hadoop分布式安装

时间 2019-11-18

原文原文链接

先配置免登陆html

先准备了7个节点java

192.168.101.172       node1          
192.168.101.206       node2
192.168.101.207       node3
192.168.101.215       node4
192.168.101.216       node5
192.168.101.217       node6
192.168.101.218       node7

每台机器都修改hosts文件以下:node

127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4
::1 localhost localhost.localdomain localhost6 localhost6.localdomain6
192.168.101.172 node1
192.168.101.206 node2
192.168.101.207 node3
192.168.101.215 node4
192.168.101.216 node5
192.168.101.217 node6
192.168.101.218 node7linux

修改主机名:git

vim /etc/sysconfig/network
NETWORKING=yes
HOSTNAME=node2

关防火墙:github

centos6:web

service iptables stop算法

centos7:shell

systemctl stop firewalld
systemctl disable firewalldexpress

建立用户:

groupadd hadoop
useradd hadoop -g hadoop
mkdir /main/bigdata
chown -R hadoop:hadoop /main/bigdata/
passwd hadoop

时间同步centos6:

yum -y installntp

ntpdate cn.pool.ntp.org

开始配置免登陆:

免密的核心思想就是：若是B服务器authorized_keys有A服务器的公钥（锁），那么A服务器能够免密登陆B服务器。

先到7台机器分别生成密钥:

# 安装ssh
yum -y install openssh-clients


su hadoop
rm -rf ~/.ssh/*
ssh-keygen -t rsa
# 一路回车,而后拷贝到受权文件
cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys

接着要让一台机器能够访问其余的全部机器:

参考了 https://blog.csdn.net/ITYang_/article/details/70144395

先在全部6台机器上执行如家命令,把各个机器的密钥发送到node1(172)机器的authorized_keys中去:


#centos6
ssh-copy-id "-p 2121 192.168.101.172"

#centos7
ssh-copy-id -p 2121 hadoop@node1



# 或更详细的:  当上一步是dsa的时候,这里必需要指定算法是id_dsa.pub

ssh-copy-id -i ~/.ssh/id_dsa.pub "-p 2121 192.168.101.172"

把咱们的公钥都传递给172这个节点. 在172(node1)上执行:

cat .ssh/authorized_keys

会看到别的机器节点密钥都传上来了

测试:

此时在别的节点上执行

ssh node1 -p 2121

发现能够免密码进入node1 !

下一步是把咱们辛苦收集到node1机器的.ssh/authorized_keys文件分发到各个节点上去:

在node1的机器上执行一个shell:

yum install expect

而后:

#!/bin/bash
SERVERS="192.168.101.172 192.168.101.206 192.168.101.207 192.168.101.215 192.168.101.216 192.168.101.217 192.168.101.218"
PASSWORD=机器密码

# 将当前机器的密钥copy到其余的节点上,为了保证下一步scp的时候能够免密登陆
auto_ssh_copy_id() {
    expect -c "set timeout -1;
        # 若是缺省端口的话  能够直接spawn ssh-copy-id $1;
        spawn ssh-copy-id \" -p 2121 $1 \";
        expect {
            *(yes/no)* {send -- yes\r;exp_continue;}
            *assword:* {send -- $2\r;exp_continue;}
            eof        {exit 0;}
        }";
}

# 循环全部的机器,开始copy
ssh_copy_id_to_all() {
    for SERVER in $SERVERS
        do
            auto_ssh_copy_id $SERVER $PASSWORD
        done
}

# 调用上面的方法
ssh_copy_id_to_all


# 循环全部的机器ip,scp把当前机器的密钥全都copy到别的机器上去(端口2121,若是缺省的话能够不填写)  用户名hadoop根据实际状况更换
for SERVER in $SERVERS
do
    scp -P 2121 ~/.ssh/authorized_keys hadoop@$SERVER:~/.ssh/
done

centos7的 copy-id命令语法变了,因此shell变为:

#!/bin/bash
SERVERS="node1 node2 node3 node4 node5 node6 node7 node8"
USERNAME=hadoop
PASSWORD=机器密码

# 将当前机器的密钥copy到其余的节点上,为了保证下一步scp的时候能够免密登陆
auto_ssh_copy_id() {
    expect -c "set timeout -1;
        # 若是缺省端口的话  能够直接spawn ssh-copy-id $1;
        spawn ssh-copy-id  -p 2121 $2@$1 ;
        expect {
            *(yes/no)* {send -- yes\r;exp_continue;}
            *assword:* {send -- $3\r;exp_continue;}
            eof        {exit 0;}
        }";
}

# 循环全部的机器,开始copy
ssh_copy_id_to_all() {
    for SERVER in $SERVERS
        do
            auto_ssh_copy_id $SERVER  $USERNAME $PASSWORD
        done
}

# 调用上面的方法
ssh_copy_id_to_all


# 循环全部的机器ip,scp把当前机器的密钥全都copy到别的机器上去(端口2121,若是缺省的话能够不填写)  用户名hadoop根据实际状况更换
for SERVER in $SERVERS
do
    scp -P 2121 ~/.ssh/authorized_keys $USERNAME@$SERVER:~/.ssh/
done

可是实际状况在使用非root用户的时候发现了很奇怪的现象:

centos6小插曲:

node1收集到别的6个节点的密钥以后,反向ssh-copy-id回去以后,没法ssh免密码登陆对方,scp的时候仍然须要输入密码....

可是别的节点删掉本地的~/.ssh/authorized_keys 以后就能够了.....

结果我就在其余6个节点执行了rm ~/.ssh/authorized_keys以后,再在node1运行这个shell就能够不用输入密码了 ....

root用户没有这个问题...

好奇怪. 有读者解决了以后能够和我沟通下,共同窗习.

在这台node1机器分别执行ssh nodeX -p xxxx 以后选yes,会缓存knowhost

而后scp这个文件到其余的机器上去:

#!/bin/bash
SERVERS="node1 node2 node3 node4 node5 node6 node7 node8"
for SERVER in $SERVERS
do
    scp -P 2121 ~/.ssh/known_hosts hadoop@$SERVER:~/.ssh/
done

这样其余的机器就不用首次输入yes了

至此,机器面密码登陆搞定.

安装Java8:

cd /main
tar -zvxf soft/jdk-8u171-linux-x64.tar.gz -C .

vim /etc/profile

# 最后加上
export JAVA_HOME=/main/jdk1.8.0_171
export JRE_HOME=$JAVA_HOME/jre
export CLASSPATH=$JAVA_HOME/lib  
export PATH=:$PATH:$JAVA_HOME/bin:$JRE_HOME/bin


rm -rf /usr/bin/java*
# 当即生效
source /etc/profile

zookeeper安装配置省略 ,基本就是解压改路径,看心情改端口,设置myid就ok

# The number of milliseconds of each tick
tickTime=2000
# The number of ticks that the initial 
# synchronization phase can take
initLimit=10
# The number of ticks that can pass between 
# sending a request and getting an acknowledgement
syncLimit=5
# the directory where the snapshot is stored.
# do not use /tmp for storage, /tmp here is just 
# example sakes.
dataDir=/main/zookeeper/data
dataLogDir=/main/zookeeper/logs
# the port at which the clients will connect
clientPort=2181
# the maximum number of client connections.
# increase this if you need to handle more clients
#maxClientCnxns=60
#
# Be sure to read the maintenance section of the 
# administrator guide before turning on autopurge.
#
# http://zookeeper.apache.org/doc/current/zookeeperAdmin.html#sc_maintenance
#
# The number of snapshots to retain in dataDir
#autopurge.snapRetainCount=3
# Purge task interval in hours
# Set to "0" to disable auto purge feature
#autopurge.purgeInterval=1
server.1=192.168.101.173:2888:3888
server.2=192.168.101.183:2888:3888
server.3=192.168.101.193:2888:3888

安装hadoop:

安装5个节点

准备在node1-2五台机器安装,其中node1/node2做为master, node3-5做为slave

在node1的机器上进行配置:

cd /main
tar -zvxf soft/hadoop-2.8.4.tar.gz -C .
chown -R hadoop:hadoop /main/hadoop-2.8.4/

su hadoop
mkdir /main/hadoop-2.8.4/data
mkdir /main/hadoop-2.8.4/data/journal
mkdir /main/hadoop-2.8.4/data/tmp
mkdir /main/hadoop-2.8.4/data/hdfs
mkdir /main/hadoop-2.8.4/data/hdfs/namenode
mkdir /main/hadoop-2.8.4/data/hdfs/datanode

继续修改/main/hadoop-2.8.4/etc/hadoop/下的:

hadoop-env.sh

mapred-env.sh

yarn-env.sh

三个文件设置java_home

export JAVA_HOME=/main/jdk1.8.0_171

hadoop-env.sh最后一行加上ssh的端口

export HADOOP_SSH_OPTS="-p 2121"

修改/main/hadoop-2.8.4/etc/hadoop/slaves文件

node3
node4
node5

修改/main/hadoop-2.8.4/etc/hadoop/core-site.xml

<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<!--
  Licensed under the Apache License, Version 2.0 (the "License");
  you may not use this file except in compliance with the License.
  You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

  Unless required by applicable law or agreed to in writing, software
  distributed under the License is distributed on an "AS IS" BASIS,
  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  See the License for the specific language governing permissions and
  limitations under the License. See accompanying LICENSE file.
-->

<!-- Put site-specific property overrides in this file. -->

<configuration>
    <!-- fs.defaultFS须要设置成HDFS的逻辑服务名（需与hdfs-site.xml中的dfs.nameservices一致） -->
    <property>  
        <name>fs.defaultFS</name>  
        <value>hdfs://hjbdfs</value>  
    </property>  
    <property>  
        <name>hadoop.tmp.dir</name>  
        <value>/main/hadoop-2.8.4/data/tmp</value>  
    </property>  
    <property>  
        <name>hadoop.http.staticuser.user</name>  
        <value>hadoop</value>  
    </property>  
    <property>  
        <name>ha.zookeeper.quorum</name>  
        <value>192.168.101.173:2181,192.168.101.173:2181,192.168.101.173:2181</value>  
    </property>
</configuration>

更多参数详见: http://hadoop.apache.org/docs/current/hadoop-project-dist/hadoop-common/core-default.xml

修改 hdfs-site.xml:

注意故障切换sshfence节点的端口若是ssh不是默认的22端口,须要设置为sshfence([[username][:port]])好比 sshfence(hadoop:2121) 不然active的namenode挂了以后,sshfence没法进去到另一台机器去,致使没法自动切换主备.

<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<!--
  Licensed under the Apache License, Version 2.0 (the "License");
  you may not use this file except in compliance with the License.
  You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

  Unless required by applicable law or agreed to in writing, software
  distributed under the License is distributed on an "AS IS" BASIS,
  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  See the License for the specific language governing permissions and
  limitations under the License. See accompanying LICENSE file.
-->

<!-- Put site-specific property overrides in this file. -->

<configuration>
    <property>
        <name>dfs.replication</name>
        <value>3</value>
    </property>
    <property>
        <name>dfs.permissions.enabled</name>
        <value>false</value>
    </property>
    <!-- HDFS NN的逻辑名称，使用上面crore中fs.defaultFS设置的hjbdfs -->
    <property>
        <name>dfs.nameservices</name>
        <value>hjbdfs</value>
    </property>
    <property>
        <name>dfs.blocksize</name>
        <value>134217728</value>
    </property>
    <!-- 给定服务逻辑名称myhdfs的节点列表 -->
    <property>
        <name>dfs.ha.namenodes.hjbdfs</name>
        <value>nn1,nn2</value>
    </property>
    <!-- nn1的RPC通讯地址，nn1所在地址  -->
    <property>
        <name>dfs.namenode.rpc-address.hjbdfs.nn1</name>
        <value>node1:8020</value>
    </property>
    <!-- nn1的http通讯地址，外部访问地址 -->
    <property>
        <name>dfs.namenode.http-address.hjbdfs.nn1</name>
        <value>node1:50070</value>
    </property>
    <!-- nn2的RPC通讯地址，nn2所在地址 -->
    <property>
        <name>dfs.namenode.rpc-address.hjbdfs.nn2</name>
        <value>node2:8020</value>
    </property>
    <!-- nn2的http通讯地址，外部访问地址 -->
    <property>
        <name>dfs.namenode.http-address.hjbdfs.nn2</name>
        <value>node2:50070</value>
    </property>
    <!-- 指定NameNode的元数据在JournalNode日志上的存放位置(通常和zookeeper部署在一块儿) -->
    <!--  设置一组 journalNode 的 URI 地址，active NN 将 edit log 写入这些JournalNode，而 standby NameNode 读取这些 edit log，并做用在内存中的目录树中。若是journalNode有多个节点则使用分号分割。该属性值应符合如下格式qjournal://host1:port1;host2:port2;host3:port3/journalId -->
    <property>
        <name>dfs.namenode.shared.edits.dir</name>
        <value>qjournal://node3:8485;node4:8485;node5:8485/hjbdf_journal</value>
    </property>
    <!-- 指定JournalNode在本地磁盘存放数据的位置 -->
    <property>
        <name>dfs.journalnode.edits.dir</name>
        <value>/main/hadoop-2.8.4/data/journal</value>
    </property>
    <!--指定cluster1出故障时，哪一个实现类负责执行故障切换  -->
    <property>
        <name>dfs.client.failover.proxy.provider.hjbdfs</name>
        <value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
    </property>
    <!--解决HA集群脑裂问题（即出现两个 master 同时对外提供服务，致使系统处于不一致状态）。在 HDFS HA中，JournalNode 只容许一个 NameNode 写数据，不会出现两个 active NameNode 的问题.这是配置自动切换的方法，有多种使用方法，具体能够看官网，在文末会给地址，这里是远程登陆杀死的方法  -->
    <property>
        <name>dfs.ha.fencing.methods</name>
        <value>sshfence(hadoop:2121)</value>
        <description>how to communicate in the switch process</description>
    </property>
    <!-- 这个是使用sshfence隔离机制时才须要配置ssh免登录 -->
    <property>
        <name>dfs.ha.fencing.ssh.private-key-files</name>
        <value>/home/hadoop/.ssh/id_rsa</value>
    </property>
    <!-- 配置sshfence隔离机制超时时间，这个属性同上，若是你是用脚本的方法切换，这个应该是能够不配置的 -->
    <property>
        <name>dfs.ha.fencing.ssh.connect-timeout</name>
        <value>30000</value>
    </property>
    <!-- 这个是开启自动故障转移，若是你没有自动故障转移，这个能够先不配 -->
    <property>
        <name>dfs.ha.automatic-failover.enabled</name>
        <value>true</value>
    </property>
    <property>
        <name>dfs.datanode.data.dir</name>
        <value>/main/hadoop-2.8.4/data/hdfs/datanode</value>
    </property>
    <property>
        <name>dfs.namenode.name.dir</name>
        <value>/main/hadoop-2.8.4/data/hdfs/namenode</value>
    </property>
</configuration>

参考 http://ju.outofmemory.cn/entry/95494

https://www.cnblogs.com/meiyuanbao/p/3545929.html

官方参数: http://hadoop.apache.org/docs/current/hadoop-project-dist/hadoop-hdfs/hdfs-default.xml

修改mapred-site.xml.template名称为mapred-site.xml并修改:

<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<!--
  Licensed under the Apache License, Version 2.0 (the "License");
  you may not use this file except in compliance with the License.
  You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

  Unless required by applicable law or agreed to in writing, software
  distributed under the License is distributed on an "AS IS" BASIS,
  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  See the License for the specific language governing permissions and
  limitations under the License. See accompanying LICENSE file.
-->

<!-- Put site-specific property overrides in this file. -->

<configuration>
    <property>    
        <name>mapreduce.framework.name</name>    
        <value>yarn</value>    
    </property>    
    <property>    
        <name>mapreduce.jobhistory.address</name>    
        <value>node1:10020</value>    
    </property>    
    <property>    
        <name>mapreduce.jobhistory.webapp.address</name>    
        <value>node1:19888</value>    
    </property>    
</configuration>

配置yarn-site.xml:

<?xml version="1.0"?>
<!--
  Licensed under the Apache License, Version 2.0 (the "License");
  you may not use this file except in compliance with the License.
  You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

  Unless required by applicable law or agreed to in writing, software
  distributed under the License is distributed on an "AS IS" BASIS,
  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
  See the License for the specific language governing permissions and
  limitations under the License. See accompanying LICENSE file.
-->
<configuration>
    <property>  
        <name>yarn.nodemanager.aux-services</name>  
        <value>mapreduce_shuffle</value>  
    </property>
    <!-- Site specific YARN configuration properties -->
    <!--启用resourcemanager ha-->  
    <!--是否开启RM ha，默认是开启的-->  
    <property>  
       <name>yarn.resourcemanager.ha.enabled</name>  
       <value>true</value>  
    </property>  
    <!--声明两台resourcemanager的地址-->  
    <property>  
       <name>yarn.resourcemanager.cluster-id</name>  
       <value>rmcluster</value>  
    </property>  
    <property>  
       <name>yarn.resourcemanager.ha.rm-ids</name>  
       <value>rm1,rm2</value>  
    </property>  
    <property>  
       <name>yarn.resourcemanager.hostname.rm1</name>  
       <value>node1</value>  
    </property>  
    <property>  
       <name>yarn.resourcemanager.hostname.rm2</name>  
       <value>node2</value>  
    </property>  
   
    <!--指定zookeeper集群的地址-->   
    <property>  
       <name>yarn.resourcemanager.zk-address</name>  
        <value>192.168.101.173:2181,192.168.101.173:2181,192.168.101.173:2181</value>  
    </property>  
    <!--启用自动恢复，当任务进行一半，rm坏掉，就要启动自动恢复，默认是false-->   
    <property>  
       <name>yarn.resourcemanager.recovery.enabled</name>  
       <value>true</value>  
    </property>  
   
    <!--指定resourcemanager的状态信息存储在zookeeper集群，默认是存放在FileSystem里面。-->   
    <property>  
       <name>yarn.resourcemanager.store.class</name>  
       <value>org.apache.hadoop.yarn.server.resourcemanager.recovery.ZKRMStateStore</value>  
    </property> 

</configuration>

以上基础的都配置完了,打包:

tar -zcvf  hadoop-2.8.4.ready.tar.gz  /main/hadoop-2.8.4

而后用scp把包拷贝到其余4个节点

scp -P 2121 hadoop-2.8.4.ready.tar.gz hadoop@node2:/main

去另外4个节点解压重命名文件夹为/main/hadoop-2.8.4便可

而后继续配置node1 node2这两个master

分别在node1和node2的yarn-site.xml上添加yarn.resourcemanager.ha.id :

相似与zookeeper的myid

<property>  
   <name>yarn.resourcemanager.ha.id</name>  
   <value>rm1</value>  
</property>

<property>  
   <name>yarn.resourcemanager.ha.id</name>  
   <value>rm2</value>  
</property>

启动journal:

三个slave节点启动journal

[hadoop@node5 hadoop-2.8.4]$ /main/hadoop-2.8.4/sbin/hadoop-daemon.sh start journalnode
starting journalnode, logging to /main/hadoop-2.8.4/logs/hadoop-hadoop-journalnode-node5.out
[hadoop@node5 hadoop-2.8.4]$ jps
2272 Jps
2219 JournalNode

启动namenode:

格式化一个master的namenode: journal要起来才能格式化

不能再次格式化或者在另外的节点再次格式化,不然会致使nn和dn的namespaceID不一致而报错!!!

/main/hadoop-2.8.4/bin/hdfs namenode -format

启动一个 namenode:

[hadoop@node1 hadoop]$ /main/hadoop-2.8.4/sbin/hadoop-daemon.sh start namenode
starting namenode, logging to /main/hadoop-2.8.4/logs/hadoop-hadoop-namenode-node1.out
[hadoop@node1 hadoop]$ jps
7536 Jps
7457 NameNode

[hadoop@node1 hadoop]$ ps -ef|grep namenode
hadoop    7457     1 10 09:20 pts/4    00:00:08 /main/jdk1.8.0_171/bin/java -Dproc_namenode -Xmx1000m -Djava.net.preferIPv4Stack=true -Dhadoop.log.dir=/main/hadoop-2.8.4/logs -Dhadoop.log.file=hadoop.log -Dhadoop.home.dir=/main/hadoop-2.8.4 -Dhadoop.id.str=hadoop -Dhadoop.root.logger=INFO,console -Djava.library.path=/main/hadoop-2.8.4/lib/native -Dhadoop.policy.file=hadoop-policy.xml -Djava.net.preferIPv4Stack=true -Djava.net.preferIPv4Stack=true -Djava.net.preferIPv4Stack=true -Dhadoop.log.dir=/main/hadoop-2.8.4/logs -Dhadoop.log.file=hadoop-hadoop-namenode-node1.log -Dhadoop.home.dir=/main/hadoop-2.8.4 -Dhadoop.id.str=hadoop -Dhadoop.root.logger=INFO,RFA -Djava.library.path=/main/hadoop-2.8.4/lib/native -Dhadoop.policy.file=hadoop-policy.xml -Djava.net.preferIPv4Stack=true -Dhadoop.security.logger=INFO,RFAS -Dhdfs.audit.logger=INFO,NullAppender -Dhadoop.security.logger=INFO,RFAS -Dhdfs.audit.logger=INFO,NullAppender -Dhadoop.security.logger=INFO,RFAS -Dhdfs.audit.logger=INFO,NullAppender -Dhadoop.security.logger=INFO,RFAS org.apache.hadoop.hdfs.server.namenode.NameNode

此时应该能够访问node1的hdfs界面,状态为standby:

http://192.168.101.172:50070/

而后另一个master先同步上一个master namenode的元数据信息再启动:

主要是为了同步data/hdfs/namenode/这些信息(包括namespaceID),不然两个节点不一致会报错

/main/hadoop-2.8.4/bin/hdfs namenode -bootstrapStandby

会有成功下在到文件的日志:

18/06/28 11:01:56 WARN common.Util: Path /main/hadoop-2.8.4/data/hdfs/namenode should be specified as a URI in configuration files. Please update hdfs configuration.
18/06/28 11:01:56 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
=====================================================
About to bootstrap Standby ID nn2 from:
           Nameservice ID: hjbdfs
        Other Namenode ID: nn1
  Other NN's HTTP address: http://node1:50070
  Other NN's IPC  address: node1/192.168.101.172:8020
             Namespace ID: 675265321
            Block pool ID: BP-237410497-192.168.101.172-1530153904905
               Cluster ID: CID-604da42a-d0a8-403b-b073-68c857c9b772
           Layout version: -63
       isUpgradeFinalized: true
=====================================================
Re-format filesystem in Storage Directory /main/hadoop-2.8.4/data/hdfs/namenode ? (Y or N) Y
18/06/28 11:02:06 INFO common.Storage: Storage directory /main/hadoop-2.8.4/data/hdfs/namenode has been successfully formatted.
18/06/28 11:02:06 WARN common.Util: Path /main/hadoop-2.8.4/data/hdfs/namenode should be specified as a URI in configuration files. Please update hdfs configuration.
18/06/28 11:02:06 WARN common.Util: Path /main/hadoop-2.8.4/data/hdfs/namenode should be specified as a URI in configuration files. Please update hdfs configuration.
18/06/28 11:02:06 INFO namenode.FSEditLog: Edit logging is async:true
18/06/28 11:02:06 INFO namenode.TransferFsImage: Opening connection to http://node1:50070/imagetransfer?getimage=1&txid=0&storageInfo=-63:675265321:1530153904905:CID-604da42a-d0a8-403b-b073-68c857c9b772&bootstrapstandby=true
18/06/28 11:02:06 INFO namenode.TransferFsImage: Image Transfer timeout configured to 60000 milliseconds
18/06/28 11:02:06 INFO namenode.TransferFsImage: Transfer took 0.01s at 0.00 KB/s
18/06/28 11:02:06 INFO namenode.TransferFsImage: Downloaded file fsimage.ckpt_0000000000000000000 size 323 bytes.
18/06/28 11:02:06 INFO util.ExitUtil: Exiting with status 0
18/06/28 11:02:06 INFO namenode.NameNode: SHUTDOWN_MSG: 
/************************************************************
SHUTDOWN_MSG: Shutting down NameNode at node2/192.168.101.206
************************************************************/

接着启动:

/main/hadoop-2.8.4/sbin/hadoop-daemon.sh start namenode

web界面的状态一样也是standby

手动强制让一个节点变为active的主节点,有问题,会抛出EOFExcption,致使namenode节点挂掉:

[hadoop@node1 hadoop-2.8.4]$ /main/hadoop-2.8.4/bin/hdfs haadmin -transitionToActive nn1 --forcemanual

使用zookeeper自动接管namenode:

先把整个集群关闭，zookeeper不关，输入bin/hdfs zkfc -formatZK，格式化ZKFC

关闭

/main/hadoop-2.8.4/sbin/stop-dfs.sh

启动

/main/hadoop-2.8.4/bin/hdfs zkfc -formatZK

执行完毕以后zookeepr多出来一个hadoop-ha的节点:

启动整个集群:

/main/hadoop-2.8.4/sbin/start-dfs.sh

能够看到这个命令前后启动了namenode datanode journalnode 并启动了zkfc,进行了注册:

[hadoop@node1 hadoop-2.8.4]$ /main/hadoop-2.8.4/sbin/start-dfs.sh 
18/06/28 11:57:59 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
Starting namenodes on [node1 node2]
node1: starting namenode, logging to /main/hadoop-2.8.4/logs/hadoop-hadoop-namenode-node1.out
node2: starting namenode, logging to /main/hadoop-2.8.4/logs/hadoop-hadoop-namenode-node2.out
node3: starting datanode, logging to /main/hadoop-2.8.4/logs/hadoop-hadoop-datanode-node3.out
node4: starting datanode, logging to /main/hadoop-2.8.4/logs/hadoop-hadoop-datanode-node4.out
node5: starting datanode, logging to /main/hadoop-2.8.4/logs/hadoop-hadoop-datanode-node5.out
Starting journal nodes [node3 node4 node5]
node3: starting journalnode, logging to /main/hadoop-2.8.4/logs/hadoop-hadoop-journalnode-node3.out
node4: starting journalnode, logging to /main/hadoop-2.8.4/logs/hadoop-hadoop-journalnode-node4.out
node5: starting journalnode, logging to /main/hadoop-2.8.4/logs/hadoop-hadoop-journalnode-node5.out
18/06/28 11:58:17 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
Starting ZK Failover Controllers on NN hosts [node1 node2]
node1: starting zkfc, logging to /main/hadoop-2.8.4/logs/hadoop-hadoop-zkfc-node1.out
node2: starting zkfc, logging to /main/hadoop-2.8.4/logs/hadoop-hadoop-zkfc-node2.out

能够看到zookeeper上已经注册好了服务:

zookeeper上两个节点的内容:

[zk: localhost:2181(CONNECTED) 10] get /hadoop-ha/hjbdfs/ActiveBreadCrumb

hjbdfsnn2node2 �>(�>
cZxid = 0x30000000a
ctime = Thu Jun 28 11:49:36 CST 2018
mZxid = 0x30000000a
mtime = Thu Jun 28 11:49:36 CST 2018
pZxid = 0x30000000a
cversion = 0
dataVersion = 0
aclVersion = 0
ephemeralOwner = 0x0
dataLength = 26
numChildren = 0

[zk: localhost:2181(CONNECTED) 16] get /hadoop-ha/hjbdfs/ActiveStandbyElectorLock

hjbdfsnn1node1 �>(�>
cZxid = 0x3000001a7
ctime = Thu Jun 28 11:54:13 CST 2018
mZxid = 0x3000001a7
mtime = Thu Jun 28 11:54:13 CST 2018
pZxid = 0x3000001a7
cversion = 0
dataVersion = 0
aclVersion = 0
ephemeralOwner = 0x163f1ef62ad008c
dataLength = 26
numChildren = 0

此时的NameNode:

[hadoop@node1 hadoop-2.8.4]$ jps
14885 NameNode
15191 DFSZKFailoverController
15321 Jps

[hadoop@node2 hadoop-2.8.4]$ jps
18850 NameNode
19059 Jps
18952 DFSZKFailoverController

3个DataNode都同样:

[hadoop@node3 hadoop-2.8.4]$ jps
5409 DataNode
5586 Jps
5507 JournalNode

此时能够发现一个节点是active的,一个节点是standby的.

初步判断安装成功,接下来是测试:

后来重启节点以后,active的节点是node1,开始zookeeper对NameNode的自动故障切换测试:

kill掉主节点Node1的NN进程:

[hadoop@node1 hadoop-2.8.4]$ jps
18850 NameNode 18952 DFSZKFailoverController
19103 Jps

能够看到故障切换的hadoop-hadoop-zkfc-node2.log日志:

2018-06-28 16:07:40,181 INFO org.apache.hadoop.ha.NodeFencer: ====== Beginning Service Fencing Process... ======
2018-06-28 16:07:40,181 INFO org.apache.hadoop.ha.NodeFencer: Trying method 1/1: org.apache.hadoop.ha.SshFenceByTcpPort(hadoop:2121)
2018-06-28 16:07:40,220 INFO org.apache.hadoop.ha.SshFenceByTcpPort: Connecting to node1...
2018-06-28 16:07:40,222 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: Connecting to node1 port 2121
2018-06-28 16:07:40,229 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: Connection established
2018-06-28 16:07:40,236 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: Remote version string: SSH-2.0-OpenSSH_5.3
2018-06-28 16:07:40,236 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: Local version string: SSH-2.0-JSCH-0.1.54
2018-06-28 16:07:40,236 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: CheckCiphers: aes256-ctr,aes192-ctr,aes128-ctr,aes256-cbc,aes192-cbc,aes128-cbc,3des-ctr,arcfour,arcfour128,arcfour256
2018-06-28 16:07:40,635 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: CheckKexes: diffie-hellman-group14-sha1,ecdh-sha2-nistp256,ecdh-sha2-nistp384,ecdh-sha2-nistp521
2018-06-28 16:07:40,725 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: CheckSignatures: ecdsa-sha2-nistp256,ecdsa-sha2-nistp384,ecdsa-sha2-nistp521
2018-06-28 16:07:40,729 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: SSH_MSG_KEXINIT sent
2018-06-28 16:07:40,729 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: SSH_MSG_KEXINIT received
2018-06-28 16:07:40,730 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: server: diffie-hellman-group-exchange-sha256,diffie-hellman-group-exchange-sha1,diffie-hellman-group14-sha1,diffie-hellman-group1-sha1
2018-06-28 16:07:40,730 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: server: ssh-rsa,ssh-dss
2018-06-28 16:07:40,730 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: server: aes128-ctr,aes192-ctr,aes256-ctr,arcfour256,arcfour128,aes128-cbc,3des-cbc,blowfish-cbc,cast128-cbc,aes192-cbc,aes256-cbc,arcfour,rijndael-cbc@lysator.liu.se
2018-06-28 16:07:40,730 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: server: aes128-ctr,aes192-ctr,aes256-ctr,arcfour256,arcfour128,aes128-cbc,3des-cbc,blowfish-cbc,cast128-cbc,aes192-cbc,aes256-cbc,arcfour,rijndael-cbc@lysator.liu.se
2018-06-28 16:07:40,730 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: server: hmac-md5,hmac-sha1,umac-64@openssh.com,hmac-sha2-256,hmac-sha2-512,hmac-ripemd160,hmac-ripemd160@openssh.com,hmac-sha1-96,hmac-md5-96
2018-06-28 16:07:40,730 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: server: hmac-md5,hmac-sha1,umac-64@openssh.com,hmac-sha2-256,hmac-sha2-512,hmac-ripemd160,hmac-ripemd160@openssh.com,hmac-sha1-96,hmac-md5-96
2018-06-28 16:07:40,730 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: server: none,zlib@openssh.com
2018-06-28 16:07:40,731 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: server: none,zlib@openssh.com
2018-06-28 16:07:40,731 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: server: 
2018-06-28 16:07:40,731 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: server: 
2018-06-28 16:07:40,731 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: client: ecdh-sha2-nistp256,ecdh-sha2-nistp384,ecdh-sha2-nistp521,diffie-hellman-group14-sha1,diffie-hellman-group-exchange-sha256,diffie-hellman-group-exchange-sha1,diffie-hellman-group1-sha1
2018-06-28 16:07:40,731 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: client: ssh-rsa,ssh-dss,ecdsa-sha2-nistp256,ecdsa-sha2-nistp384,ecdsa-sha2-nistp521
2018-06-28 16:07:40,731 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: client: aes128-ctr,aes128-cbc,3des-ctr,3des-cbc,blowfish-cbc,aes192-ctr,aes192-cbc,aes256-ctr,aes256-cbc
2018-06-28 16:07:40,731 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: client: aes128-ctr,aes128-cbc,3des-ctr,3des-cbc,blowfish-cbc,aes192-ctr,aes192-cbc,aes256-ctr,aes256-cbc
2018-06-28 16:07:40,731 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: client: hmac-md5,hmac-sha1,hmac-sha2-256,hmac-sha1-96,hmac-md5-96
2018-06-28 16:07:40,731 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: client: hmac-md5,hmac-sha1,hmac-sha2-256,hmac-sha1-96,hmac-md5-96
2018-06-28 16:07:40,731 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: client: none
2018-06-28 16:07:40,732 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: client: none
2018-06-28 16:07:40,732 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: client: 
2018-06-28 16:07:40,732 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: client: 
2018-06-28 16:07:40,732 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: server->client aes128-ctr hmac-md5 none
2018-06-28 16:07:40,732 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: kex: client->server aes128-ctr hmac-md5 none
2018-06-28 16:07:40,769 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: SSH_MSG_KEXDH_INIT sent
2018-06-28 16:07:40,770 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: expecting SSH_MSG_KEXDH_REPLY
2018-06-28 16:07:40,804 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: ssh_rsa_verify: signature true
2018-06-28 16:07:40,811 WARN org.apache.hadoop.ha.SshFenceByTcpPort.jsch: Permanently added 'node1' (RSA) to the list of known hosts.
2018-06-28 16:07:40,811 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: SSH_MSG_NEWKEYS sent
2018-06-28 16:07:40,811 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: SSH_MSG_NEWKEYS received
2018-06-28 16:07:40,817 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: SSH_MSG_SERVICE_REQUEST sent
2018-06-28 16:07:40,818 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: SSH_MSG_SERVICE_ACCEPT received
2018-06-28 16:07:40,820 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: Authentications that can continue: gssapi-with-mic,publickey,keyboard-interactive,password
2018-06-28 16:07:40,820 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: Next authentication method: gssapi-with-mic
2018-06-28 16:07:40,824 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: Authentications that can continue: publickey,keyboard-interactive,password
2018-06-28 16:07:40,824 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: Next authentication method: publickey
2018-06-28 16:07:40,902 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: Authentication succeeded (publickey).
2018-06-28 16:07:40,902 INFO org.apache.hadoop.ha.SshFenceByTcpPort: Connected to node1 2018-06-28 16:07:40,902 INFO org.apache.hadoop.ha.SshFenceByTcpPort: Looking for process running on port 8020
2018-06-28 16:07:40,950 INFO org.apache.hadoop.ha.SshFenceByTcpPort: Indeterminate response from trying to kill service. Verifying whether it is running using nc...
2018-06-28 16:07:40,966 WARN org.apache.hadoop.ha.SshFenceByTcpPort: nc -z node1 8020 via ssh: bash: nc: command not found
2018-06-28 16:07:40,968 INFO org.apache.hadoop.ha.SshFenceByTcpPort: Verified that the service is down. 2018-06-28 16:07:40,968 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: Disconnecting from node1 port 2121
2018-06-28 16:07:40,972 INFO org.apache.hadoop.ha.NodeFencer: ====== Fencing successful by method org.apache.hadoop.ha.SshFenceByTcpPort(hadoop:2121) ======
2018-06-28 16:07:40,972 INFO org.apache.hadoop.ha.ActiveStandbyElector: Writing znode /hadoop-ha/hjbdfs/ActiveBreadCrumb to indicate that the local node is the most recent active...
2018-06-28 16:07:40,973 INFO org.apache.hadoop.ha.SshFenceByTcpPort.jsch: Caught an exception, leaving main loop due to Socket closed
2018-06-28 16:07:40,979 INFO org.apache.hadoop.ha.ZKFailoverController: Trying to make NameNode at node2/192.168.101.206:8020 active... 2018-06-28 16:07:41,805 INFO org.apache.hadoop.ha.ZKFailoverController: Successfully transitioned NameNode at node2/192.168.101.206:8020 to active state

启动Yarn,测试resourcemanager ha ,node1输入:

[hadoop@node1 sbin]$ /main/hadoop-2.8.4/sbin/start-yarn.sh 
starting yarn daemons
starting resourcemanager, logging to /main/hadoop-2.8.4/logs/yarn-hadoop-resourcemanager-node1.out
node3: starting nodemanager, logging to /main/hadoop-2.8.4/logs/yarn-hadoop-nodemanager-node3.out
node4: starting nodemanager, logging to /main/hadoop-2.8.4/logs/yarn-hadoop-nodemanager-node4.out
node5: starting nodemanager, logging to /main/hadoop-2.8.4/logs/yarn-hadoop-nodemanager-node5.out

node2启动resourcemanager:

[hadoop@node2 sbin]$ /main/hadoop-2.8.4/sbin/yarn-daemon.sh start resourcemanager
starting resourcemanager, logging to /main/hadoop-2.8.4/logs/yarn-hadoop-resourcemanager-node2.out

浏览器敲node2的resourcemanager地址http://192.168.101.206:8088 会自动跳转到 http://node1:8088 也就是http://192.168.101.172:8088/cluster

测试HDFS文件上传/下载/删除:

新建了一个 word_i_have_a_dream.txt 里面放的马丁路德金的英文演讲稿.

[hadoop@node1 hadoop-2.8.4]$ /main/hadoop-2.8.4/bin/hadoop fs -put word_i_have_a_dream.txt /word.txt
18/06/28 16:39:28 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable

[hadoop@node1 hadoop-2.8.4]$ /main/hadoop-2.8.4/bin/hadoop fs -ls /word.txt
18/06/28 16:39:43 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
-rw-r--r--   3 hadoop supergroup       4805 2018-06-28 16:39 /word.txt


[hadoop@node1 hadoop-2.8.4]$ /main/hadoop-2.8.4/bin/hadoop fs -rm /word.txt
18/06/28 16:39:59 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
Deleted /word.txt

[hadoop@node1 hadoop-2.8.4]$ /main/hadoop-2.8.4/bin/hadoop fs -ls /word.txt
18/06/28 16:40:07 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
ls: `/word.txt': No such file or directory

下载是get命令,更多见官网或其余博客如https://www.cnblogs.com/lzfhope/p/6952869.html

运行经典wordcount测试:

[hadoop@node1 hadoop-2.8.4]$ /main/hadoop-2.8.4/bin/hadoop jar /main/hadoop-2.8.4/share/hadoop/mapreduce/hadoop-mapreduce-examples-2.8.4.jar wordcount /word.txt /wordoutput
18/06/28 16:50:33 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
18/06/28 16:50:36 INFO input.FileInputFormat: Total input files to process : 1
18/06/28 16:50:36 INFO mapreduce.JobSubmitter: number of splits:1
18/06/28 16:50:36 INFO mapreduce.JobSubmitter: Submitting tokens for job: job_1530173899165_0001
18/06/28 16:50:37 INFO impl.YarnClientImpl: Submitted application application_1530173899165_0001
18/06/28 16:50:37 INFO mapreduce.Job: The url to track the job: http://node1:8088/proxy/application_1530173899165_0001/
18/06/28 16:50:37 INFO mapreduce.Job: Running job: job_1530173899165_0001
18/06/28 16:50:49 INFO mapreduce.Job: Job job_1530173899165_0001 running in uber mode : false
18/06/28 16:50:49 INFO mapreduce.Job:  map 0% reduce 0%
18/06/28 16:50:58 INFO mapreduce.Job:  map 100% reduce 0%
18/06/28 16:51:07 INFO mapreduce.Job:  map 100% reduce 100%
18/06/28 16:51:08 INFO mapreduce.Job: Job job_1530173899165_0001 completed successfully
18/06/28 16:51:08 INFO mapreduce.Job: Counters: 49
    File System Counters
        FILE: Number of bytes read=4659
        FILE: Number of bytes written=330837
        FILE: Number of read operations=0
        FILE: Number of large read operations=0
        FILE: Number of write operations=0
        HDFS: Number of bytes read=4892
        HDFS: Number of bytes written=3231
        HDFS: Number of read operations=6
        HDFS: Number of large read operations=0
        HDFS: Number of write operations=2
    Job Counters 
        Launched map tasks=1
        Launched reduce tasks=1
        Data-local map tasks=1
        Total time spent by all maps in occupied slots (ms)=6187
        Total time spent by all reduces in occupied slots (ms)=6411
        Total time spent by all map tasks (ms)=6187
        Total time spent by all reduce tasks (ms)=6411
        Total vcore-milliseconds taken by all map tasks=6187
        Total vcore-milliseconds taken by all reduce tasks=6411
        Total megabyte-milliseconds taken by all map tasks=6335488
        Total megabyte-milliseconds taken by all reduce tasks=6564864
    Map-Reduce Framework
        Map input records=32
        Map output records=874
        Map output bytes=8256
        Map output materialized bytes=4659
        Input split bytes=87
        Combine input records=874
        Combine output records=359
        Reduce input groups=359
        Reduce shuffle bytes=4659
        Reduce input records=359
        Reduce output records=359
        Spilled Records=718
        Shuffled Maps =1
        Failed Shuffles=0
        Merged Map outputs=1
        GC time elapsed (ms)=194
        CPU time spent (ms)=1860
        Physical memory (bytes) snapshot=444248064
        Virtual memory (bytes) snapshot=4178436096
        Total committed heap usage (bytes)=317718528
    Shuffle Errors
        BAD_ID=0
        CONNECTION=0
        IO_ERROR=0
        WRONG_LENGTH=0
        WRONG_MAP=0
        WRONG_REDUCE=0
    File Input Format Counters 
        Bytes Read=4805
    File Output Format Counters 
        Bytes Written=3231

能够看到HDFS上/wordoutput下生成了输出文件:

[hadoop@node1 hadoop-2.8.4]$ /main/hadoop-2.8.4/bin/hadoop fs -ls /
18/06/28 16:52:32 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
Found 3 items
drwx------   - hadoop supergroup          0 2018-06-28 16:50 /tmp
-rw-r--r--   3 hadoop supergroup       4805 2018-06-28 16:43 /word.txt
drwxr-xr-x   - hadoop supergroup          0 2018-06-28 16:51 /wordoutput

[hadoop@node1 hadoop-2.8.4]$ /main/hadoop-2.8.4/bin/hadoop fs -ls /wordoutput
18/06/28 16:52:49 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
Found 2 items
-rw-r--r--   3 hadoop supergroup          0 2018-06-28 16:51 /wordoutput/_SUCCESS
-rw-r--r--   3 hadoop supergroup       3231 2018-06-28 16:51 /wordoutput/part-r-00000

[hadoop@node1 hadoop-2.8.4]$ /main/hadoop-2.8.4/bin/hadoop fs -ls /wordoutput/_SUCCESS
18/06/28 16:53:05 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
-rw-r--r--   3 hadoop supergroup          0 2018-06-28 16:51 /wordoutput/_SUCCESS

能够看到_SUCCESS大小是0 因此内容在part-r-00000上,撸到本地看一眼:

[hadoop@node1 hadoop-2.8.4]$ /main/hadoop-2.8.4/bin/hadoop fs -ls /wordoutput/part-r-00000
18/06/28 16:54:43 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
-rw-r--r--   3 hadoop supergroup       3231 2018-06-28 16:51 /wordoutput/part-r-00000

[hadoop@node1 hadoop-2.8.4]$ /main/hadoop-2.8.4/bin/hadoop fs -get /wordoutput/part-r-00000 word_success.txt
18/06/28 16:54:57 WARN util.NativeCodeLoader: Unable to load native-hadoop library for your platform... using builtin-java classes where applicable
[hadoop@node1 hadoop-2.8.4]$ vim word_success.txt

生成了咱们想要的结果

接下来试试kill掉NN master测试下HA:

注意:

　　若是报错连接被拒绝,就要去确认以前另外一个master namenode自身的服务是否监听正常,防火墙是否关闭,以及它的hosts文件与hdfs-site.xml文件中namenode的配置是否一致.

　若是配置文件中是主机名,那么每一个节点都要在hosts中映射全部集群节点的ip,自身主机名要映射为内网ip,不能映射为127.0.0.1,不然会致使自身服务绑定在127.0.0.1:xxxx上,局域网其余节点没法连接!!!

　　若是遇到问题须要从新格式化NameNode,须要清理全部节点的老信息,不然会由于老的DataNode节点的namespaceID不一样而没法正确启动:

rm -rf /main/hadoop-2.8.4/data/journal/*
rm -rf /main/hadoop-2.8.4/data/hdfs/namenode/*
rm -rf /main/hadoop-2.8.4/data/hdfs/datanode/*
rm -rf /main/hadoop-2.8.4/logs/*

　　安装若遇到问题,在/main/hadoop-2.8.4/logs目录下的各类*.log文件有详细的日志

HBASE安装

HMaster没有单点问题，HBase中能够启动多个HMaster，经过Zookeeper的Master Election机制保证总有一个Master运行。因此这里要配置HBase高可用的话，只须要启动两个HMaster，让Zookeeper本身去选择一个Master Acitve。

5台机器都解压到/main下:

tar -zvxf /main/soft/hbase-1.2.6.1-bin.tar.gz -C /main/

在Hadoop配置的基础上，配置环境变量HBASE_HOME、hbase-env.sh

vim /etc/profile 设置以下:

#java  以前hadoop时候设置的
export JAVA_HOME=/main/jdk1.8.0_171
export JRE_HOME=$JAVA_HOME/jre
export CLASSPATH=$JAVA_HOME/lib  
export PATH=:$PATH:$JAVA_HOME/bin:$JRE_HOME/bin
#hbase
export HBASE_HOME=/main/hbase-1.2.6.1
export  PATH=$HBASE_HOME/bin:$PATH

HBASE的hbase-env.sh设置以下:

设置javahome 关闭Hbase自带的zk 使用咱们本身安装的zk 同时设置ssh的端口

export JAVA_HOME=/main/jdk1.8.0_171

export HBASE_MANAGES_ZK=false

export HBASE_SSH_OPTS="-p 2121"

设置hbase-site.xml

官网有配置文件说明:

http://abloz.com/hbase/book.html

https://hbase.apache.org/2.0/book.html#example_config

https://hbase.apache.org/2.0/book.html#config.files

<?xml version="1.0"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<configuration>
    <!--命名空间-->
    <property>
        <name>hbase.rootdir</name>
        <!--这里要和hadoop的HDFS的servicename名字一致,不然会报错!-->
        <value>hdfs://hjbdfs/hbase</value>
        <description>The directory shared by RegionServers.</description>
    </property>
    <property>
        <name>hbase.cluster.distributed</name>
        <value>true</value>
    </property>
    <property>
        <name>hbase.master.port</name>
        <value>16000</value>
        <description>The port the HBase Master should bind to.</description>
    </property>
    <property>
        <name>hbase.zookeeper.quorum</name>
        <value>192.168.101.173,192.168.101.183,192.168.101.193</value>
        <description>逗号分割的zk服务器地址
　　　　　　Comma separated list of servers in the ZooKeeper Quorum.
      For example, "host1.mydomain.com,host2.mydomain.com,host3.mydomain.com".
      By default this is set to localhost for local and pseudo-distributed modes
      of operation. For a fully-distributed setup, this should be set to a full
      list of ZooKeeper quorum servers. If HBASE_MANAGES_ZK is set in hbase-env.sh
      this is the list of servers which we will start/stop ZooKeeper on.
      </description>
    </property>
    <property>
        <name>hbase.zookeeper.property.clientPort</name>
        <value>2181</value>
        <description>Property from ZooKeeper's config zoo.cfg.The port at which the clients will connect.</description>
    </property>
    <property>
        <name>hbase.zookeeper.property.dataDir</name>
        <value>/main/zookeeper/data</value>
        <description>zk配置文件zoo.cfg中的data目录地址
      Property from ZooKeeper config zoo.cfg.The directory where the snapshot is stored.</description>
    </property>
    <property>
        <name>hbase.tmp.dir</name>
        <value>/main/hbase-1.2.6.1/hbase/tmp</value>
    </property>
</configuration>

拷贝haddop的配置文件到HBASE目录下,关联两者:

[hadoop@node1 hbase-1.2.6.1]$ cp /main/hadoop-2.8.4/etc/hadoop/core-site.xml /main/hbase-1.2.6.1/conf/
[hadoop@node1 hbase-1.2.6.1]$ cp /main/hadoop-2.8.4/etc/hadoop/hdfs-site.xml /main/hbase-1.2.6.1/conf/

vim regionservers

node3
node4
node5

官网说明:

regionservers: A plain-text file containing a list of hosts which should run a RegionServer in your HBase cluster. By default this file contains the single entry localhost. It should contain a list of hostnames or IP addresses, one per line, and should only contain localhost if each node in your cluster will run a RegionServer on its localhost interface.

vim backup-masters

在node1的机器上写node2 这样的话,之后在node1上start集群,就会把node2做为备份的Master

同理能够在node2上写node1,这样之后能够同时在两个节点进行操做

node2

官网说明:

backup-masters: Not present by default. A plain-text file which lists hosts on which the Master should start a backup Master process, one host per line.

启动HMaster 任意一个master节点都可:

[hadoop@node2 bin]$

/main/hbase-1.2.6.1/bin

/start-hbase.sh starting master, logging to /main/hbase-1.2.6.1/bin/../logs/hbase-hadoop-master-node2.out Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0 Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0 node5: starting regionserver, logging to /main/hbase-1.2.6.1/bin/../logs/hbase-hadoop-regionserver-node5.out node3: starting regionserver, logging to /main/hbase-1.2.6.1/bin/../logs/hbase-hadoop-regionserver-node3.out node4: starting regionserver, logging to /main/hbase-1.2.6.1/bin/../logs/hbase-hadoop-regionserver-node4.out node3: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0 node3: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0 node4: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0 node4: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0 node5: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0 node5: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0 node1: starting master, logging to /main/hbase-1.2.6.1/bin/../logs/hbase-hadoop-master-node1.out node1: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0 node1: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0

在第一个HMaster的节点上执行:

[hadoop@node1 hbase-1.2.6.1]$ /main/hbase-1.2.6.1/bin/start-hbase.sh 
starting master, logging to /main/hbase-1.2.6.1/bin/../logs/hbase-hadoop-master-node1.out
Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0
Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0
node3: starting regionserver, logging to /main/hbase-1.2.6.1/bin/../logs/hbase-hadoop-regionserver-node3.out
node4: starting regionserver, logging to /main/hbase-1.2.6.1/bin/../logs/hbase-hadoop-regionserver-node4.out
node5: starting regionserver, logging to /main/hbase-1.2.6.1/bin/../logs/hbase-hadoop-regionserver-node5.out
node3: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0
node3: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0
node4: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0
node4: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0
node5: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option PermSize=128m; support was removed in 8.0
node5: Java HotSpot(TM) 64-Bit Server VM warning: ignoring option MaxPermSize=128m; support was removed in 8.0

会根据配置文件自动去SSH到RegionServer进行启动,而自身节点若是不是RegionServer,启动完也不会有变化.

如今开始进入第二个HMaster节点,手动再启动一个Hmaster

RegionServer上启动先后的进程变化:

[hadoop@node3 hbase-1.2.6.1]$ jps
7460 DataNode
8247 Jps
7562 JournalNode
7660 NodeManager
[hadoop@node3 hbase-1.2.6.1]$ vim conf/hbase-env.sh 
[hadoop@node3 hbase-1.2.6.1]$ jps
7460 DataNode
8408 Jps
7562 JournalNode
8300 HRegionServer
7660 NodeManager

从节点是node1:

主节点是node2 :

咱们去node2 kill掉进程,测试master的灾备:

[hadoop@node2 bin]$ jps
3809 Jps
1412 NameNode
3607 HMaster
1529 DFSZKFailoverController
[hadoop@node2 bin]$ kill 3607
[hadoop@node2 bin]$ jps
3891 Jps
1412 NameNode
1529 DFSZKFailoverController

node1成功变成了主节点:

能够用./hbase-daemon.sh start master命令再次启动挂掉的master

以上安装完毕以后,node1 node2:

[hadoop@node1 hbase-1.2.6.1]$ jps
31458 NameNode
31779 DFSZKFailoverController
5768 Jps
5482 HMaster
31871 ResourceManager

node3-5:

[hadoop@node3 hbase-1.2.6.1]$ jps
9824 Jps
9616 HRegionServer
7460 DataNode
7562 JournalNode
7660 NodeManager

Spark的安装:

在另一堆集群上,也如上安装了hadoop

192.168.210.114 node1
192.168.210.115 node2
192.168.210.116 node3
192.168.210.117 node4
192.168.210.134 node5
192.168.210.135 node6
192.168.210.136 node7
192.168.210.137 node8

spark安装在1-5上 spark依赖于hadoop:

因此下载的时候须要注意http://spark.apache.org/downloads.html 选择和hadoop兼容的版本:

同时须要下载其对应的scala版本: 去http://spark.apache.org/docs/latest/ 看看读应的scala版本

因此咱们下载了scala 2.11.11 解压到/main下设置好 PATH便可.

下载spark-2.3.1-bin-hadoop2.7.tgz解压到/main后:

编辑 spark-env.sh,根据实际状况设置:

#!/usr/bin/env bash

#
# Licensed to the Apache Software Foundation (ASF) under one or more
# contributor license agreements. See the NOTICE file distributed with
# this work for additional information regarding copyright ownership.
# The ASF licenses this file to You under the Apache License, Version 2.0
# (the "License"); you may not use this file except in compliance with
# the License. You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
#

# This file is sourced when running various Spark programs.
# Copy it as spark-env.sh and edit that to configure Spark for your site.

# Options read when launching programs locally with
# ./bin/run-example or ./bin/spark-submit
# - HADOOP_CONF_DIR, to point Spark towards Hadoop configuration files
# - SPARK_LOCAL_IP, to set the IP address Spark binds to on this node
# - SPARK_PUBLIC_DNS, to set the public dns name of the driver program

# Options read by executors and drivers running inside the cluster
# - SPARK_LOCAL_IP, to set the IP address Spark binds to on this node
# - SPARK_PUBLIC_DNS, to set the public DNS name of the driver program
# - SPARK_LOCAL_DIRS, storage directories to use on this node for shuffle and RDD data
# - MESOS_NATIVE_JAVA_LIBRARY, to point to your libmesos.so if you use Mesos

# Options read in YARN client/cluster mode
# - SPARK_CONF_DIR, Alternate conf dir. (Default: ${SPARK_HOME}/conf)
# - HADOOP_CONF_DIR, to point Spark towards Hadoop configuration files
# - YARN_CONF_DIR, to point Spark towards YARN configuration files when you use YARN
# - SPARK_EXECUTOR_CORES, Number of cores for the executors (Default: 1).
# - SPARK_EXECUTOR_MEMORY, Memory per Executor (e.g. 1000M, 2G) (Default: 1G)
# - SPARK_DRIVER_MEMORY, Memory for Driver (e.g. 1000M, 2G) (Default: 1G)

# Options for the daemons used in the standalone deploy mode
# - SPARK_MASTER_HOST, to bind the master to a different IP address or hostname
# - SPARK_MASTER_PORT / SPARK_MASTER_WEBUI_PORT, to use non-default ports for the master
# - SPARK_MASTER_OPTS, to set config properties only for the master (e.g. "-Dx=y")
# - SPARK_WORKER_CORES, to set the number of cores to use on this machine
# - SPARK_WORKER_MEMORY, to set how much total memory workers have to give executors (e.g. 1000m, 2g)
# - SPARK_WORKER_PORT / SPARK_WORKER_WEBUI_PORT, to use non-default ports for the worker
# - SPARK_WORKER_DIR, to set the working directory of worker processes
# - SPARK_WORKER_OPTS, to set config properties only for the worker (e.g. "-Dx=y")
# - SPARK_DAEMON_MEMORY, to allocate to the master, worker and history server themselves (default: 1g).
# - SPARK_HISTORY_OPTS, to set config properties only for the history server (e.g. "-Dx=y")
# - SPARK_SHUFFLE_OPTS, to set config properties only for the external shuffle service (e.g. "-Dx=y")
# - SPARK_DAEMON_JAVA_OPTS, to set config properties for all daemons (e.g. "-Dx=y")
# - SPARK_DAEMON_CLASSPATH, to set the classpath for all daemons
# - SPARK_PUBLIC_DNS, to set the public dns name of the master or workers

# Generic options for the daemons used in the standalone deploy mode
# - SPARK_CONF_DIR Alternate conf dir. (Default: ${SPARK_HOME}/conf)
# - SPARK_LOG_DIR Where log files are stored. (Default: ${SPARK_HOME}/logs)
# - SPARK_PID_DIR Where the pid file is stored. (Default: /tmp)
# - SPARK_IDENT_STRING A string representing this instance of spark. (Default: $USER)
# - SPARK_NICENESS The scheduling priority for daemons. (Default: 0)
# - SPARK_NO_DAEMONIZE Run the proposed command in the foreground. It will not output a PID file.
# Options for native BLAS, like Intel MKL, OpenBLAS, and so on.
# You might get better performance to enable these options if using native BLAS (see SPARK-21305).
# - MKL_NUM_THREADS=1 Disable multi-threading of Intel MKL
# - OPENBLAS_NUM_THREADS=1 Disable multi-threading of OpenBLAS
export JAVA_HOME=/main/server/jdk1.8.0_11
export SCALA_HOME=/main/scala-2.11.11
export HADOOP_HOME=/main/hadoop-2.8.4
export HADOOP_CONF_DIR=/main/hadoop-2.8.4/etc/hadoop
export SPARK_WORKER_MEMORY=4g
export SPARK_EXECUTOR_MEMORY=4g
export SPARK_DRIVER_MEMORY=4G
export SPARK_WORKER_CORES=2
export SPARK_DAEMON_JAVA_OPTS="-Dspark.deploy.recoveryMode=ZOOKEEPER -Dspark.deploy.zookeeper.url=192.168.210.38:2181,192.168.210.58:2181,192.168.210.78:2181 -Dspark.deploy.zookeeper.dir=/spark"
export SPARK_SSH_OPTS="-p 2121"

vim slaves

node1
node2
node3
node4
node5

而后便可启动,先启动全部的slaves,而后手动启动两个master:

start-all会自动把本身做为master启动,而后再去启动slaves文件中全部的worker

[hadoop@node1 conf]$ /main/spark-2.3.1-bin-hadoop2.7/sbin/start-all.sh 
starting org.apache.spark.deploy.master.Master, logging to /main/spark-2.3.1-bin-hadoop2.7/logs/spark-hadoop-org.apache.spark.deploy.master.Master-1-node1.out
node2: starting org.apache.spark.deploy.worker.Worker, logging to /main/spark-2.3.1-bin-hadoop2.7/logs/spark-hadoop-org.apache.spark.deploy.worker.Worker-1-node2.out
node1: starting org.apache.spark.deploy.worker.Worker, logging to /main/spark-2.3.1-bin-hadoop2.7/logs/spark-hadoop-org.apache.spark.deploy.worker.Worker-1-node1.out
node3: starting org.apache.spark.deploy.worker.Worker, logging to /main/spark-2.3.1-bin-hadoop2.7/logs/spark-hadoop-org.apache.spark.deploy.worker.Worker-1-node3.out
node4: starting org.apache.spark.deploy.worker.Worker, logging to /main/spark-2.3.1-bin-hadoop2.7/logs/spark-hadoop-org.apache.spark.deploy.worker.Worker-1-node4.out
node5: starting org.apache.spark.deploy.worker.Worker, logging to /main/spark-2.3.1-bin-hadoop2.7/logs/spark-hadoop-org.apache.spark.deploy.worker.Worker-1-node5.out

[hadoop@node1 conf]$ jps
11299 Master
11411 Worker
5864 NameNode
6184 DFSZKFailoverController
11802 Jps
6301 ResourceManager
6926 HMaster

其余机器都有worker进程,而后为了HA咱们再去node2启动一个Spark的Master

[hadoop@node2 conf]$ jps
5536 Jps
2209 DFSZKFailoverController
2104 NameNode
2602 HMaster
5486 Worker
[hadoop@node2 conf]$ /main/spark-2.3.1-bin-hadoop2.7/sbin/start-master.sh 
starting org.apache.spark.deploy.master.Master, logging to /main/spark-2.3.1-bin-hadoop2.7/logs/spark-hadoop-org.apache.spark.deploy.master.Master-1-node2.out
[hadoop@node2 conf]$ jps
5568 Master
2209 DFSZKFailoverController
2104 NameNode
2602 HMaster
5486 Worker
5631 Jps

去看看两个master:

另外一个是空的:

Storm安装:

先抄一张图:

下载apache-storm-1.2.2.tar.gz 解压到node6-8节点/main下,设置:

export STORM_HOME=/main/apache-storm-1.2.2

vim storm.yaml 以下,3台相同:

 storm.zookeeper.servers:
     - "192.168.210.38"
     - "192.168.210.58"
     - "192.168.210.78"
 storm.zookeeper.port: 2181
 storm.local.dir: "/main/apache-storm-1.2.2/data"
 nimbus.seeds: ["192.168.210.135"]
 supervisor.slots.ports:
     - 6700
     - 6701
     - 6702
     - 6703

 storm.health.check.dir: "healthchecks"
 storm.health.check.timeout.ms: 5000

其中storm.local.dir指定的目录须要提早建立，supervisor.slots.ports配置的端口数量决定了每台supervisor机器的worker数量，每一个worker会有本身的监听端口用于监放任务。

建立目录:

mkdir  /main/apache-storm-1.2.2/data

主控节点(能够是多个,见nimbus.seeds: ["192.168.210.135"] 配置)

先启动主控:

# 启动主控
nohup /main/apache-storm-1.2.2/bin/storm nimbus &
# 启动主控 ui
nohup /main/apache-storm-1.2.2/bin/storm ui &
# 启动supervisor
nohup /main/apache-storm-1.2.2/bin/storm supervisor &

最后在另外两台机器启动supervisor:

nohup /main/apache-storm-1.2.2/bin/storm supervisor &

主控节点的ui页面能够看到全部supervisor信息

而后从github下载storm源码(或者从storm的apache主页下载src的zip解压),切换到对应的分支,如1.1.x或1.x或2.x. 目前2.x仍是快照版,咱们安装的是1.2.2,切换分支到1.x以后,

特别是github下载的源码,必定要执行一遍:

mvn clean install -DskipTests

由于github上的代码可能更新,相关的jar并无发布到maven仓库,因此上面的命令就编译了一堆storm依赖包到本地mavan仓库.

ps:上述过程不要使用maven第三方镜像,不然极可能会出错!

咱们切换到1.x后发现最新的是storm-starter-1.2.3-SNAPSHOT版本,测试一下服务器的1.2.2是否兼容:

把storm-starter-1.2.3-SNAPSHOT.jar传到服务器,运行一下demo试试:

[hadoop@node6 main]$ /main/apache-storm-1.2.2/bin/storm jar storm-starter-1.2.3-SNAPSHOT.jar org.apache.storm.starter.WordCountTopology word-count
Running: /main/server/jdk1.8.0_11/bin/java -client -Ddaemon.name= -Dstorm.options= -Dstorm.home=/main/apache-storm-1.2.2 -Dstorm.log.dir=/main/apache-storm-1.2.2/logs -Djava.library.path=/usr/local/lib:/opt/local/lib:/usr/lib -Dstorm.conf.file= -cp /main/apache-storm-1.2.2/*:/main/apache-storm-1.2.2/lib/*:/main/apache-storm-1.2.2/extlib/*:storm-starter-1.2.3-SNAPSHOT.jar:/main/apache-storm-1.2.2/conf:/main/apache-storm-1.2.2/bin -Dstorm.jar=storm-starter-1.2.3-SNAPSHOT.jar -Dstorm.dependency.jars= -Dstorm.dependency.artifacts={} org.apache.storm.starter.WordCountTopology word-count
955  [main] WARN  o.a.s.u.Utils - STORM-VERSION new 1.2.2 old null
991  [main] INFO  o.a.s.StormSubmitter - Generated ZooKeeper secret payload for MD5-digest: -7683379793985025786:-5178094576454792625
1122 [main] INFO  o.a.s.u.NimbusClient - Found leader nimbus : node6:6627
1182 [main] INFO  o.a.s.s.a.AuthUtils - Got AutoCreds []
1190 [main] INFO  o.a.s.u.NimbusClient - Found leader nimbus : node6:6627
1250 [main] INFO  o.a.s.StormSubmitter - Uploading dependencies - jars...
1251 [main] INFO  o.a.s.StormSubmitter - Uploading dependencies - artifacts...
1251 [main] INFO  o.a.s.StormSubmitter - Dependency Blob keys - jars : [] / artifacts : []
1289 [main] INFO  o.a.s.StormSubmitter - Uploading topology jar storm-starter-1.2.3-SNAPSHOT.jar to assigned location: /main/apache-storm-1.2.2/data/nimbus/inbox/stormjar-5ed677a5-9af1-4b5e-8467-83f637e00506.jar
Start uploading file 'storm-starter-1.2.3-SNAPSHOT.jar' to '/main/apache-storm-1.2.2/data/nimbus/inbox/stormjar-5ed677a5-9af1-4b5e-8467-83f637e00506.jar' (106526828 bytes)
[==================================================] 106526828 / 106526828
File 'storm-starter-1.2.3-SNAPSHOT.jar' uploaded to '/main/apache-storm-1.2.2/data/nimbus/inbox/stormjar-5ed677a5-9af1-4b5e-8467-83f637e00506.jar' (106526828 bytes)
2876 [main] INFO  o.a.s.StormSubmitter - Successfully uploaded topology jar to assigned location: /main/apache-storm-1.2.2/data/nimbus/inbox/stormjar-5ed677a5-9af1-4b5e-8467-83f637e00506.jar
2876 [main] INFO  o.a.s.StormSubmitter - Submitting topology word-count in distributed mode with conf {"storm.zookeeper.topology.auth.scheme":"digest","storm.zookeeper.topology.auth.payload":"-7683379793985025786:-5178094576454792625","topology.workers":3,"topology.debug":true}
2876 [main] WARN  o.a.s.u.Utils - STORM-VERSION new 1.2.2 old 1.2.2
4091 [main] INFO  o.a.s.StormSubmitter - Finished submitting topology: word-count

显示提交人物完毕,接下来看看UI:

也能够中止这个Topology:

[hadoop@node6 main]$ /main/apache-storm-1.2.2/bin/storm kill word-count
Running: /main/server/jdk1.8.0_11/bin/java -client -Ddaemon.name= -Dstorm.options= -Dstorm.home=/main/apache-storm-1.2.2 -Dstorm.log.dir=/main/apache-storm-1.2.2/logs -Djava.library.path=/usr/local/lib:/opt/local/lib:/usr/lib -Dstorm.conf.file= -cp /main/apache-storm-1.2.2/*:/main/apache-storm-1.2.2/lib/*:/main/apache-storm-1.2.2/extlib/*:/main/apache-storm-1.2.2/extlib-daemon/*:/main/apache-storm-1.2.2/conf:/main/apache-storm-1.2.2/bin org.apache.storm.command.kill_topology word-count