hadoop 3.0安装
节点角色分配：
vm1 namenode1 --- zookeeper1 ZKFC1 jornalnode1 resourcemanager
vm2 namenode2 datanode1 zookeeper2 ZKFC2 jornalnode2 resourcemanager
vm3 --- datanode2 zookeeper3 --- jornalnode3 ---
https://blog.csdn.net/qq_31583183/article/details/89575981
zookeeper下载：http://archive.apache.org/dist/zookeeper/
1 安装java环境
tar -xf jdk-8u131-linux-x64.tar.gz -C /usr/local/
vi /etc/profile
export JAVA_HOME=/usr/local/jdk1.8.0_131
export CLASS_PATH=$JAVA_HOME/bin/lib:$JAVA_HOME/jre/lib:$JAVA_HOME/lib/tool.jar
export PATH=${JAVA_HOME}/bin:$PATH
source /etc/profile
2 ssh互信，免密登录
192.168.1.84 vm1 namenode
192.168.1.85 vm2 datnode
192.168.1.86 vm3 datanode
创建hadoop用户：(namenode和datanode均需要)
useradd hadoop
passwd hadoop
在namenode上生成秘钥：
# su - hadoop
# ssh-keygen
在/home/hadoop/.ssh/目录下生成：id_rsa（私钥）和id_rsa.pub（公钥）
将公钥id_rsa.pub上传至各datanode节点上
在各datanode节点上：（复制namenode上的id_rsa.pub到authorized_keys）
# vi /home/hadoop/.ssh/authorized_keys
文件权限一定是600：
chmod 700 /home/hadoop/.ssh
chmod 600 /home/hadoop/.ssh/authorized_keys
3 zookeeper集群搭建
zookeeper下载：http://archive.apache.org/dist/zookeeper/
https://www.aboutyun.com/thread-26880-1-1.html
（3台zk机器均需要如下操作）
下载解压：
tar -xf zookeeper-3.3.5.tar.gz -C /root/hdfs/
cd zookeeper-3.3.5/
创建zookeeper的存放数据的文件夹：
mkdir data
配置zookeeper的环境变量：
vim /etc/profile
export JAVA_HOME=/usr/local/jdk1.8.0_131
export PATH=$PATH:$JAVA_HOME/bin
export CLASSPATH=.:$JAVA_HOME/lib/tools.jar:$JAVA_HOME/lib/dt.jar
export ZOOKEEPER_HOME=/root/hdfs/zookeeper-3.3.5
export PATH=$PATH:$JAVA_HOME/bin:$ZOOKEEPER_HOME/bin
修改配置文件：
cd /root/hdfs/zookeeper-3.3.5/conf
先把zoo_sample.cfg改为zoo.cfg
mv zoo_sample.cfg zoo.cfg
#修改的dataDir文件路径，配置临时文件路径内容即可，配置的文件路径需要提前创建好
# example sakes.
dataDir=/root/hdfs/zookeeper-3.3.5/data
# the port at which the clients will connect
clientPort=2181
# 配置server 几台机器就配置几台即可，注意server的数字，我们在后续需要用到
server.1=vm1:2888:3888
server.2=vm2:2888:3888
server.3=vm3:2888:3888
(以上配置在其它zk节点同样操作)
创建需要的文件夹:
vm1机器上：echo 1 > /root/hdfs/zookeeper-3.3.5/data/myid
vm2机器上：echo 2 > /root/hdfs/zookeeper-3.3.5/data/myid
vm3机器上：echo 3 > /root/hdfs/zookeeper-3.3.5/data/myid
启动并查看状态(3台zk均需启动):
zkServer.sh start
zkServer.sh status
# zkServer.sh status
JMX enabled by default
Using config: /root/hdfs/zookeeper-3.3.5/bin/../conf/zoo.cfg
Mode: follower
-----
Mode: leader
4 安装hadoop
# su - hadoop
# cd /home/hadoop/
下载hadoop包：
https://archive.apache.org/dist/hadoop/common/
https://mirrors.cnnic.cn/apache/hadoop/common/
tar -xf hadoop-3.2.2.tar.gz
cd /home/hadoop/hadoop-3.2.2/etc/hadoop/
修改配置：
vi hadoop-env.sh
export JAVA_HOME=/usr/local/jdk1.8.0_131
export HDFS_NAMENODE_USER=root
export HDFS_DATANODE_USER=root
export HDFS_SECONDARYNAMENODE_USER=root
export HDFS_ZKFC_USER=root
export HDFS_JOURNALNODE_USER=root
vi core-site.xml
<configuration>
<property>
<name>fs.defaultFS</name>
<value>hdfs://ns1</value>
</property>
<property>
<name>dfs.nameservices</name>
<value>ns1</value>
</property>
<property>
<name>hadoop.tmp.dir</name>
<value>/hadoop/tmp</value>
</property>
<property>
<name>ha.zookeeper.quorum</name>
<value>vm1:2181,vm1:2181,vm1:2181</value>
</property>
</configuration>
vi hdfs-site.xml
<configuration>
<property>
<name>dfs.replication</name>
<value>3</value>
</property>
<property>
<name>dfs.nameservices</name>
<value>ns1</value>
</property>
<property>
<name>dfs.ha.namenodes.ns1</name>
<value>nn1,nn2</value>
</property>
<property>
<name>dfs.namenode.rpc-address.ns1.nn1</name>
<value>vm1:8020</value>
</property>
<property>
<name>dfs.namenode.rpc-address.ns1.nn2</name>
<value>vm2:8020</value>
</property>
<property>
<name>dfs.namenode.http-address.ns1.nn1</name>
<value>vm1:50070</value>
</property>
<property>
<name>dfs.namenode.http-address.ns1.nn2</name>
<value>vm2:50070</value>
</property>
<property>
<name>dfs.namenode.shared.edits.dir</name>
<value>qjournal://vm1:8485;vm2:8485;vm3:8485/ns1</value>
</property>
<property>
<name>dfs.client.failover.proxy.provider.ns1</name>
<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
</property>
<property>
<name>dfs.ha.fencing.methods</name>
<value>sshfence</value>
</property>
<property>
<name>dfs.ha.fencing.ssh.private-key-files</name>
<value>/root/.ssh/id_rsa</value>
</property>
<property>
<name>dfs.name.dir</name>
<value>/hadoop/namedata</value>
</property>
<property>
<name>dfs.data.dir</name>
<value>/hadoop/data</value>
</property>
<property>
<name>dfs.journalnode.edits.dir</name>
<value>/hadoop/journalnode</value>
</property>
<property>
<name>dfs.ha.automatic-failover.enabled</name>
<value>true</value>
</property>
</configuration>
vi mapred-site.xml
<configuration>
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
</configuration>
vi yar-site.xml
<configuration>
<!-- Site specific YARN configuration properties -->
<property>
<name>yarn.resourcemanager.ha.enabled</name>
<value>true</value>
</property>
<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>vm1</value>
</property>
<property>
<name>yarn.resourcemanager.hostname.rm2</name>
<value>vm2</value>
</property>
<property>
<name>yarn.resourcemanager.zk-address</name>
<value>vm1:2181,vm2:2181,vm3:2181</value>
</property>
<property>
<name>yarn.resourcemanager.recovery.enabled</name>
<value>true</value>
</property>
<property>
<name>yarn.resourcemanager.store.class</name>
<value>org.apache.hadoop.yarn.server.resourcemanager.recovery.ZKRMStateStore</value>
</property>
</configuration>
vi /home/hadoop/hadoop-3.2.2/etc/hadoop/workers
vm2
vm3
启动hadoop:
(1)journalnode 节点上执行
hadoop/sbin/hadoop-daemon.sh start journalnode
使用jps 能看到 JournalNode
(2)在第一个namenode上格式化namenode，格式化zk，启动namenode
bin/hdfs namenode -format
bin/hdfs zkfc -formatZK
sbin/hadoop-daemon.sh start namenode
bin/hdfs --daemon start zkfc
jps 命令可以看到 NameNode DFSZKFailoverController
(3)在第二个namenode上，同步数据，启动第二个namenode
bin/hdfs namenode -bootstrapStandby
sbin/hadoop-daemon.sh start namenode
bin/hdfs --daemon start zkfc
jps 会发现有namenode DFSZKFailoverController
(4)所有datanode上启动 datanode
sbin/hadoop-daemon.sh start datanode
jps 会发现有datanode
查看namenode的状态：
# hdfs haadmin -getAllServiceState
vm1:8020 active
vm2:8020 standby
# hdfs haadmin -getServiceState nn1
active
# hdfs haadmin -getServiceState nn2
standby
ps -ef |grep hadoop
root 107962 1 3 09:27 pts/3 00:01:07 /usr/local/jdk1.8.0_131/bin/java -Dproc_journalnode -Djava.net.preferIPv4Stack=true -Dyarn.log.dir=/root/hdfs/hadoop-3.2.2/logs -Dyarn.log.file=hadoop-root-journalnode-vm1.log -Dyarn.home.dir=/root/hdfs/hadoop-3.2.2 -Dyarn.root.logger=INFO,console -Djava.library.path=/root/hdfs/hadoop-3.2.2/lib/native -Dhadoop.log.dir=/root/hdfs/hadoop-3.2.2/logs -Dhadoop.log.file=hadoop-root-journalnode-vm1.log -Dhadoop.home.dir=/root/hdfs/hadoop-3.2.2 -Dhadoop.id.str=root -Dhadoop.root.logger=INFO,RFA -Dhadoop.policy.file=hadoop-policy.xml -Dhadoop.security.logger=INFO,NullAppender org.apache.hadoop.hdfs.qjournal.server.JournalNode
root 109743 1 6 09:31 pts/3 00:01:52 /usr/local/jdk1.8.0_131/bin/java -Dproc_namenode -Djava.net.preferIPv4Stack=true -Dhdfs.audit.logger=INFO,NullAppender -hadoop.security.logger=INFO,RFAS -Dyarn.log.dir=/root/hdfs/hadoop-3.2.2/logs -Dyarn.log.file=hadoop-root-namenode-vm1.log -Dyarn.home.dir=/root/hdfs/hadoop-3.2.2 -Dyarn.root.logger=INFO,console -Djava.library.path=/root/hdfs/hadoop-3.2.2/lib/native -Dhadoop.log.dir=/root/hdfs/hadoop-3.2.2/logs -Dhadoop.log.file=hadoop-root-namenode-vm1.log -Dhadoop.home.dir=/root/hdfs/hadoop-3.2.2 -Dhadoop.id.str=root -Dhadoop.root.logger=INFO,RFA -Dhadoop.policy.file=hadoop-policy.xml org.apache.hadoop.hdfs.server.namenode.NameNode
root 111512 1 3 09:34 pts/3 00:01:01 /usr/local/jdk1.8.0_131/bin/java -Dproc_zkfc -Djava.net.preferIPv4Stack=true -Dyarn.log.dir=/root/hdfs/hadoop-3.2.2/logs -Dyarn.log.file=hadoop-root-zkfc-vm1.log -Dyarn.home.dir=/root/hdfs/hadoop-3.2.2 -Dyarn.root.logger=INFO,console -Djava.library.path=/root/hdfs/hadoop-3.2.2/lib/native -Dhadoop.log.dir=/root/hdfs/hadoop-3.2.2/logs -Dhadoop.log.file=hadoop-root-zkfc-vm1.log -Dhadoop.home.dir=/root/hdfs/hadoop-3.2.2 -Dhadoop.id.str=root -Dhadoop.root.logger=INFO,RFA -Dhadoop.policy.file=hadoop-policy.xml -Dhadoop.security.logger=INFO,NullAppender org.apache.hadoop.hdfs.tools.DFSZKFailoverController
root 71116 1 7 09:45 pts/2 00:01:18 /usr/local/jdk1.8.0_131/bin/java -Dproc_datanode -Djava.net.preferIPv4Stack=true -Dhadoop.security.logger=ERROR,RFAS -Dyarn.log.dir=/root/hdfs/hadoop-3.2.2/logs -Dyarn.log.file=hadoop-root-datanode-vm3.log -Dyarn.home.dir=/root/hdfs/hadoop-3.2.2 -Dyarn.root.logger=INFO,console -Djava.library.path=/root/hdfs/hadoop-3.2.2/lib/native -Dhadoop.log.dir=/root/hdfs/hadoop-3.2.2/logs -Dhadoop.log.file=hadoop-root-datanode-vm3.log -Dhadoop.home.dir=/root/hdfs/hadoop-3.2.2 -Dhadoop.id.str=root -Dhadoop.root.logger=INFO,RFA -Dhadoop.policy.file=hadoop-policy.xml org.apache.hadoop.hdfs.server.datanode.DataNode
