楔子
Hadoop的学习 学习可以参考 此处视频
学习过程中 HDFS搭建还是基于Hadoop3 ,
但是在YARN的搭建就变为了 2.6.5版本 ?
环境节点分配
NN-1 | NN-2 | DN(DataNode) | ZK | ZKFC | JNN(JournalNode) | RS(ResourceManager) | NM(NodeManager) | |
---|---|---|---|---|---|---|---|---|
had1 | * | * | * | |||||
had2 | * | * | * | * | * | * | ||
had3 | * | * | * | * | * | |||
had4 | * | * | * | * |
配置
1 配置在$HADOOP_HOME/etc/hadoop目录下面
[root@had1 hadoop]# cd $HADOOP_HOME
2 修改五个配置文件
hdfs配置需要
- hadoop-env.sh
- core-site.xml
- hdfs-site.xml
yarn配置
- yarn-site.xml
- mapred-site.xml
修改workers(添加节点的主机名称)
had2
had3
had4
3 配置免密登录
ssh-keygen #数个回车键
# 将公钥拷贝到其他节点,包括本机
ssh-copy-id root@had1
ssh-copy-id root@had2
ssh-copy-id root@had3
ssh-copy-id root@had4
YARN启动
之前已经启动过 hdfs 此处单单是启动yarn
start-yarn.sh
启动情况
HA测试
Overview 'had2:8020' (active)# 表示had2节点的namenode处于 active
Overview 'had1:8020' (standby)# 表示had1节点的namenode处于 备用
kill 进程ID
hdfs --daemon start namenode#启动namenode
YARN HA测试
kill 进程ID
yarn --daemon start resourcemanager #启动
问题
启动yarn遇到如下问题
[root@had1 hadoop]# start-yarn.sh
Starting resourcemanagers on [ had3 had4]
ERROR: Attempting to operate on yarn resourcemanager as root
ERROR: but there is no YARN_RESOURCEMANAGER_USER defined. Aborting operation.
Starting nodemanagers
ERROR: Attempting to operate on yarn nodemanager as root
ERROR: but there is no YARN_NODEMANAGER_USER defined. Aborting operation.
在hadoop-env.sh添加相关用户
# yarn需要
export YARN_RESOURCEMANAGER_USER=root
export YARN_NODEMANAGER_USER=root
全部配置文件内容
配置文件可以从此处下载
1 hadoop-env.sh
export JAVA_HOME=/usr/local/soft/jdk
export HDFS_NAMENODE_USER=root
export HDFS_DATANODE_USER=root
#export HDFS_SECONDARYNAMENODE_USER=root
export HDFS_ZKFC_USER=root
export HDFS_JOURNALNODE_USER=root
# yarn需要
export YARN_RESOURCEMANAGER_USER=root
export YARN_NODEMANAGER_USER=root
2 core-site.xml
<configuration>
<!-- 指定hadoop临时目录,默认在/tmp/{$user}目录下,不安全,每次开机都会被清空-->
<property>
<name>hadoop.tmp.dir</name>
<value>/opt/data/hadoop_ha</value>
</property>
<property>
<name>hadoop.http.staticuser.user</name>
<value>root</value>
</property>
<!-- 指定zookeeper地址 -->
<property>
<name>ha.zookeeper.quorum</name>
<value>had2:2181,had3:2181,had4:2181</value>
</property>
<!-- 指定hdfs的nameservice名称空间为nameservices 与hdfs-site.xml (dfs.nameservices)保持一致 -->
<property>
<name>fs.defaultFS</name>
<value>hdfs://mycluster</value>
</property>
</configuration>
3 hdfs-site.xml
<configuration>
<!--文件副本数量 -->
<property>
<name>dfs.replication</name>
<value>2</value>
</property>
<!-- NamoNode HA配置 -->
<!-- core-site.xml 保持一致 -->
<property>
<name>dfs.nameservices</name>
<value>mycluster</value>
<description>指定hdfs的nameservice为 mycluster ,需要和core-site.xml中的保持一致</description>
</property>
<property>
<name>dfs.ha.namenodes.mycluster</name>
<value>nn1,nn2</value>
<description>ns命名空间下有两个NameNode,逻辑代号,随便起名字,分别是nn1,nn2</description>
</property>
<!-- nn1的RPC通信地址 -->
<property>
<name>dfs.namenode.rpc-address.mycluster.nn1</name>
<value>had1:8020</value>
</property>
<!-- nn2的RPC通信地址 -->
<property>
<name>dfs.namenode.rpc-address.mycluster.nn2</name>
<value>had2:8020</value>
</property>
<!-- n1的http通信地址 -->
<property>
<name>dfs.namenode.http-address.mycluster.nn1</name>
<value>had1:9870</value>
</property>
<property>
<name>dfs.namenode.http-address.mycluster.nn2</name>
<value>had2:9870</value>
</property>
<!-- JNN(JournalNode)节点配置 -->
<property>
<name>dfs.namenode.shared.edits.dir</name>
<value>qjournal://had1:8485;had2:8485;had3:8485/mycluster</value>
</property>
<!-- 故障转移代理类 -->
<!-- 配置失败自动切换实现方式,使用内置的zkfc -->
<property>
<name>dfs.client.failover.proxy.provider.mycluster</name>
<value>org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider</value>
</property>
<property>
<name>dfs.ha.fencing.methods</name>
<value>sshfence</value>
<description>配置隔离机制,多个机制用换行分割,先执行sshfence,执行失败后执行shell(/bin/true),/bin/true会直接返回0表示成功</description>
</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>
<!-- 指定JournalNode在本地磁盘存放数据的位置 -->
<property>
<name>dfs.journalnode.edits.dir</name>
<value>/opt/data/hadoopha/journal</value>
</property>
<!--namenode高可用主备切换配置 -->
<property>
<name>dfs.ha.automatic-failover.enabled</name>
<value>true</value>
</property>
</configuration>
4 yarn-site.xml
<configuration>
<!--NodeManager上运行的附属服务,需配置成mapreduce_shuffle,才可运行MapReduce程序-->
<property>
<name>yarn.nodemanager.aux-services</name>
<value>mapreduce_shuffle</value>
</property>
<!-- 开启RM高可用 -->
<property>
<name>yarn.resourcemanager.ha.enabled</name>
<value>true</value>
</property>
<!-- 指定RM的cluster id,一组高可用的rm共同的逻辑id -->
<property>
<name>yarn.resourcemanager.cluster-id</name>
<value>yarn-ha</value>
</property>
<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>had3</value>
</property>
<property>
<name>yarn.resourcemanager.hostname.rm2</name>
<value>had4</value>
</property>
<!-- HTTP访问的端口号 -->
<property>
<name>yarn.resourcemanager.webapp.address.rm1</name>
<value>had3:8088</value>
</property>
<property>
<name>yarn.resourcemanager.webapp.address.rm2</name>
<value>had4:8088</value>
</property>
<property>
<name>yarn.resourcemanager.zk-address</name>
<value>had2:2181,had3:2181,had4:2181</value>
</property>
<!-- 开启日志聚合 -->
<property>
<name>yarn.log-aggregation-enable</name>
<value>true</value>
</property>
<!-- 日志聚合HDFS目录 -->
<property>
<name>yarn.nodemanager.remote-app-log-dir</name>
<value>/data/hadoop/yarn-logs</value>
</property>
<!-- 日志保存时间3days,单位秒 -->
<property>
<name>yarn.log-aggregation.retain-seconds</name>
<value>259200</value>
</property>
</configuration>
5 mapred-site.xml
<configuration>
<!-- 指定mr框架为yarn方式 -->
<property>
<name>mapreduce.framework.name</name>
<value>yarn</value>
</property>
<!-- 历史日志服务jobhistory相关配置 -->
<property>
<name>mapreduce.jobhistory.address</name>
<value>had3:10020</value>
<description>历史服务器端口号</description>
</property>
<property>
<name>mapreduce.jobhistory.webapp.address</name>
<value>had3:19888</value>
<description>历史服务器的WEB UI端口号</description>
</property>
<property>
<name>mapreduce.jobhistory.joblist.cache.size</name>
<value>2000</value>
<description>内存中缓存的historyfile文件信息(主要是job对应的文件目录)</description>
</property>
</configuration>