数据治理之集群迁移数据
准备两套集群,我这使用apache集群和CDH集群。
启动集群
启动完毕后,将apache集群中,hive库里dwd,dws,ads三个库的数据迁移到CDH集群
在apache集群里hosts加上CDH Namenode对应域名并分发给各机器
[root@hadoop101 ~]# vim /etc/hosts
[root@hadoop101 ~]# scp /etc/hosts hadoop102:/etc/
[root@hadoop101 ~]# scp /etc/hosts hadoop103:/etc/
因为集群都是HA模式,所以需要在apache集群上配置CDH集群,让distcp能识别出CDH的nameservice
[root@hadoop101 hadoop]# vim /opt/module/hadoop-3.1.3/etc/hadoop/hdfs-site.xml
dfs.nameservices
mycluster,nameservice1
dfs.internal.nameservices
mycluster
dfs.ha.namenodes.mycluster
nn1,nn2,nn3
dfs.namenode.rpc-address.mycluster.nn1
hadoop101:8020
dfs.namenode.rpc-address.mycluster.nn2
hadoop102:8020
dfs.namenode.rpc-address.mycluster.nn3
hadoop103:8020
dfs.ha.namenodes.nameservice1
namenode30,namenode37
dfs.namenode.rpc-address.nameservice1.namenode30
hadoop104:8020
dfs.namenode.rpc-address.nameservice1.namenode37
hadoop106:8020
dfs.namenode.http-address.nameservice1.namenode30
hadoop104:9870
dfs.namenode.http-address.nameservice1.namenode37
hadoop106:9870
dfs.client.failover.proxy.provider.nameservice1
org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider
dfs.namenode.http-address.mycluster.nn1
hadoop101:9870
dfs.namenode.http-address.mycluster.nn2
hadoop102:9870
dfs.namenode.http-address.mycluster.nn3
hadoop103:9870
dfs.client.failover.proxy.provider.mycluster
org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider
修改CDH hosts
[root@hadoop101 ~]# vim /etc/hosts
进行分发,这里的hadoop104,hadoop105,hadoop106,分别对应apache的hadoop101,hadoop102,hadoop103
[root@hadoop101 ~]# scp /etc/hosts hadoop102:/etc/
[root@hadoop101 ~]# scp /etc/hosts hadoop103:/etc/
同样修改CDH集群配置,在所有hdfs-site.xml文件里修改配置
dfs.nameservices
mycluster,nameservice1
dfs.internal.nameservices
nameservice1
dfs.ha.namenodes.mycluster
nn1,nn2,nn3
dfs.namenode.rpc-address.mycluster.nn1
hadoop104:8020
dfs.namenode.rpc-address.mycluster.nn2
hadoop105:8020
dfs.namenode.rpc-address.mycluster.nn3
hadoop106:8020
dfs.namenode.http-address.mycluster.nn1
hadoop104:9870
dfs.namenode.http-address.mycluster.nn2
hadoop105:9870
dfs.namenode.http-address.mycluster.nn3
hadoop106:9870
dfs.client.failover.proxy.provider.mycluster
org.apache.hadoop.hdfs.server.namenode.ha.ConfiguredFailoverProxyProvider
最后注意:重点由于我的Apahce集群和CDH集群3台集群都是hadoop101,hadoop102,hadoop103,所以要关闭域名访问,使用ip访问
CDH把钩去了
apache设置为false
再使用hadoop distcp命令进行迁移,-Dmapred.job.queue.name指定队列,默认是default队列。上面配置集群都配了的话,那么在CDH和apache集群下都可以执行这个命令
[root@hadoop101 hadoop]# hadoop distcp -Dmapred.job.queue.name=hive webhdfs://mycluster:9070/user/hive/warehouse/dwd.db/ hdfs://nameservice1/user/hive/warehouse
会启动一个mr任务,正在迁移
查看cdh 9870 http地址
数据已经成功迁移。数据迁移成功之后,接下来迁移hive表结构,编写shell脚本
[root@hadoop101 module]# vim exportHive.sh
#!/bin/bash
hive -e “use dwd;show tables”>tables.txt
cat tables.txt |while read eachline
do
hive -e “use dwd;show create table $eachline”>>tablesDDL.txt
echo “;” >> tablesDDL.txt
done
执行脚本后将tablesDDL.txt文件分发到CDH集群下
[root@hadoop101 module]# scp tablesDDL.txt hadoop104:/opt/module/
然后CDH下导入此表结构,先进到CDH的hive里创建dwd库
[root@hadoop101 module]# hive
hive> create database dwd;
创建数据库后,边界tablesDDL.txt,在最上方加上use dwd;
并且将createtab_stmt都替换成空格
[root@hadoop101 module]# sed -i s"#createtab_stmt# #g" tablesDDL.txt
最后执行hive -f命令将表结构导入
[root@hadoop101 module]# hive -f tablesDDL.txt
最后将表的分区重新刷新下,只有刷新分区才能把数据读出来,编写脚本
[root@hadoop101 module]# vim msckPartition.sh
#!/bin/bash
hive -e “use dwd;show tables”>tables.txt
cat tables.txt |while read eachline
do
hive -e “use dwd;MSCK REPAIR TABLE $eachline”
done
[root@hadoop101 module]# chmod 777 msckPartition.sh
[root@hadoop101 module]# ./msckPartition.sh
刷完分区后,查询表数据