hdp2.4搭建

1、集群安装

2、配置Standby NameNode 、Standby HBase Master

3、集群扩容节点、添加zk、jn

一、集群安装

虚拟机目录 /data/www/html/hbase

httpd配置文件修改下面三行路径 vi /etc/httpd/conf/httpd.conf (默认是/var/www/html/)

DocumentRoot "/data/www/html"
<Directory "/data/www">
<Directory "/data/www/html">
重启 /bin/systemctl restart httpd.service (启动httpd /bin/systemctl start httpd.service)

浏览器访问 http://192.168.159.11/hbase/

关闭透明大页

[root@hdp11 ~]# cat /sys/kernel/mm/transparent_hugepage/enabled
[always] madvise never
[root@hdp11 ~]# echo never > /sys/kernel/mm/transparent_hugepage/enabled
[root@hdp11 ~]# cat /sys/kernel/mm/transparent_hugepage/enabled
always madvise [never]

yum文件

[root@hdp1 hbase]# cat /etc/yum.repos.d/hdp.repo

[AMBARI-2.2.1.0]
name=AMBARI-2.2.1.0
baseurl=http://hdp1/hbase/AMBARI-2.2.1.0/centos7/2.2.1.0-161/
enabled=1
gpgcheck=0

[HDP-2.2.1.0]
name=HDP-2.2.1.0
baseurl=http://hdp1/hbase/HDP/centos7/2.x/updates/2.4.0.0/
enabled=1
gpgcheck=0

[HDP-uti]
name=HDP-uti
baseurl=http://hdp1/hbase/HDP-UTILS-1.1.0.20/repos/centos7/
enabled=1
gpgcheck=0

所有节点

yum -y install ambari-agent
ambari-agent start
chkconfig --add ambari-agent

修改配置文件hostname /etc/ambari-agent/conf/ambari-agent.ini service ambari-agent restart

master节点 yum install ambari-server -y

初始化ambari-server ambari-server setup

一路默认（jdk选[3] Custom JDK 填JAVA_HOME地址数据库选[1] - PostgreSQL (Embedded)）

ambari账号密码都是admin（或者ambari）

Database name (ambari):
Postgres schema (ambari):
Username (ambari):
Enter Database Password (bigdata):

[root@hdp9 ~]# ambari-server setup
Using python  /usr/bin/python2
Setup ambari-server
Checking SELinux...
SELinux status is 'disabled'
Customize user account for ambari-server daemon [y/n] (n)? 
Adjusting ambari-server permissions and ownership...
Checking firewall status...
Redirecting to /bin/systemctl status iptables.service

Checking JDK...
Do you want to change Oracle JDK [y/n] (n)? y
[1] Oracle JDK 1.8 + Java Cryptography Extension (JCE) Policy Files 8
[2] Oracle JDK 1.7 + Java Cryptography Extension (JCE) Policy Files 7
[3] Custom JDK
==============================================================================
Enter choice (1): 3
WARNING: JDK must be installed on all hosts and JAVA_HOME must be valid on all hosts.
WARNING: JCE Policy files are required for configuring Kerberos security. If you plan to use Kerberos,please make sure JCE Unlimited Strength Jurisdiction Policy Files are valid on all hosts.
Path to JAVA_HOME: /var/java/jdk1.8.0_151/
Validating JDK on Ambari Server...done.
Completing setup...
Configuring database...
Enter advanced database configuration [y/n] (n)? y
Configuring database...
==============================================================================
Choose one of the following options:
[1] - PostgreSQL (Embedded)
[2] - Oracle
[3] - MySQL
[4] - PostgreSQL
[5] - Microsoft SQL Server (Tech Preview)
[6] - SQL Anywhere
==============================================================================
Enter choice (1): 
Database name (ambari): 
Postgres schema (ambari): 
Username (ambari): admin
Enter Database Password (ambari): 
Re-enter password: 
Default properties detected. Using built-in database.
Configuring ambari database...
Checking PostgreSQL...
Configuring local database...
Connecting to local database...done.
Configuring PostgreSQL...
Restarting PostgreSQL
Extracting system views...
.......
Adjusting ambari-server permissions and ownership...
Ambari Server 'setup' completed successfully.

启动service ambari-server start 页面访问http://192.168.159.19:8080/ admin admin

ambari-server start   启动  
ambari-server stop   停止
ambari-server restart   重启
	
ambari-agent start  启动
ambari-agent stop  停止
ambari-agent status 查看状态

Name your cluster kzx_pay_hf

HDP 2.4

Advanced Repository Options (参考 /etc/yum.repos.d/hdp.repo)

Choose Services

HDFS HBase ZooKeeper Ambari Metrics

这里注意 failed disk tolerance设置，测试环境如果只有一个磁盘又设置1会导致datanode启动失败

报org.apache.hadoop.util.DiskChecker$DiskErrorException: Invalid volume failure config value: 1

【HDFS问题】datanode启动失败的问题

集群安装完毕！！！

二、配置Standby NameNode 、Standby HBase Master

为集群配置 Standby NameNode 、Standby HBase Master

HDFS --> Service Actions --> Add HBase Master 为hbase增加 Standby HBase Maste（添加完需要start，先显示hbase master，启动完会变成Standby HBase Maste）
Hbase --> Service Actions --> Enable NameNode HA（需停掉hbase），填Nameservice ID: ，然后
集群配置 Standby NameNode 、Standby HBase Master 完毕！！！

三、集群扩容节点、添加zk、jn

扩容集群：添加2个节点，添加zk、JournalNode

现有集群hdp1、hdp2、hdp3准备2个新节点bjc-hdp4 bjc-hdp5 并初始化 (修改好/etc/ambari-agent/conf/ambari-agent.ini)

192.168.159.11 hdp1.kzx.com hdp1 
192.168.159.12 hdp2.kzx.com hdp2
192.168.159.13 hdp3.kzx.com hdp3

192.168.159.14 bjc-hdp4.kzx.com bjc-hdp4
192.168.159.15 bjc-hdp5.kzx.com bjc-hdp5

192.168.159.16 hdp6.kzx.com hdp6
192.168.159.17 hdp7.kzx.com hdp7
192.168.159.18 hdp8.kzx.com hdp8

192.168.159.19 hdp9.kzx.com hdp9
192.168.159.20 hdp10.kzx.com hdp10
192.168.159.21 hdp11.kzx.com hdp11

Ambari界面Hosts-->Actions-->Add New Hosts

如果Status显示失败，检查ambari-agent配置并重启再recheck试试 systemctl restart ambari-agent.service（service ambari-agent restart）

等待红色警告消失

安装zk

Add ZooKeeper Server

添加好2个zk

这时候先别启动新增的两个zk

这里没有要重启zk

不要用这个restart All，应该一个个节点进行重启affected的应用。

不用用restart ALL

逐个重启rs和hmaster后就要解决zk重启的问题

（停1个非leader的zk，保留2个旧zk）重启全部zk ，10s内重启好，不会影响集群

while :;do echo -n " ";date; for i in hdp1 hdp2 hdp3 bjc-hdp4 bjc-hdp5 ;do echo -n "$i:"; echo mntr|nc $i 2181|grep zk_server_state;done;sleep 2;done

安装jn

以下两条命令操作加jn（红色是需要修改的）：
curl -u admin:admin -H 'X-Requested-By: ambari' -X POST http://hdp1:8080/api/v1/clusters/kzx_pay/hosts/bjc-hdp4.kzx.com/host_components/JOURNALNODE

curl -u admin:admin -H 'X-Requested-By: ambari' -X PUT -d '{"RequestInfo":{"context":"Install JournalNode"},"Body":{"HostRoles":{"state":"INSTALLED"}}}' http://hdp1:8080/api/v1/clusters/kzx_pay/hosts/bjc-hdp4.kzx.com/host_components/JOURNALNODE

HDFS
Configs
Advanced
Custom hdfs-site
dfs.namenode.shared.edits.dir
qjournal://hdp1.kzx.com:8485;hdp2.kzx.com:8485;hdp3.kzx.com:8485;bjc-hdp4:8485;bjc-hdp5:8485/kzxpay
qjournal://bjc-hdp4.kzx.com:8485;bjc-hdp5.kzx.com:8485;hdp1.kzx.com:8485;hdp2.kzx.com:8485;hdp3.kzx.com:8485/kzxpay

save

Restart Required: 17 Components on 5 Hosts

逐个重启 jn先不重启（3个jn running 2个stop 重启一个就只有2个running了）

再先启动两个新jn 最后逐个restart旧的3个jn

hdp1 ll /hadoop/hdfs/journal
新jn节点 mkdir -pv /hadoop/hdfs/journal/kzxpay
hdp1 scp -r /hadoop/hdfs/journal/kzxpay/* hdp5:/hadoop/hdfs/journal/kzxpay/
chown hdfs:hadoop -R /hadoop/hdfs/journal/
界面启动然后删除lock文件检查是否有最新lock文件生成 rm -f /hadoop/hdfs/journal/kzxpay/in_use.lock
cat /hadoop/hdfs/journal/kzxpay/in_use.lock
ls -lrt /hadoop/hdfs/journal/kzxpay/current/*inprogress*
重启 nn1 nn2

posted on 2024-04-29 13:14 寒星12345678999 阅读(14) 评论(2) 编辑收藏举报