Storm之——Storm on Yarn 部署

转载请注明出处：http://blog.csdn.net/l1028386804/article/details/79172573

1. 环境介绍

一、节点与服务映射关系

ip	host	服务
192.168.209.121	master	Namenode、NodeManager、DataNode、zookeeper
192.168.209.122	slave1	ResurceManager、NodeManager、DataNode、zookeeper
192.168.209.123	slave2	NodeManager、DataNode、zookeeper

二、安装Zookeeper集群

参考博文《Hadoop之——Hadoop2.5.2 HA高可靠性集群搭建(Hadoop+Zookeeper)》或   《Storm之——搭建Storm集群》

三、安装Hadoop

1、伪分布式安装

请参考博文：《Hadoop之——Hadoop2.4.1伪分布搭建》

2、 集群安装

请参考博文《Hadoop之——CentOS + hadoop2.5.2分布式环境配置》

3、 高可用集群安装

请参考博文《Hadoop之——Hadoop2.5.2 HA高可靠性集群搭建(Hadoop+Zookeeper)前期准备》和《Hadoop之——Hadoop2.5.2 HA高可靠性集群搭建(Hadoop+Zookeeper)》

这篇博文中，我也是在单节点上安装的Hadoop，将HBase和Hadoop安装在了同一台服务器上。由于HBase的运行依赖于Zookeeper，所以，在同一台服务器上，又安装了单节点的Zookeeper。

四、安装Storm集群

参考博文《Storm之——搭建Storm集群》

五、安装maven

1、下载、解压maven

wget http://mirrors.tuna.tsinghua.edu.cn/apache/maven/maven-3/3.3.9/binaries/apache-maven-3.3.9-bin.tar.gz
tar -xzvf apache-maven-3.3.9-bin.tar.gz
ln -s apache-maven-3.3.9/ maven

2、配置Maven环境变量

export MAVEN_HOME=/application/maven
export PATH=.:$MAVEN_HOME/bin:$JAVA_HOME/bin:$HADOOP_HOME/bin:$HADOOP_HOME/sbin:$PATH
source /etc/profile

3、验证，查看maven版本号

mvn -v
Apache Maven 3.3.9 (bb52d8502b132ec0a5a3f4c09453c07478323dc5; 2015-11-11T00:41:47+08:00)
Maven home: /application/maven
Java version: 1.7.0_79, vendor: Oracle Corporation
Java home: /application/jdk1.7.0_79/jre
Default locale: en_US, platform encoding: UTF-8
OS name: "linux", version: "2.6.32-504.el6.x86_64", arch: "amd64", family: "unix"

六、安装storm-yarn

1、下载storm-yarn

wget https://github.com/yahoo/storm-yarn/archive/master.zip

重命名下载的压缩包为storm-yarn.zip

mv master.zip storm-yarn.zip

解压

unzip storm-yarn.zip

2、修改pom.xml

根据hadoop的版本号修改storm-yarn-master/pom.xml

<properties>
    <storm.version>0.9.6</storm.version>
    <hadoop.version>2.6.0</hadoop.version>
</properties>

因为我的环境中hadoop的版本是2.6.0，没有安装storm（也不用安装），所以pom.xml没有修改。

3、编译打包storm-yarn-master

mvn package -DskipTests
打包成功
[INFO] --- maven-jar-plugin:2.4:jar (default-jar) @ storm-yarn ---
[INFO] ----------------------------------------------------------------------
[INFO] BUILD SUCCESS
[INFO] ----------------------------------------------------------------------
[INFO] Total time: 08:41 min
[INFO] Finished at: 2018-01-26T10:16:27+08:00
[INFO] Final Memory: 14M/34M
[INFO] ----------------------------------------------------------------------

4、解压lib/storm.zip并移动解压后的目录

解压storm-yarn-master/lib/storm.zip

unzip storm-yarn-master/lib/storm.zip 

[root@master lib]# ls
storm-0.9.0-wip21  storm-0.9.0-wip21.zip  storm-starter-0.0.1-SNAPSHOT.jar  storm.zip

将解压后生成的storm-0.9.0-wip21移到和storm-yarn-master同级的目录

[root@master lib]mv storm-0.9.0-wip21 ../..
[root@master lib]# cd ../..
[root@master application]# ls
maven   storm-yarn.zip  apache-maven-3.3.9  tools   hadoop  zookeeper   hadoop-2.6.0    java    storm-0.9.0-wip21   zookeeper-3.3.6 jdk1.7.0_79 storm-yarn-master   

5、配置环境变量

export PATH=.:/application/storm-0.9.0-wip21/bin:/application/storm-yarn-master/bin:$MAVEN_HOME/bin::$JAVA_HOME/bin:$HADOOP_HOME/bin:$HADOOP_HOME/sbin:$PATH

使用source /etc/profile使之生效

6、编辑 storm-0.9.0-wip21/conf/storm.yaml

修改zookeeper地址

storm.zookeeper.servers:
   - "master"
   - "slave1"
   - "slave2"

7、压缩storm.zip并上传到hdfs上

压缩storm-0.9.0-wip21

zip -r storm.zip storm-0.9.0-wip21 

将压缩后的文件storm.zip上传到hdfs上的目录/lib/storm/0.9.0-wip21/

hadoop fs -mkdir -p /lib/storm/0.9.0-wip21
hdfs dfs -put storm.zip /lib/storm/0.9.0-wip21/
hdfs dfs -ls /lib/storm/0.9.0-wip21/
Found 1 items
-rw-r--r--   3 root supergroup   17141097 2016-08-26 10:27 /lib/storm/0.9.0-wip21/storm.zip

七、运行storm-yarn

1、启动storm-yarn

[root@master application]# storm-yarn launch storm-0.9.0-wip21/conf/storm.yaml 
SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/application/hadoop-2.6.0/share/hadoop/common/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/application/storm-0.9.0-wip21/lib/logback-classic-1.0.6.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]
18/01/26 10:28:28 INFO client.RMProxy: Connecting to ResourceManager at slave1/192.168.40.133:8032
18/01/26 10:28:29 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
18/01/26 10:28:30 INFO yarn.StormOnYarn: Copy App Master jar from local filesystem and add to local environment
18/01/26 10:28:32 INFO yarn.StormOnYarn: Set the environment for the application master
18/01/26 10:28:32 INFO yarn.StormOnYarn: YARN CLASSPATH COMMAND = [[yarn, classpath]]
18/01/26 10:28:32 INFO yarn.StormOnYarn: Using JAVA_HOME = [/application/java]
18/01/26 10:28:32 INFO yarn.StormOnYarn: Setting up app master command:[/application/java/bin/java, -Dstorm.home=./storm/storm-0.9.0-wip21/, -Dlogfile.name=<LOG_DIR>/master.log, com.yahoo.storm.yarn.MasterServer, 1><LOG_DIR>/stdout, 2><LOG_DIR>/stderr]
18/01/26 10:28:32 INFO impl.YarnClientImpl: Submitted application application_1516932395528_0002
application_1516932395528_0002

通过http://192.168.209.121:8088查看yarn的appplication运行情况

2、获取nimbus地址

得到storm的属性信息，并保存到/root/.storm/storm.yaml中

[root@master application]# mkdir -p /root/.storm/
[root@master application]# storm-yarn getStormConfig -appId application_1516932395528_0002  -output ~/.storm/storm.yaml
SLF4J: Class path contains multiple SLF4J bindings.
SLF4J: Found binding in [jar:file:/application/hadoop-2.6.0/share/hadoop/common/lib/slf4j-log4j12-1.7.5.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: Found binding in [jar:file:/application/storm-0.9.0-wip21/lib/logback-classic-1.0.6.jar!/org/slf4j/impl/StaticLoggerBinder.class]
SLF4J: See http://www.slf4j.org/codes.html#multiple_bindings for an explanation.
SLF4J: Actual binding is of type [org.slf4j.impl.Log4jLoggerFactory]
18/01/26 10:30:13 INFO client.RMProxy: Connecting to ResourceManager at slave1/192.168.209.122:8032
18/01/26 10:30:13 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
18/01/26 10:30:14 INFO yarn.StormOnYarn: application report for application_1516932395528_0002 :slave2:9000
18/01/26 10:30:14 INFO yarn.StormOnYarn: Attaching to slave2:9000 to talk to app master application_1516932395528_0002
{10:35:18}18/01/26 10:30:14 INFO yarn.StormMasterCommand: storm.yaml downloaded into /root/.storm/storm.yaml

根据/root/.storm/storm.yaml获取nimbus地址

[root@master .storm]# cat ~/.storm/storm.yaml | grep nimbus.host
nimbus.host: 192.168.209.123

3、通过nimbus访问storm UI

通过访问http://192.168.209.123:7070访问storm UI

4、提交任务wordcount

[root@master storm-yarn-master]# storm jar lib/storm-starter-0.0.1-SNAPSHOT.jar storm.starter.WordCountTopology WordCountTopology -c nimbus.host=192.168.40.134
561  [main] INFO  backtype.storm.StormSubmitter - Jar not uploaded to master yet. Submitting jar...
573  [main] INFO  backtype.storm.StormSubmitter - Uploading topology jar lib/storm-starter-0.0.1-SNAPSHOT.jar to assigned location: storm-local/nimbus/inbox/stormjar-12a737f5-82e9-43a6-b422-c1740cc36635.jar
596  [main] INFO  backtype.storm.StormSubmitter - Successfully uploaded topology jar to assigned location: storm-local/nimbus/inbox/stormjar-12a737f5-82e9-43a6-b422-c1740cc36635.jar
596  [main] INFO  backtype.storm.StormSubmitter - Submitting topology WordCountTopology in distributed mode with conf {"nimbus.host":"192.168.40.134","topology.workers":3,"topology.debug":true}
948  [main] INFO  backtype.storm.StormSubmitter - Finished submitting topology: WordCountTopology

查看Storm UI
查看topology wordcount的运行情况
查看各节点jps

八、关闭

1、关闭storm 提交的任务

[root@master ~]# storm kill WordCountTopology
2277 [main] INFO  backtype.storm.thrift - Connecting to Nimbus at 192.168.209.123:6627
2387 [main] INFO  backtype.storm.command.kill-topology - Killed topology: WordCountTopology

2、关闭storm on yarn

[root@master ~]# yarn application -kill application_1516932395528_0002
Killing application application_1516932395528_0002