.apache.org</url> <properties> <project.build.sourceEncoding>UTF-8</project.build.sourceEncoding> </properties> <dependencies> <dependency> <groupId>org.apache.spark</groupId> <artifactId>spark-core_2.10</artifactId> <version>1.5.1</version> </dependency> <dependency> <groupId>org.apache.spark</groupId> <artifactId>spark-sql_2.10</artifactId> <version>1.5.1</version> </dependency> <dependency> <groupId>org.apache.spark</groupId> <artifactId>spark-streaming_2.10</artifactId> <version>1.5.2</version> </dependency> <dependency> <groupId>org.apache.spark</groupId> <artifactId>spark-streaming_2.10</artifactId> <version>1.5.2</version> </dependency> <dependency> <groupId>org.apache.spark</groupId> <artifactId>spark-streaming-flume_2.10</artifactId> <version>1.5.2</version> </dependency> <dependency> <groupId>org.apache.flume</groupId> <artifactId>flume-ng-sdk</artifactId> <version>1.5.2</version> </dependency> </dependencies> <build> <plugins> <plugin> <artifactId>maven-assembly-plugin</artifactId> <configuration> <descriptorRefs> <descriptorRef>jar-with-dependencies</descriptorRef> </descriptorRefs> </configuration> <executions> <execution> <id>make-assembly</id> <phase>package</phase> <goals> <goal>single</goal> </goals> </execution> </executions> </plugin> </plugins> </build> </project>
Java测试代码
public final class FlumeEventCount {
public static void main(String[] args) {
String host = args[0];
int port = Integer.parseInt(args[1]);
Duration batchInterval = new Duration(Integer.parseInt(args[2]));
SparkConf sparkConf = new SparkConf().setAppName("JavaFlumeEventCount");
JavaStreamingContext ssc = new JavaStreamingContext(sparkConf,batchInterval);
JavaReceiverInputDStream<SparkFlumeEvent> flumeStream = FlumeUtils.createStream(ssc, host, port);
System.out.println("flumeStream.count():"+flumeStream.count());;
flumeStream.count().map(new Function<Long, String>() {
private static final long serialVersionUID = -572435064083746235L;
public String call(Long in) {
System.out.println("Flume test ....."+in);
return "Received " + in + " flume events....";
}
}).print();
ssc.start();
ssc.awaitTermination();
}
}
打成jar包，启动Spark streaming程序
spark-submit --class cn.test.FlumeEventCount --master spark://192.168.1.10:7077 sparkTest-0.0.1-jar-with-dependencies.jar 192.168.1.11 23004 5000
运行agent：
cd到flume安装目录，执行。
bin/flume-ng agent -n agent -c conf -f conf/spool-8.51.conf -Dflume.root.logger=DEBUG,console
复制文件到监控目录：
cp spool-test.txt /data/apache-flume-1.6.0-bin/spooldir/
查看提交Spark 任务输出：
------------------------------------------- Time: 1472202305000 ms ------------------------------------------- Received 120 flume events....
基于拉模式
Java代码：
import org.apache.spark.SparkConf;
import org.apache.spark.api.java.function.Function;
import org.apache.spark.streaming.Durations;
import org.apache.spark.streaming.StreamingContext;
import org.apache.spark.streaming.api.java.JavaReceiverInputDStream;
import org.apache.spark.streaming.api.java.JavaStreamingContext;
import org.apache.spark.streaming.flume.FlumeUtils;
import org.apache.spark.streaming.flume.SparkFlumeEvent;
public class SparkStreamingFlume2 {
public static void main(String[] args) {
SparkConf conf = new SparkConf().setAppName("Streaming...");
StreamingContext streamingContext = new StreamingContext(conf,Durations.seconds(30));
JavaStreamingContext ssc = new JavaStreamingContext(streamingContext);
String host = args[0];
int port = Integer.parseInt(args[1]);
JavaReceiverInputDStream<SparkFlumeEvent> pollingStream = FlumeUtils.createPollingStream(ssc, host, port);
pollingStream.count().map(new Function<Long, String>() {
public String call(Long v1) throws Exception {
return "Received " + v1 + " flume events.";
}
}).print();
ssc.start();
ssc.awaitTermination();
}
}
flume配置文件，在flume的conf目录创建一个flume-pull.conf文件，
数据来源是netcat：
a1.sources = r1 a1.channels = c1 a1.sinks = k1 a1.sources.r1.type = netcat a1.sources.r1.bind = 192.168.11.16 a1.sources.r1.port = 22222 a1.sources.r1.channels = c1 a1.channels.c1.type = memory a1.channels.c1.capacity = 2000 a1.channels.c1.transactionCapacity = 1000 a1.sinks.k1.type = org.apache.spark.streaming.flume.sink.SparkSink a1.sinks.k1.hostname = 192.168.11.16 a1.sinks.k1.port = 11111 a1.sinks.k1.channel = c1
数据来源是文件夹：
a1.sources = r1 a1.channels = c1 a1.sinks = k1 a1.sources.r1.type = spooldir a1.sources.r1.spoolDir=/data/apache-flume-1.6.0-bin/spooldir a1.sources.r1.fileHeader = true a1.sources.r1.deletePolicy =immediate a1.sources.r1.channels = c1 a1.channels.c1.type = memory a1.channels.c1.capacity = 2000 a1.channels.c1.transactionCapacity = 1000 a1.sinks.k1.type = org.apache.spark.streaming.flume.sink.SparkSink a1.sinks.k1.hostname = 192.168.11.16 a1.sinks.k1.port = 11111 a1.sinks.k1.channel = c1
这里使用的是文件夹作为数据来源。
由于用到了agent的sink是 org.apache.spark.streaming.flume.sink.SparkSink类型，需要把spark-streaming-flume-sink_2.10-1.5.2.jar复制到flume的lib目录，否则，会报找不到org.apache.spark.streaming.flume.sink.SparkSink类的错误。
org.apache.flume.FlumeException: Unable to load sink type: org.apache.spark.streaming.flume.sink.SparkSink, class: org.apache.spark.streaming.flume.sink.SparkSink
at org.apache.flume.sink.DefaultSinkFactory.getClass(DefaultSinkFactory.java:71)
at org.apache.flume.sink.DefaultSinkFactory.create(DefaultSinkFactory.java:43)
at org.apache.flume.node.AbstractConfigurationProvider.loadSinks(AbstractConfigurationProvider.java:410)
at org.apache.flume.node.AbstractConfigurationProvider.getConfiguration(AbstractConfigurationProvider.java:98)
at org.apache.flume.node.PollingPropertiesFileConfigurationProvider$FileWatcherRunnable.run(PollingPropertiesFileConfigurationProvider.java:140)
at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
at java.util.concurrent.FutureTask.runAndReset(FutureTask.java:304)
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.access$301(ScheduledThreadPoolExecutor.java:178)
at java.util.concurrent.ScheduledThreadPoolExecutor$ScheduledFutureTask.run(ScheduledThreadPoolExecutor.java:293)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:744)
Caused by: java.lang.ClassNotFoundException: org.apache.spark.streaming.flume.sink.SparkSink
at java.net.URLClassLoader$1.run(URLClassLoader.java:366)
at java.net.URLClassLoader$1.run(URLClassLoader.java:355)
at java.security.AccessController.doPrivileged(Native Method)
at java.net.URLClassLoader.findClass(URLClassLoader.java:354)
at java.lang.ClassLoader.loadClass(ClassLoader.java:425)
at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:308)
at java.lang.ClassLoader.loadClass(ClassLoader.java:358)
at java.lang.Class.forName0(Native Method)
at java.lang.Class.forName(Class.java:190)
at org.apache.flume.sink.DefaultSinkFactory.getClass(DefaultSinkFactory.java:69)
... 11 more
先启动flume，
bin/flume-ng agent --conf conf --conf-file conf/flume-pull.conf --name a1 -Dflume.root.logger=INFO,console
注意，flume的–name参数项要跟配置项的agent名一致，配置文件不要弄错。
在控制台看到如下如下信息，agent分别启动了channel、sink、source，说明，agent启动成功
2016-08-30 15:20:45,990 (lifecycleSupervisor-1-0) [INFO - org.apache.flume.instrumentation.MonitoredCounterGroup.start(MonitoredCounterGroup.java:96)] Component type: CHANNEL, name: c1 started 2016-08-30 15:20:45,991 (conf-file-poller-0) [INFO - org.apache.flume.node.Application.startAllComponents(Application.java:173)] Starting Sink k1 2016-08-30 15:20:45,991 (conf-file-poller-0) [INFO - org.apache.flume.node.Application.startAllComponents(Application.java:184)] Starting Source r1 2016-08-30 15:20:45,992 (lifecycleSupervisor-1-0) [INFO - org.apache.flume.source.SpoolDirectorySource.start(SpoolDirectorySource.java:78)] SpoolDirectorySource source starting with directory: /data/apache-flume-1.6.0-bin/spooldir 2016-08-30 15:20:45,992 (lifecycleSupervisor-1-1) [INFO - org.apache.spark.streaming.flume.sink.Logging$class.logInfo(Logging.scala:47)] Starting Spark Sink: k1 on port: 11111 and interface: 192.168.11.16 with pool size: 10 and transaction timeout: 60. 2016-08-30 15:20:46,021 (lifecycleSupervisor-1-0) [INFO - org.apache.flume.instrumentation.MonitoredCounterGroup.register(MonitoredCounterGroup.java:120)] Monitored counter group for type: SOURCE, name: r1: Successfully registered new MBean. 2016-08-30 15:20:46,021 (lifecycleSupervisor-1-0) [INFO - org.apache.flume.instrumentation.MonitoredCounterGroup.start(MonitoredCounterGroup.java:96)] Component type: SOURCE, name: r1 started 2016-08-30 15:20:46,462 (lifecycleSupervisor-1-1) [INFO - org.apache.spark.streaming.flume.sink.Logging$class.logInfo(Logging.scala:47)] Starting Avro server for sink: k1 2016-08-30 15:20:46,464 (SinkRunner-PollingRunner-DefaultSinkProcessor) [INFO - org.apache.spark.streaming.flume.sink.Logging$class.logInfo(Logging.scala:47)] Blocking Sink Runner, sink will continue to run..
再提交Spark Streaming任务，
spark-submit --class cn.test.SparkStreamingFlume2 --master spark://192.168.8.51:7077 sparkTest-0.0.1-jar-with-dependencies.jar 192.168.11.16 11111
192.168.11.16就是启动了agent的ip。
看到agent的控制台输出了Spark Streaming任务已经连接了agent的消息：
2016-08-30 15:21:10,896 (New I/O server boss #1 ([id: 0xbad2f716, /192.168.11.16:11111])) [INFO - org.apache.avro.ipc.NettyServer$NettyServerAvroHandler.handleUpstream(NettyServer.java:171)] [id: 0x4e25ab4e, /192.168.8.55:55293 => /192.168.11.16:11111] OPEN 2016-08-30 15:21:10,898 (New I/O worker #1) [INFO - org.apache.avro.ipc.NettyServer$NettyServerAvroHandler.handleUpstream(NettyServer.java:171)] [id: 0x4e25ab4e, /192.168.8.55:55293 => /192.168.11.16:11111] BOUND: /192.168.11.16:11111 2016-08-30 15:21:10,898 (New I/O worker #1) [INFO - org.apache.avro.ipc.NettyServer$NettyServerAvroHandler.handleUpstream(NettyServer.java:171)] [id: 0x4e25ab4e, /192.168.8.55:55293 => /192.168.11.16:11111] CONNECTED: /192.168.8.55:55293
在启动agent的机器，复制一些文件到/data/apache-flume-1.6.0-bin/spooldir，cp data.txt /data/apache-flume-1.6.0-bin/spooldir/
看到Spark 任务输出，说明测试成功。
------------------------------------------- Time: 1472541990000 ms ------------------------------------------- Received 54 flume events.
如果agent出现以下错误
org.apache.flume.ChannelException: Take list for MemoryTransaction, capacity 100 full, consider committing more frequently, increasing capacity, or increasing thread count
at org.apache.flume.channel.MemoryChannel$MemoryTransaction.doTake(MemoryChannel.java:96)
at org.apache.flume.channel.BasicTransactionSemantics.take(BasicTransactionSemantics.java:113)
at org.apache.flume.channel.BasicChannelSemantics.take(BasicChannelSemantics.java:95)
at org.apache.spark.streaming.flume.sink.TransactionProcessor$$anonfun$populateEvents$1$$anonfun$apply$1.apply$mcV$sp(TransactionProcessor.scala:123)
at scala.util.control.Breaks.breakable(Breaks.scala:37)
at org.apache.spark.streaming.flume.sink.TransactionProcessor$$anonfun$populateEvents$1.apply(TransactionProcessor.scala:119)
at org.apache.spark.streaming.flume.sink.TransactionProcessor$$anonfun$populateEvents$1.apply(TransactionProcessor.scala:113)
at scala.Option.foreach(Option.scala:236)
at org.apache.spark.streaming.flume.sink.TransactionProcessor.populateEvents(TransactionProcessor.scala:113)
at org.apache.spark.streaming.flume.sink.TransactionProcessor.call(TransactionProcessor.scala:243)
at org.apache.spark.streaming.flume.sink.TransactionProcessor.call(TransactionProcessor.scala:43)
at java.util.concurrent.FutureTask.run(FutureTask.java:262)
at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
at java.lang.Thread.run(Thread.java:744)
那就设置一下channel的内存好了
a1.channels.c1.capacity = 2000 a1.channels.c1.transactionCapacity = 1000
获取Flume数据：示例：
pollingStream.map(new Function<SparkFlumeEvent, String>() {
public String call(SparkFlumeEvent v1) throws Exception {
return new String(v1.event().getBody().array());
}
}).print();
测试拉模式
首先，我们将Spark Streaming 任务停掉，然后将文件复制到监控文件夹下，cp ../conf/* .
这时候Flume的日志是：
2016-08-30 16:47:49,727 (New I/O worker #1) [INFO - org.apache.avro.ipc.NettyServer$NettyServerAvroHandler.handleUpstream(NettyServer.java:171)] [id: 0x4c760232, /192.168.8.56:58196 :> /192.168.11.16:11111] DISCONNECTED 2016-08-30 16:47:49,728 (New I/O worker #1) [INFO - org.apache.avro.ipc.NettyServer$NettyServerAvroHandler.handleUpstream(NettyServer.java:171)] [id: 0x4c760232, /192.168.8.56:58196 :> /192.168.11.16:11111] UNBOUND 2016-08-30 16:47:49,728 (New I/O worker #1) [INFO - org.apache.avro.ipc.NettyServer$NettyServerAvroHandler.handleUpstream(NettyServer.java:171)] [id: 0x4c760232, /192.168.8.56:58196 :> /192.168.11.16:11111] CLOSED 2016-08-30 16:47:49,728 (New I/O worker #1) [INFO - org.apache.avro.ipc.NettyServer$NettyServerAvroHandler.channelClosed(NettyServer.java:209)] Connection to /192.168.8.56:58196 disconnected.
再提交Spark Streaming任务，看到Flume控制台如下输出：Flume重新
2016-08-30 16:48:49,726 (Spark Sink Processor Thread - 2) [WARN - org.apache.spark.streaming.flume.sink.Logging$class.logWarning(Logging.scala:59)] Spark was unable to successfully process the events. Transaction is being rolled back.
Spark控制台输出：
------------------------------------------- Time: 1472546940000 ms ------------------------------------------- Received 54 flume events.
这个测试说明，Flume基于拉模式下，数据不会丢失。
