HDFS的HA
一、集群的规划
Zookeeper集群:
192.168.223.112 (bigdata112)
192.168.223.113 (bigdata113)
192.168.223.114 (bigdata114)
Hadoop集群:
192.168.223.112 (bigdata112) NameNode1 ResourceManager1 Journalnode
192.168.223.113 (bigdata113) NameNode2 ResourceManager2 Journalnode
192.168.223.114 (bigdata114) DataNode1 NodeManager1
192.168.223.115 (bigdata115) DataNode2 NodeManager2
二、准备工作
- 安装JDK
- 配置环境变量
- 配置免密码登录
- 配置主机名
三、配置Zookeeper
在主节点(bigdata112)上配置ZooKeeper
(*)配置/root/training/zookeeper-3.4.6/conf/zoo.cfg文件
dataDir=/root/training/zookeeper-3.4.6/tmp
server.1=bigdata112:2888:3888
server.2=bigdata113:2888:3888
server.3=bigdata114:2888:3888
(*)在/root/training/zookeeper-3.4.6/tmp目录下创建一个myid的空文件
echo 1 > /root/training/zookeeper-3.4.6/tmp/myid
(*)将配置好的zookeeper拷贝到其他节点,同时修改各自的myid文件
scp -r /root/training/zookeeper-3.4.6/ bigdata113:/root/training
scp -r /root/training/zookeeper-3.4.6/ bigdata114:/root/training
四、安装Hadoop集群(在bigdata112上安装)
修改hadoo-env.sh
export JAVA_HOME=/root/training/jdk1.8.0_144
修改core-site.xml
<configuration>
<!-- 指定hdfs的nameservice为ns1 -->
<property>
<name>fs.defaultFS</name>
<value>hdfs://ns1</value>
</property>
<!-- 指定hadoop临时目录 -->
<property>
<name>hadoop.tmp.dir</name>
<value>/root/training/hadoop-2.7.3/tmp</value>
</property>
<!-- 指定zookeeper地址 -->
<property>
<name>ha.zookeeper.quorum</name>
<value>bigdata112:2181,bigdata113:2181,bigdata114:2181</value>
</property>
</configuration>
修改hdfs-site.xml(配置这个nameservice中有几个namenode)
<configuration>
<!--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>bigdata112:9000</value>
</property>
<!-- nn1的http通信地址 -->
<property>
<name>dfs.namenode.http-address.ns1.nn1</name>
<value>bigdata112:50070</value>
</property>
<!-- nn2的RPC通信地址 -->
<property>
<name>dfs.namenode.rpc-address.ns1.nn2</name>
<value>bigdata113:9000</value>
</property>
<!-- nn2的http通信地址 -->
<property>
<name>dfs.namenode.http-address.ns1.nn2</name>
<value>bigdata113:50070</value>
</property>
<!-- 指定NameNode的日志在JournalNode上的存放位置 -->
<property>
<name>dfs.namenode.shared.edits.dir</name>
<value>qjournal://bigdata112:8485;bigdata113:8485;/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>/root/training/hadoop-2.7.3/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>
<!-- 使用sshfence隔离机制时需要ssh免登陆 -->
<property>
<name>dfs.ha.fencing.ssh.private-key-files</name>
<value>/root/.ssh/id_rsa</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>
</configuration>
修改mapred-site.xml
<configuration>
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
</configuration>
修改yarn-site.xml
<configuration>
<!-- 开启RM高可靠 -->
<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>bigdata112</value>
</property>
<property>
<name>yarn.resourcemanager.hostname.rm2</name>
<value>bigdata113</value>
</property>
<!-- 指定zk集群地址 -->
<property>
<name>yarn.resourcemanager.zk-address</name>
<value>bigdata112:2181,bigdata113:2181,bigdata114:2181</value>
</property>
<!--MapReduce运行方式:shuffle洗牌-->
<property>
<name>yarn.nodemanager.aux-services</name>
<value>mapreduce_shuffle</value>
</property>
</configuration>
修改slaves
bigdata114
bigdata115
将配置好的hadoop拷贝到其他节点
scp -r /root/training/hadoop-2.7.3/ root@bigdata113:/root/training/
scp -r /root/training/hadoop-2.7.3/ root@bigdata114:/root/training/
scp -r /root/training/hadoop-2.7.3/ root@bigdata115:/root/training/
五、启动Zookeeper集群
六、在bigdata112和bigdata113上启动journalnode
hadoop-daemon.sh start journalnode
七、格式化HDFS(在bigdata112上执行)
1. hdfs namenode -format
2. 将/root/training/hadoop-2.7.3/tmp拷贝到bigdata113的/root/training/hadoop-2.7.3/tmp下
3. 格式化zookeeper
hdfs zkfc -formatZK
日志:17/07/13 00:34:33 INFO ha.ActiveStandbyElector: Successfully created /hadoop-ha/ns1 in ZK
八、在bigdata112上启动Hadoop集群
start-all.sh
bigdata113上的ResourceManager需要单独启动
命令:yarn-daemon.sh start resourcemanager
HDFS的联盟
规划
NameNode:bigdata112 bigdata113
DataNode: bigdata114 bigdata115
hadoop-env.sh JAVA_HOME /root/training/jdk1.8.0_144
core-site.xml
<property>
<name>hadoop.tmp.dir</name>
<value>/root/training/hadoop-2.7.3/tmp</value>
</property>
mapred-site.xml
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
yarn-site.xml
<property>
<name>yarn.resourcemanager.hostname</name>
<value>bigdata112</value>
</property>
<property>
<name>yarn.nodemanager.aux-services</name>
<value>mapreduce_shuffle</value>
</property>
slaves
bigdata114
bigdata115
hdfs-site.xml
<property>
<name>dfs.nameservices</name>
<value>ns1,ns2</value>
</property>
<property>
<name>dfs.namenode.rpc-address.ns1</name>
<value>192.168.223.112:9000</value>
</property>
<property>
<name>dfs.namenode.http-address.ns1</name>
<value>192.168.223.112:50070</value>
</property>
<property>
<name>dfs.namenode.secondaryhttp-address.ns1</name>
<value>192.168.223.112:50090</value>
</property>
<property>
<name>dfs.namenode.rpc-address.ns2</name>
<value>192.168.223.113:9000</value>
</property>
<property>
<name>dfs.namenode.http-address.ns2</name>
<value>192.168.223.113:50070</value>
</property>
<property>
<name>dfs.namenode.secondaryhttp-address.ns2</name>
<value>192.168.223.113:50090</value>
</property>
<property>
<name>dfs.replication</name>
<value>2</value>
</property>
<property>
<name>dfs.permissions</name>
<value>false</value>
</property>
增加路由规则-----> core-site.xml中
<property>
<name>fs.viewfs.mounttable.xdl1.homedir</name>
<value>/home</value>
</property>
<property>
<name>fs.viewfs.mounttable.xdl1.link./movie</name>
<value>hdfs://192.168.223.112:9000/movie</value>
</property>
<property>
<name>fs.viewfs.mounttable.xdl1.link./mp3</name>
<value>hdfs://192.168.223.113:9000/mp3</value>
</property>
<property>
<name>fs.default.name</name>
<value>viewfs://xdl1</value>
</property>
注意:如果路由规则太多了,不方便维护 ----> 单独创建一个xml文件(mountTable.xml)
参考官方的文档:http://hadoop.apache.org/docs/r2.7.6/hadoop-project-dist/hadoop-hdfs/Federation.html
http://hadoop.apache.org/docs/r2.7.6/hadoop-project-dist/hadoop-hdfs/ViewFs.html
复制到其他节点
scp -r hadoop-2.7.3/ root@bigdata113:/root/training
scp -r hadoop-2.7.3/ root@bigdata114:/root/training
scp -r hadoop-2.7.3/ root@bigdata115:/root/training
对每个NameNode(bigdata112 bigdata113)进行格式刷
注意ID号
hdfs namenode -format -clusterId xdl1
启动
在对应NameNode上,事先创建号目录
hadoop fs -mkdir hdfs://192.168.223.112:9000/movie
hadoop fs -mkdir hdfs://192.168.223.113:9000/mp3
一定注意
[root@bigdata112 training]# hdfs dfs -ls / 查看到的是路由规则,不是目录
Found 2 items
-r-xr-xr-x - root root 0 2018-05-11 20:39 /movie
-r-xr-xr-x - root root 0 2018-05-11 20:39 /mp3