环境变量参考【CentOS6-大数据套件HA安装(1)统一环境配置)】
此处所有机器防火墙关闭,实际可根据需要调整。软件包统一在/usr/local/soft 目录,安装目录为:/hadoop
安装hadoop
- 解压
tar -xzvf /usr/local/soft/hadoop-2.7.6.tar.gz -C /hadoop/
- 创建tmp、journal目录
mkdir /hadoop/hadoop-2.7.6/tmp
mkdir -p /hadoop/hadoop-2.7.6/dfs/name
mkdir -p /hadoop/hadoop-2.7.6/dfs/data
mkdir /hadoop/hadoop-2.7.6/journal
vi hadoop-2.7.6/etc/hadoop/hadoop-env.sh
export JAVA_HOME=/usr/java/jdk1.8.0_172-amd64
- 配置 yarn-env.sh
vi hadoop-2.7.6/etc/hadoop/yarn-env.sh
export JAVA_HOME=/usr/java/jdk1.8.0_172-amd64
- 配置 core-site.xml
vi hadoop-2.7.6/etc/hadoop/core-site.xml
<configuration>
<!--
<property>
<name>fs.default.name</name>
<value>hdfs://manager202:9000</value>
</property>
-->
<property>
<name>fs.defaultFS</name>
<value>hdfs://ns1</value>
</property>
<property>
<name>hadoop.tmp.dir</name>
<value>/hadoop/hadoop-2.7.6/tmp</value>
<description>Abase for other temporary directories.</description>
</property>
<property>
<name>ha.zookeeper.quorum</name>
<value>hadoop101:2181,hadoop102:2181,hadoop103:2181</value>
</property>
<!--
<property>
<name>io.file.buffer.size</name>
<value>131072</value>
</property
<property>
<name>hadoop.proxyuser.hduser.hosts</name>
<value>*</value>
</property>
<property>
<name>hadoop.proxyuser.hduser.groups</name>
<value>*</value>
</property>
-->
</configuration>
- 配置 hdfs-site.xml
vi hadoop-2.7.6/etc/hadoop/hdfs-site.xml
<configuration>
<property>
<name>dfs.name.dir</name>
<value>/hadoop/hadoop-2.7.6/dfs/name</value>
</property>
<property>
<name>dfs.data.dir</name>
<value>/hadoop/hadoop-2.7.6/dfs/data</value>
</property>
<property>
<name>dfs.replication</name>
<value>3</value>
</property>
<!--指定hdfs的nameservice为ns1,需要和core-site.xml中的保持一致 -->
<property>
<name>dfs.nameservices</name>
<value>ns1</value>
</property>
<!-- ns1下面有两个NameNode,分别是nn1,nn2 -->
<property>
<name>dfs.ha.namenodes.ns1</name>
<value>nn1,nn2</value>
</property>
<!-- nn1的RPC通信地址 -->
<property>
<name>dfs.namenode.rpc-address.ns1.nn1</name>
<value>manager202:9000</value>
</property>
<!-- nn1的http通信地址 -->
<property>
<name>dfs.namenode.http-address.ns1.nn1</name>
<value>manager202:50070</value>
</property>
<!-- nn2的RPC通信地址 -->
<property>
<name>dfs.namenode.rpc-address.ns1.nn2</name>
<value>manager203:9000</value>
</property>
<!-- nn2的http通信地址 -->
<property>
<name>dfs.namenode.http-address.ns1.nn2</name>
<value>manager203:50070</value>
</property>
<!-- 指定NameNode的元数据在JournalNode上的存放位置 -->
<property>
<name>dfs.namenode.shared.edits.dir</name>
<value>qjournal://hadoop101:8485;hadoop102:8485;hadoop103:8485/ns1</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>/hadoop/hadoop-2.7.6/journal</value>
</property>
<!-- 开启NameNode失败自动切换 -->
<property>
<name>dfs.ha.automatic-failover.enabled</name>
<value>true</value>
</property>
<!-- 配置失败自动切换实现方式 -->
<property>
<name>dfs.client.failover.proxy.provider.ns1</name>
<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
</property>
<!-- 配置隔离机制 -->
<property>
<name>dfs.ha.fencing.methods</name>
<value>sshfence</value>
</property>
<!-- 使用隔离机制时需要ssh免登陆 -->
<property>
<name>dfs.ha.fencing.ssh.private-key-files</name>
<value>/home/hadoop/.ssh/id_rsa</value>
</property>
</configuration>
- 配置 mapred-site.xml
vi hadoop-2.7.6/etc/hadoop/mapred-site.xml
<configuration>
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
</configuration>
- 配置 yarn-site.xml
vi hadoop-2.7.6/etc/hadoop/yarn-site.xml
<configuration>
<!-- 开启RM高可用 -->
<property>
<name>yarn.resourcemanager.ha.enabled</name>
<value>true</value>
</property>
<!-- 指定RM的cluster id -->
<property>
<name>yarn.resourcemanager.cluster-id</name>
<value>yrc</value>
</property>
<!-- 指定RM的名字 -->
<property>
<name>yarn.resourcemanager.ha.rm-ids</name>
<value>rm1,rm2</value>
</property>
<!-- 分别指定RM的地址 -->
<property>
<name>yarn.resourcemanager.hostname.rm1</name>
<value>manager202</value>
</property>
<property>
<name>yarn.resourcemanager.hostname.rm2</name>
<value>manager203</value>
</property>
<!-- 指定zk集群地址 -->
<property>
<name>yarn.resourcemanager.zk-address</name>
<value>hadoop101:2181,hadoop102:2181,hadoop103:2181</value>
</property>
<property>
<name>yarn.nodemanager.aux-services</name>
<value>mapreduce_shuffle</value>
</property>
<!-- 指定可用内存和CPU -->
<property>
<name>yarn.nodemanager.resource.memory-mb</name>
<value>12288</value>
</property>
<property>
<name>yarn.nodemanager.resource.cpu-vcores</name>
<value>6</value>
</property>
</configuration>
- 配置 slaves
vi hadoop-2.7.6/etc/hadoop/slaves
hadoop101
hadoop102
hadoop103
- 配置
pid
存储位置
# 默认会放在tmp目录,但是超过10天以后会删除,导致调用stop脚本时无法关闭的问题
vi hadoop-daemon.sh
HADOOP_PID_DIR=/data/hadoop/pid
vi yarn-daemon.sh
YARN_PID_DIR=/data/hadoop/pid
vi mr-jobhistory-daemon.sh
HADOOP_MAPRED_PID_DIR=/data/hadoop/pid
- 向各节点分发Hadoop程序
scp -r hadoop-2.7.6 hadoop@manager203:/hadoop/
scp -r hadoop-2.7.6 hadoop@hadoop101:/hadoop/
scp -r hadoop-2.7.6 hadoop@hadoop102:/hadoop/
scp -r hadoop-2.7.6 hadoop@hadoop103:/hadoop/
启动或者初始化必须确认先启动了zookeeper
初始化hadoop
- 启动 journalnode (执行一次)
# 分别在hadoop101、hadoop102、hadoop103上执行
./hadoop-2.7.6/sbin/hadoop-daemon.sh start journalnode
- 格式化 HDFS (执行一次)
# 在manager202上执行命令:
./hadoop-2.7.6/bin/hdfs namenode -format
# 然后将 /hadoop/hadoop-2.7.6/dfs/name/* 拷贝到manager203的/hadoop/hadoop-2.7.6/dfs/name/下
scp -r hadoop-2.7.6/dfs/name/* hadoop@manager203:/hadoop/hadoop-2.7.6/dfs/name/
##也可以这样,建议hdfs namenode -bootstrapStandby
- 格式化 ZKFC (执行一次)
# 在manager202上执行
./hadoop-2.7.6/bin/hdfs zkfc -formatZK
启动
- 启动 HDFS
# 在manager202上执行
./hadoop-2.7.6/sbin/start-dfs.sh
- 启动 YARN
# 在manager202上执行
./hadoop-2.7.6/sbin/start-yarn.sh
# manager203上执行
./hadoop-2.7.6/sbin/yarn-daemon.sh start resourcemanager
- 测试集群工作状态的一些指令
# 查看hdfs的各节点状态信息
./hadoop-2.7.6/bin/hdfs dfsadmin -report
# 查看namenode高可用
./hadoop-2.7.6/bin/hdfs haadmin -getServiceState nn1
# 查看ResourceManager高可用
./hadoop-2.7.6/bin/yarn rmadmin –getServiceState rm1
# 单独启动一个namenode进程
./hadoop-2.7.6/sbin/hadoop-daemon.sh start namenode
# 单独启动一个zkfc进程
./hadoop-2.7.6/hadoop-daemon.sh start zkfc