2.联邦模式配置---扩容，负载均衡
原理图 两个集群---目的：扩容
HA联邦模式解决了单纯HA模式的性能瓶颈（主要指Namenode、ResourceManager），将整个HA集群划分为两个以上的集群，不同的集群之间通过Federation进行连接，使得HA集群拥有了横向扩展的能力。理论上，在该模式下，能够通过增加计算节点以处理无限增长的数据。联邦模式下的配置在原HA模式的基础上做了部分调整。
配置过程
federation
cp -r local/ federation
1.规划集群
ns1:nn1(s101) + nn2(s102)
ns2:nn3(s103) + nn4(s014)
2.准备
[nn1 ~ nn4 ]ssh 所有节点.
3.停止整个集群
4.配置文件
4.1)s101和s102的hdfs-site.xml配置
[hadoop/federation/hdfs-site.xml]
<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<configuration>
<property>
<name>dfs.nameservices</name>
<value>ns1,ns2</value>
</property>
<!-- **************ns1********************* -->
<property>
<name>dfs.ha.namenodes.ns1</name>
<value>nn1,nn2</value>
</property>
<property>
<name>dfs.namenode.rpc-address.ns1.nn1</name>
<value>s101:8020</value>
</property>
<property>
<name>dfs.namenode.rpc-address.ns1.nn2</name>
<value>s102:8020</value>
</property>
<property>
<name>dfs.namenode.http-address.ns1.nn1</name>
<value>s101:50070</value>
</property>
<property>
<name>dfs.namenode.http-address.ns1.nn2</name>
<value>s102:50070</value>
</property>
<property>
<name>dfs.client.failover.proxy.provider.ns1</name>
<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
</property>
<!-- **************ns2********************* -->
<property>
<name>dfs.ha.namenodes.ns2</name>
<value>nn3,nn4</value>
</property>
<property>
<name>dfs.namenode.rpc-address.ns2.nn3</name>
<value>s103:8020</value>
</property>
<property>
<name>dfs.namenode.rpc-address.ns2.nn4</name>
<value>s104:8020</value>
</property>
<property>
<name>dfs.namenode.http-address.ns2.nn3</name>
<value>s103:50070</value>
</property>
<property>
<name>dfs.namenode.http-address.ns2.nn4</name>
<value>s104:50070</value>
</property>
<property>
<name>dfs.client.failover.proxy.provider.ns2</name>
<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
</property>
<!--***********************************************-->
<property>
<name>dfs.namenode.shared.edits.dir</name>
<value>qjournal://s102:8485;s103:8485;s104:8485/ns1</value>
</property>
<property>
<name>dfs.ha.fencing.methods</name>
<value>
sshfence
shell(/bin/true)
</value>
</property>
<property>
<name>dfs.ha.fencing.ssh.private-key-files</name>
<value>/home/centos/.ssh/id_rsa</value>
</property>
<property>
<name>dfs.ha.automatic-failover.enabled</name>
<value>true</value>
</property>
<property>
<name>dfs.replication</name>
<value>3</value>
</property>
</configuration>
4.2)s103和s104的hdfs-site.xml配置
[hadoop/federation/hdfs-site.xml]
<?xml version="1.0" encoding="UTF-8"?>
<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>
<configuration>
<property>
<name>dfs.nameservices</name>
<value>ns1,ns2</value>
</property>
<!-- **************ns1********************* -->
<property>
<name>dfs.ha.namenodes.ns1</name>
<value>nn1,nn2</value>
</property>
<property>
<name>dfs.namenode.rpc-address.ns1.nn1</name>
<value>s101:8020</value>
</property>
<property>
<name>dfs.namenode.rpc-address.ns1.nn2</name>
<value>s102:8020</value>
</property>
<property>
<name>dfs.namenode.http-address.ns1.nn1</name>
<value>s101:50070</value>
</property>
<property>
<name>dfs.namenode.http-address.ns1.nn2</name>
<value>s102:50070</value>
</property>
<property>
<name>dfs.client.failover.proxy.provider.ns1</name>
<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
</property>
<!-- **************ns2********************* -->
<property>
<name>dfs.ha.namenodes.ns2</name>
<value>nn3,nn4</value>
</property>
<property>
<name>dfs.namenode.rpc-address.ns2.nn3</name>
<value>s103:8020</value>
</property>
<property>
<name>dfs.namenode.rpc-address.ns2.nn4</name>
<value>s104:8020</value>
</property>
<property>
<name>dfs.namenode.http-address.ns2.nn3</name>
<value>s103:50070</value>
</property>
<property>
<name>dfs.namenode.http-address.ns2.nn4</name>
<value>s104:50070</value>
</property>
<property>
<name>dfs.client.failover.proxy.provider.ns2</name>
<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
</property>
<!--***********************************************-->
<property>
<name>dfs.namenode.shared.edits.dir</name>
<value>qjournal://s102:8485;s103:8485;s104:8485/ns2</value>
</property>
<property>
<name>dfs.ha.fencing.methods</name>
<value>
sshfence
shell(/bin/true)
</value>
</property>
<property>
<name>dfs.ha.fencing.ssh.private-key-files</name>
<value>/home/centos/.ssh/id_rsa</value>
</property>
<property>
<name>dfs.ha.automatic-failover.enabled</name>
<value>true</value>
</property>
<property>
<name>dfs.replication</name>
<value>3</value>
</property>
</configuration>
4.3)s101 ~ s104的core-site.xml配置文件
[hadoop/federation/core-site.xml]
<?xml version="1.0"?>
<configuration xmlns:xi="http://www.w3.org/2001/XInclude">
<xi:include href="mountTable.xml" />
<property>
<name>fs.defaultFS</name>
<value>viewfs://ClusterX</value>
</property>
<property>
<name>dfs.journalnode.edits.dir</name>
<value>/home/centos/hadoop/federation/journalnode</value>
</property>
<property>
<name>hadoop.tmp.dir</name>
<value>/home/centos/hadoop/federation</value>
</property>
<property>
<name>ha.zookeeper.quorum</name>
<value>s102:2181,s103:2181,s104:2181</value>
</property>
</configuration>
4.4)mountTable.xml 挂载表文件
[hadoop/federation/mountTable.xml]
<configuration>
<property>
<name>fs.viewfs.mounttable.ClusterX.homedir</name>
<value>/home</value>
</property>
<property>
<name>fs.viewfs.mounttable.ClusterX.link./home</name>
<value>hdfs://ns1/home</value>
</property>
<property>
<name>fs.viewfs.mounttable.ClusterX.link./tmp</name>
<value>hdfs://ns2/tmp</value>
</property>
<property>
<name>fs.viewfs.mounttable.ClusterX.link./projects/foo</name>
<value>hdfs://ns1/projects/foo</value>
</property>
<property>
<name>fs.viewfs.mounttable.ClusterX.link./projects/bar</name>
<value>hdfs://ns2/projects/bar</value>
</property>
</configuration>
5.操作
5.1)删除所有节点的日志和本地临时目录
$>xcall.sh rm -rf /soft/hadoop/logs/*
$>xcall.sh rm -rf /home/centos/hadoop/federation/*
5.2)修改所有节点的hadoop软连接
$>xcall.sh ln -sfT /soft/hadoop/etc/federation /soft/hadoop/etc/hadoop
5.3)对ns1集群进行格式化以及初始工作
a)启动jn集群
登录s102 ~ s104，启动jounalnode进程。
$>hadoop-daemon.sh start journalnode
b)格式化nn1节点
[s101]
$>hdfs namenode -format
c)复制s101的元数据到s102下.
[s101]
$>scp -r
d)在s102上执行引导过程
#s101启动名称节点
$>hadoop-daemon.sh start namenode
# s102执行引导,不要重格(N)
$>hdfs namenode -bootstrapStandby
e)在s102上初始化编辑日志到jn集群(N)
$>hdfs namenode -initializeSharedEdits
f)在s102对zookeeper格式化zkfc(选择Y).
$>hdfs zkfc -formatZK
g)启动s101和s102的namenode和zkfc进程。
[s101]
$>hadoop-daemon.sh start zkfc
[s102]
$>hadoop-daemon.sh start namenode
$>hadoop-daemon.sh start zkfc
h)测试webui
5.4)对ns2集群进行格式化以及初始工作
a)格式nn3，切记使用-clusterId属性，保持和ns1的一致。
[s103]
$>hdfs namenode -format -clusterId CID-e16c5e2f-c0a5-4e51-b789-008e36b7289a
b)复制s103的元数据到s104上。
$>scp -r /home/centos/hadoop/federation centos@s104:/home/centos/hadoop/
c)在s104引导
#在s103启动namenode
$>hadoop-daemon.sh start namenode
#在s104执行引导
$>hdfs namenode -bootstrapStandby
d)在s104上初始化编辑日志
$>hdfs namenode -initializeSharedEdits
e)在s104对zookeeper格式化zkfc(选择Y).
$>hdfs zkfc -formatZK
f)启动s103和s104的namenode和zkfc进程。
[s103]
$>hadoop-daemon.sh start zkfc
[s104]
$>hadoop-daemon.sh start namenode
$>hadoop-daemon.sh start zkfc
5.5)停止集群
$>stop-dfs.sh
5.6)重启dfs集群
$>start-dfs.sh
5.7)创建目录
# 注意加p参数
$>hdfs dfs -mkdir -p /home/data
#上传文件,考察webui
$>hdfs dfs -put 1.txt /home/data
