1. HA容灾控制配置

HA配置过程

-----------------

0.选型

挑选指定的namenode。

s101

s105

1.配置s101和s105互相您能够ssh

2.配置文件

[=============core-site.xml============]

<?xml version="1.0"?>

<name>fs.defaultFS</name>

<value>hdfs://mycluster</value>

</property>

<name>dfs.journalnode.edits.dir</name>

<value>/home/centos/hadoop/ha/journalnode</value>

</property>

<name>hadoop.tmp.dir</name>

<value>/home/centos/hadoop/ha</value>

</property>

<name>ha.zookeeper.quorum</name>

</property>

<name>fs.trash.interval</name>

删除文件后在回收站保留的分钟数。默认0，表示禁用回收站。

</description>

</property>

<name>fs.trash.checkpoint.interval</name>

<description>两次检查回收站的间隔数(默认0分钟),0表示和fs.trash.interval相同。</description>

</property>

<name>hadoop.http.staticuser.user</name>

<value>centos</value>

</property>

<!--

<name>net.topology.node.switch.mapping.impl</name>

<value>com.it18zhang.hadoop.rackaware.MyDNSToSwitchMapping</value>

</property>

-->

<name>hadoop.proxyuser.centos.hosts</name>

</property>

<name>hadoop.proxyuser.centos.groups</name>

</property>

</configuration>

原始配置文件

<?xml version="1.0"?>

<name>fs.defaultFS</name>

<value>hdfs://mycluster</value>

</property>

<name>dfs.journalnode.edits.dir</name>

<value>/home/centos/ha/dfs/journal/node/local/data</value>

</property>

<name>hadoop.tmp.dir</name>

<value>/home/centos/ha</value>

</property>

<name>ha.zookeeper.quorum</name>

</property>

<name>hadoop.proxyuser.hue.hosts</name>

</property>

<name>hadoop.proxyuser.hue.groups</name>

</property>

<name>hadoop.proxyuser.httpfs.hosts</name>

</property>

<name>hadoop.proxyuser.httpfs.groups</name>

</property>

</configuration>

[=============hdfs-site.xml============]

<?xml version="1.0" encoding="UTF-8"?>

<?xml-stylesheet type="text/xsl" href="configuration.xsl"?>

<name>dfs.nameservices</name>

<value>mycluster</value>

</property>

<name>dfs.ha.namenodes.mycluster</name>

</property>

<name>dfs.namenode.rpc-address.mycluster.nn1</name>

</property>

<name>dfs.namenode.rpc-address.mycluster.nn2</name>

</property>

<name>dfs.namenode.http-address.mycluster.nn1</name>

</property>

<name>dfs.namenode.http-address.mycluster.nn2</name>

</property>

<name>dfs.namenode.shared.edits.dir</name>

<value>qjournal://s102:8485;s103:8485;s104:8485/mycluster</value>

</property>

<name>dfs.client.failover.proxy.provider.mycluster</name>

<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>

</property>

<name>dfs.ha.fencing.methods</name>

<value>

sshfence

shell(/bin/true)

</value>

</property>

<name>dfs.ha.fencing.ssh.private-key-files</name>

<value>/home/centos/.ssh/id_rsa</value>

</property>

<name>dfs.ha.automatic-failover.enabled</name>

</property>

<name>dfs.replication</name>

</property>

<name>dfs.namenode.secondary.http-address</name>

</property>

<name>dfs.namenode.name.dir</name>

<value>file://${hadoop.tmp.dir}/dfs/name1,file://${hadoop.tmp.dir}/dfs/name2</value>

</property>

<name>dfs.datanode.data.dir</name>

<value>file://${hadoop.tmp.dir}/dfs/data1,file://${hadoop.tmp.dir}/dfs/data2</value>

</property>

<name>dfs.namenode.fs-limits.min-block-size</name>

</property>

<name>dfs.hosts</name>

<value>/soft/hadoop/etc/hadoop/dfs_include.conf</value>

</property>

<name>dfs.hosts.exclude</name>

<value>/soft/hadoop/etc/hadoop/dfs_exclude.conf</value>

</property>

<name>dfs.webhdfs.enabled</name>

</property>

<name>dfs.permissions.enabled</name>

<value>false</value>

</property>

</configuration>

原始的配置文件

<?xml version="1.0"?>

<name>dfs.replication</name>

</property>

<name>dfs.namenode.secondary.http-address</name>

</property>

<name>dfs.namenode.name.dir</name>

<value>/home/centos/ha/dfs/name1,/home/centos/ha/dfs/name2</value>

</property>

<name>dfs.datanode.data.dir</name>

<value>/home/centos/ha/dfs/data1,/home/centos/ha/dfs/data2</value>

</property>

<name>dfs.nameservices</name>

<value>mycluster</value>

</property>

<name>dfs.ha.namenodes.mycluster</name>

</property>

<name>dfs.namenode.rpc-address.mycluster.nn1</name>

</property>

<name>dfs.namenode.rpc-address.mycluster.nn2</name>

</property>

<name>dfs.namenode.http-address.mycluster.nn1</name>

</property>

<name>dfs.namenode.http-address.mycluster.nn2</name>

</property>

<name>dfs.namenode.shared.edits.dir</name>

<value>qjournal://s102:8485;s103:8485;s104:8485/mycluster</value>

</property>

<name>dfs.client.failover.proxy.provider.mycluster</name>

<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>

</property>

<name>dfs.ha.fencing.methods</name>

<value>shell(/bin/true)</value>

</property>

<name>dfs.ha.automatic-failover.enabled</name>

</property>

</configuration>

3.分发所有节点

xsync.sh core-site.xml

xsync.sh hdfs-site.xml

重新搭建ha集群

-----------------

1.停止hadoop集群

$>stop-dfs.sh

2.删除所有log和本地文件

$>xcall.sh rm -rf /soft/hadoop/logs/*

$>xcall.sh rm -rf /home/centos/hadoop/ha/*

3.登录jn节点启动jn进程

[s102 ~ s104]

$>hadoop-daemon.sh start journalnode

4.格式化文件系统

[s101]

hdfs namenode -format

名称节点上的集群版本

数据节点上的集群版本

5.同步101名称节点数据到105

[s101]

scp -r /home/centos/hadoop/ha centos@s105:/home/centos/hadoop

6.在s105上进行引导

6.1)需要先启动s101的namenode

[s101]

hadoop-daemon.sh start namenode

6.2)执行bootstrap

[s105]

hdfs namenode -bootstrapStandby

注意：不要重新格式化 N

7)初始化编辑日志到jn集群

[没有启动nn的节点上执行]

$>hdfs namenode -initializeSharedEdits

注意：N

8)使用管理命令

#切换状态

$>hdfs haadmin -transitionToActive nn1 --forcemanual

#查询状态

$>hdfs haadmin -getServiceState nn1

#容灾演练

$>hdfs haadmin -failover nn2

9)支持自动容灾配置

9.1)引入两个组件

a)zk quorum

b)zkfc

zkfc是nn节点伴随节点(贴身保镖)，有一下三个职责:

1.监控nn的健康状况

周期ping，收达响应说明健康，否则不健康。

2.zk session管理

3.zk选举

9.2)配置自动容灾

略

9.3)格式化zkfc

$>hdfs zkfc -formatZK

hadoop-daemon.sh start zkfc

测试

posted @ 2018-09-25 20:09 star521 阅读(242) 评论(0) 编辑收藏举报

刷新页面返回顶部

star521

1. HA容灾控制配置

公告