Hadoop集群搭建ha高可用

• 1.Hadoop集群准备ok
• 2.zookeeper配置完成
• 3.vi core-site.xml
  name集群的名称
  zookeeper2181地址

<configuration>
  <property>
    <name>fs.defaultFS</name>
    <value>hdfs://ns</value>
  </property>
  <property>
    <name>hadoop.tmp.dir</name>
    <value>/opt/hadoop/tmp</value>
  </property>
  <!-- 指定zookeeper地址 -->
  <property>
    <name>ha.zookeeper.quorum</name>
    <value>hadoop001:2181,hadoop002:2181,hadoop003:2181</value>
  </property>
  <property>
    <name>hadoop.proxyuser.root.hosts</name>
    <value>*</value>
  </property>
  <property>
    <name>hadoop.proxyuser.root.groups</name>
    <value>*</value>
  </property>
  <property>
    <name>hadoop.native.lib</name>
    <value>false</value>
    <description>Should native hadoop libraries, if present, be used.
    </description>
  </property>
</configuration>

• 4.vi hdfs-site.xml
  nameservices集群下面的各个nameservice服务对应的映射
  rpc-address的地址9000
  http-address的地址50070

  journalnode的共享地址(各个datanode)
  qjournal://datanode:8485;列表/集群名

  出错处理，切换方式及隔离机制的设置

<configuration>
  <!-- 节点数，ha中可配置 -->
  <property>
    <name>dfs.replication</name>
    <value>3</value>
  </property>
  <!-- 指定名称服务，与core-site.xml中的一致 -->
  <property>
    <name>dfs.nameservices</name>
    <value>ns</value>
  </property>
  <!-- 指定名称服务下的namenode节点 -->
  <property>
    <name>dfs.ha.namenodes.ns</name>
    <value>hadoop001,hadoop002</value>
  </property>
  <!-- rpc地址，9000端口 -->
  <property>
    <name>dfs.namenode.rpc-address.ns.hadoop001</name>
    <value>hadoop001:9000</value>
  </property>
  <property>
    <name>dfs.namenode.rpc-address.ns.hadoop002</name>
    <value>hadoop002:9000</value>
  </property>
  <!-- http地址，50070端口 -->
  <property>
    <name>dfs.namenode.http-address.ns.hadoop001</name>
    <value>hadoop001:50070</value>
  </property>
  <property>
    <name>dfs.namenode.http-address.ns.hadoop002</name>
    <value>hadoop002:50070</value>
  </property>
</property>
  <!-- namenode在jornalnode（datanode）上存放数据的端口8485 -->
  <property>
    <name>dfs.namenode.shared.edits.dir</name>
    <value>qjournal://hadoop001:8485;hadoop002:8485;hadoop003:8485/ns</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.ns</name>
    <value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
  </property>
  <!-- 配置隔离机制方法，多个机制用换行分割，即每个机制暂用一行 -->
  <property>
    <name>dfs.ha.fencing.methods</name>
    <value>
      sshfence
      shell(/bin/true)
    </value>
  </property>
  <!-- 配置sshfence隔离机制超时时间 -->
    <property>
      <name>dfs.ha.fencing.ssh.connect-timeout</name>
      <value>30000</value>
    </property>
    <property>
      <name>dfs.webhdfs.enabled</name>
      <value>true</value>
    </property>
</configuration>

• 5.vi yarn-site.xml
  允许ha
  resourcemanager集群配置
  rm-ids:rm1,rm2
  rm1:nm1 及端口配置8088
  rm2:nm2 及端口配置8088
  zookeeper配置

<configuration>
<!--  HA配置 -->
<!-- 开启HA -->
  <property>
    <name>yarn.resourcemanager.ha.enabled</name>
    <value>hadoop001</value>
  </property>
<!-- 指定rm集群id -->
  <property>
    <name>yarn.resourcemanager.cluser-id</name>
    <value>cluster_id</value>
  </property>
<!-- rm集群的名称服务 -->
  <property>
    <name>yarn.resourcemanager.ha.rm-ids</name>
    <value>rm1,rm2</value>
  </property>
<!-- 指定rm映射关系 -->
  <property>
    <name>yarn.resourcemanager.hostname.rm1</name>
    <value>hadoop001</value>
  </property>
  <property>
    <name>yarn.resourcemanager.hostname.rm2</name>
    <value>hadoop002</value>
  </property>
  <property>
    <name>yarn.resourcemanager.webapp.address.rm1</name>
    <value>hadoop001:8088</value>
  </property>
  <property>
    <name>yarn.resourcemanager.webapp.address.rm2</name>
    <value>hadoop002:8088</value>
  </property>
<!-- 配置zookepper集群 -->
  <property>
    <name>yarn.resourcemanager.zk-address</name>
    <value>hadoop001:2181,hadoop002:2181,hadoop003:2181</value>
  </property>

<!-- Site specific YARN configuration properties -->
<!-- 非HA模式下指定resourcemanager
  <property>
    <name>yarn.resourcemanager.hadoop001</name>
    <value>hadoop001</value>
  </property>
-->
  <property>
    <name>yarn.nodemanager.aux-services</name>
    <value>mapreduce_shuffle</value>
  </property>
  <property>
    <name>yarn.nodemanager.aux-services.mapreduce.shuffle.class</name>
    <value>org.apache.hadoop.mapred.ShuffleHandler</value>
  </property>
  <!-- 日志聚集功能使用 -->
  <property>
    <name>yarn.log-aggregation-enable</name>
    <value>true</value>
  </property>
  <!-- 日志保留时间设置7天 -->
  <property>
    <name>yarn.log-aggregation.retain-seconds</name>
    <value>604800</value>
  </property>
</configuration>

• 6.保证三台机器一致

scp *.xml root@其他机器主机名:$PWD

• 7.确保集群处于关闭状态
  jps查看当前启用服务

• 8.删除tmp文件夹,选择节点重新format(可能需要提前启动journalnode)

hadoop-daemon.sh start journalnode

• 9.把主节点新生成的tmp文件夹复制到备用节点

scp -r tmp root@备用节点主机名:$PWD

• 10.zookeeper要启动

zkServer.sh start

• 11.启动整个集群

start-all.sh

• 12.在备用节点上启动备用resourcemanager

yarn-daemon.sh start resourcemanager

• 13.主节点格式化zkfc

hdfs zkfc -formatZK

• 14.主备节点启动zkfc

hadoop-daemon.sh start zkfc

• 15.datanode节点启动历史服务

mr-jobhistory-daemon.sh start historyserver

可以再次通过jps查看已开服务