【ambari-部署】Ambari2.6.2 HDP2.6.5 大数据集群搭建

转自： https://blog.csdn.net/weixin_33923762/article/details/90932169

hadoop 集群 master无法免密登录slave2： https://blog.csdn.net/weixin_37600848/article/details/83418515

Ambari,hadoop的配置,管理和监控项目入门： https://blog.csdn.net/wuliusir/article/details/50754859

Ambari2.6.2集成Kerberos： https://blog.51cto.com/784687488/2332072?source=dra

Flink流处理之迭代API分析： https://blog.csdn.net/yanghua_kobe/article/details/53456674

Ambari架构源码解析： https://blog.csdn.net/chengyuqiang/article/details/60963480

ambari rest api (三)： https://www.cnblogs.com/itboys/p/7004148.html

Flink Rest API使用：  https://blog.csdn.net/qq_27710549/article/details/80500857

HUE配置文件hue.ini 的zookeeper模块详解（图文详解）（分HA集群）： https://www.cnblogs.com/zlslch/p/6817538.html

Flink 通过 REST API 获取实时 task metric 汇总统计信息： https://blog.csdn.net/yuchuanchen/article/details/88406438

CDH 和ambari之间的比较： https://blog.csdn.net/qq_32649581/article/details/90899363

docker安装centos7搭建集群: https://blog.csdn.net/kk25114/article/details/88639383

hadoop集群上部署flink环境: https://blog.csdn.net/weixin_40626699/article/details/89139591

大数据基础之Ambari（3）通过Ambari部署Airflow: https://www.cnblogs.com/barneywill/p/10284804.html

Hadoop建设工具Ambari的安装部署及完整使用（一）——介绍: https://blog.csdn.net/balabalayi/article/details/64904822

flink实战--REST API与metrics监控:  https://blog.csdn.net/aA518189/article/details/88952910

Flink架构、原理与部署测试及从kafka到flink到hbase:  https://blog.csdn.net/qq_38799155/article/details/80808818

分布式集群离线配置:  https://blog.csdn.net/m0_37614660/article/details/88987862

Ubuntu 16.04安装Hadoop及Spark:  https://blog.csdn.net/u011596455/article/details/53115410

Ambari部署Hadoop环境&安装ZooKeeper :  https://my.oschina.net/qiangzigege/blog/534176

随笔分类 - ELK(Elasticsearch/Logstash/Kibana)集群搭建:  https://www.cnblogs.com/zlslch/category/972313.html

比较完整的hadoop集群组件的安装教程; https://blog.csdn.net/u010199356/article/details/87538403

本地调试opentsdb的环境搭建： https://blog.csdn.net/weixin_34242819/article/details/88281740

 

 

 

 

原文链接：http://www.jianshu.com/p/abcb22a47652

Ambari 2.6.2 中
HDFS-2.7.3 YARN-2.7.3 HIVE-1.2.1 HBASE-1.1.2 ZOOKEEPER-3.4.6 SPARK-2.3.0

注：本文基于root用户操作

一、安装环境准备

操作系统 centos7.5
hdc-data1：192.168.163.51
hdc-data2：192.168.163.52
hdc-data3：192.168.163.53
【安装环境准备每台集群机器一样的操作，或者使用scp远程拷贝】

1、主机名IP映射配置

FQDN：(Fully Qualified Domain Name)全限定域名：同时带有主机名和域名的名称。（通过符号“.”）
例如：主机名是bigserver,域名是mycompany.com,那么FQDN就是bigserver.mycompany.com。

1. vi /etc/hosts

2. #添加如下地址映射及FQDN（ambari注册时需要）

3. 192.168.163.51 hdc-data1 hdc-data1.hadoop

4. 192.168.163.52 hdc-data2 hdc-data2.hadoop

5. 192.168.163.53 hdc-data3 hdc-data3.hadoop

2、SSH免密登录配置

1. ssh-keygen

2. ssh-copy-id -i ~/.ssh/id_rsa.pub root@hdc-data1

3. ssh-copy-id -i ~/.ssh/id_rsa.pub root@hdc-data2

4. ssh-copy-id -i ~/.ssh/id_rsa.pub root@hdc-data3

image.png

3、关闭及禁止防火墙

1. systemctl stop firewalld.service

2. systemctl disable firewalld.service

4、关闭SELinux

SELinux(Security-Enhanced Linux) 是美国国家安全局（NSA）对于强制访问控制的实现，是 Linux历史上最杰出的新安全子系统。NSA是在Linux社区的帮助下开发了一种访问控制体系，在这种访问控制体系的限制下，进程只能访问那些在他的任务中所需要文件。

1. vi /etc/sysconfig/selinux

2. #修改以下内容

3. SELINUX=disabled

image.png

5、开启NTP服务

1. yum install -y ntp

2. systemctl enable ntpd

3. systemctl start ntpd

6、安装JDK

下载地址：https://www.oracle.com/technetwork/cn/java/javase/downloads/jdk8-downloads-2133151-zhs.html

通过文件传输上传到服务器，这里不可用wget直接下载。

1. mkdir -p /opt/java

2. tar -zxvf jdk-8u181-linux-x64.tar.gz -C /opt/java/

3. vi /etc/profile

4.  

5. export JAVA_HOME=/opt/java/jdk1.8.0_181

6. export JRE_HOME=$JAVA_HOME/jre

7. export CLASSPATH=.:$JAVA_HOME/lib:$JAVA_HOME/jre/lib

8. export PATH=$PATH:$JAVA_HOME/bin

远程分发到其他服务器

1. scp -r /opt/java/jdk1.8.0_181/ root@hdc-data2:/opt/java/

2. scp -r /opt/java/jdk1.8.0_181/ root@hdc-data3:/opt/java/

3.  

4. scp /etc/profile root@hdc-data2:/etc/

5. scp /etc/profile root@hdc-data3:/etc/

6.  

7. source /etc/profile

二、Ambari安装

1、制作Ambari本地yum源

选取一台机器即可，在此选择hdc-data1。

https://docs.hortonworks.com/HDPDocuments/Ambari-2.6.2.0/bk_ambari-installation/content/ch_obtaining-public-repos.html

image.png

 

image.png

 

image.png

 

需要下载资源安装包如下：

1. wget http://public-repo-1.hortonworks.com/ambari/centos7/2.x/updates/2.6.2.0/ambari-2.6.2.0-centos7.tar.gz

2. wget http://public-repo-1.hortonworks.com/HDP/centos7/2.x/updates/2.6.5.0/HDP-2.6.5.0-centos7-rpm.tar.gz

3. wget http://public-repo-1.hortonworks.com/HDP-UTILS-1.1.0.22/repos/centos7/HDP-UTILS-1.1.0.22-centos7.tar.gz

4. wget http://public-repo-1.hortonworks.com/HDP-GPL/centos7/2.x/updates/2.6.5.0/HDP-GPL-2.6.5.0-centos7-gpl.tar.gz

1.1 安装 Apache HTTP 服务器

1. yum install httpd -y

2. #开启服务并设置开机自启动

3. systemctl start httpd.service

4. systemctl enable httpd.service

1.2 安装制作yum源工具

yum install yum-utils createrepo

1.3 创建http服务器

http根目录默认是即/var/www/html/

1. mkdir -p /var/www/html/ambari

2. #cd /var/www/html/ambari

3. #将下载好的安装包上传并解压

4. tar xvf HDP-2.6.5.0-centos7-rpm.tar.gz -C /var/www/html/ambari

5. tar xvf ambari-2.6.2.0-centos7.tar.gz -C /var/www/html/ambari

6. tar xvf HDP-UTILS-1.1.0.22-centos7.tar.gz -C /var/www/html/ambari

7. tar xvf HDP-GPL-2.6.5.0-centos7-gpl.tar.gz -C /var/www/html/ambari

8.  

9. # 删除压缩包

10. rm -rf ambari-2.6.2.0-centos7.tar.gz

11. rm -rf HDP-2.6.5.0-centos7-rpm.tar.gz

12. rm -rf HDP-UTILS-1.1.0.22-centos7.tar.gz

13. rm -rf HDP-GPL-2.6.5.0-centos7-gpl.tar.gz

验证

image.png

1.4 配置ambari、HDP、HDP-UTILS的本地源

1. #yum install wget -y

2. cd /etc/yum.repos.d/

3. wget http://public-repo-1.hortonworks.com/ambari/centos7/2.x/updates/2.6.2.0/ambari.repo

4. wget http://public-repo-1.hortonworks.com/HDP/centos7/2.x/updates/2.6.5.0/hdp.repo

5. wget http://public-repo-1.hortonworks.com/HDP-GPL/centos7/2.x/updates/2.6.5.0/hdp.gpl.repo

编辑ambari.repo，修改baseurl和gpgkey

1. [root@hdc-data1 yum.repos.d]# vi ambari.repo

2.  

3. #VERSION_NUMBER=2.6.2.0-155

4. [ambari-2.6.2.0]

5. name=ambari Version - ambari-2.6.2.0

6. #baseurl=http://public-repo-1.hortonworks.com/ambari/centos7/2.x/updates/2.6.2.0

7. baseurl=http://192.168.163.51/ambari/ambari/centos7/2.6.2.0-155

8. gpgcheck=1

9. #gpgkey=http://public-repo-1.hortonworks.com/ambari/centos7/2.x/updates/2.6.2.0/RPM-GPG-KEY/RPM-GPG-KEY-Jenkins

10. gpgkey=http://192.168.163.51/ambari/ambari/centos7/2.6.2.0-155/RPM-GPG-KEY/RPM-GPG-KEY-Jenkins

11. enabled=1

12. priority=1

编辑hdp.repo，修改baseurl和gpgkey

1. [root@hdc-data1 yum.repos.d]# vi hdp.repo

2.  

3. #VERSION_NUMBER=2.6.5.0-292

4. [HDP-2.6.5.0]

5. name=HDP Version - HDP-2.6.5.0

6. #baseurl=http://public-repo-1.hortonworks.com/HDP/centos7/2.x/updates/2.6.5.0

7. baseurl=http://192.168.163.51/ambari/HDP/centos7/2.6.5.0-292

8. gpgcheck=1

9. #gpgkey=http://public-repo-1.hortonworks.com/HDP/centos7/2.x/updates/2.6.5.0/RPM-GPG-KEY/RPM-GPG-KEY-Jenkins

10. gpgkey=http://192.168.163.51/ambari/HDP/centos7/2.6.5.0-292/RPM-GPG-KEY/RPM-GPG-KEY-Jenkins

11. enabled=1

12. priority=1

13.  
14.  

15. [HDP-UTILS-1.1.0.22]

16. name=HDP-UTILS Version - HDP-UTILS-1.1.0.22

17. #baseurl=http://public-repo-1.hortonworks.com/HDP-UTILS-1.1.0.22/repos/centos7

18. baseurl=http://192.168.163.51/ambari/HDP-UTILS/centos7/1.1.0.22

19. gpgcheck=1

20. #gpgkey=http://public-repo-1.hortonworks.com/HDP/centos7/2.x/updates/2.6.5.0/RPM-GPG-KEY/RPM-GPG-KEY-Jenkins

21. gpgkey=http://192.168.163.51/ambari/HDP-UTILS/centos7/1.1.0.22/RPM-GPG-KEY/RPM-GPG-KEY-Jenkins

22. enabled=1

23. priority=1

编辑hdp.gpl.repo，修改baseurl和gpgkey

1. [root@hdc-data1 yum.repos.d]# vi hdp.gpl.repo

2.  

3. #VERSION_NUMBER=2.6.5.0-292

4. [HDP-GPL-2.6.5.0]

5. name=HDP-GPL Version - HDP-GPL-2.6.5.0

6. #baseurl=http://public-repo-1.hortonworks.com/HDP-GPL/centos7/2.x/updates/2.6.5.0

7. baseurl=http://192.168.163.51/ambari/HDP-GPL/centos7/2.6.5.0-292

8. gpgcheck=1

9. #gpgkey=http://public-repo-1.hortonworks.com/HDP-GPL/centos7/2.x/updates/2.6.5.0/RPM-GPG-KEY/RPM-GPG-KEY-Jenkins

10. gpgkey=http://192.168.163.51/ambari/HDP-GPL/centos7/2.6.5.0-292/RPM-GPG-KEY/RPM-GPG-KEY-Jenkins

11. enabled=1

12. priority=1

13.  

分发到其他机器

1. scp /etc/yum.repos.d/ambari.repo root@hdc-data2:/etc/yum.repos.d/

2. scp /etc/yum.repos.d/ambari.repo root@hdc-data3:/etc/yum.repos.d/

3. scp /etc/yum.repos.d/hdp.repo root@hdc-data2:/etc/yum.repos.d/

4. scp /etc/yum.repos.d/hdp.repo root@hdc-data3:/etc/yum.repos.d/

5. scp /etc/yum.repos.d/hdp.gpl.repo root@hdc-data2:/etc/yum.repos.d/

6. scp /etc/yum.repos.d/hdp.gpl.repo root@hdc-data3:/etc/yum.repos.d/

每台机器yum配置

1. yum clean all

2. yum makecache

3. yum list

2、安装Mysql数据库

Ambari安装会将安装等信息写入数据库，建议使用MariaDB数据库，也可以不安装而使用默认数据库PostgreSQL

2.1 安装及初始化设置

1. [root@hdc-data1 ~]# yum install mariadb-server

2. [root@hdc-data1 ~]# systemctl start mariadb

3. [root@hdc-data1 ~]# systemctl enable mariadb

4. [root@hdc-data1 ~]# mysql_secure_installation

5.  

6. #首先是设置密码，会提示先输入密码

7. Enter current password for root (enter for none):<–初次运行直接回车

8. #设置密码

9. Set root password? [Y/n] <– 是否设置root用户密码，输入y并回车或直接回车

10. New password: <– 设置root用户的密码

11. Re-enter new password: <– 再输入一次你设置的密码

12. #其他配置

13. Remove anonymous users? [Y/n] <– 是否删除匿名用户，回车

14. Disallow root login remotely? [Y/n] <–是否禁止root远程登录,回车,

15. Remove test database and access to it? [Y/n] <– 是否删除test数据库，回车

16. Reload privilege tables now? [Y/n] <– 是否重新加载权限表，回车

【可选】修改mysql端口（生产环境安全考虑）
查看端口

image.png

 

image.png

 

2.2 安装完成后创建ambari数据库及用户

1. #进入mysql shell

2. mysql -uroot -p

1. create database ambari character set utf8 ;

2. CREATE USER 'ambari'@'%'IDENTIFIED BY 'ambari123';

3. GRANT ALL PRIVILEGES ON *.* TO 'ambari'@'%';

4. FLUSH PRIVILEGES;

如果要安装Hive，创建Hive数据库和用户

1. create database hive character set utf8 ;

2. CREATE USER 'hive'@'%'IDENTIFIED BY 'hive123';

3. GRANT ALL PRIVILEGES ON *.* TO 'hive'@'%';

4. FLUSH PRIVILEGES;

如果要安装Oozie，创建Oozie数据库和用户

1. create database oozie character set utf8 ;

2. CREATE USER 'oozie'@'%'IDENTIFIED BY 'oozie123';

3. GRANT ALL PRIVILEGES ON *.* TO 'oozie'@'%';

4. FLUSH PRIVILEGES;

3、安装 ambari

yum install ambari-server

4、下载mysql驱动

http://central.maven.org/maven2/mysql/mysql-connector-java/5.1.40/
将mysql-connector-Java.jar复制到/usr/share/java目录下

1. mkdir /usr/share/java

2. cp mysql-connector-java-5.1.40.jar /usr/share/java/mysql-connector-java.jar

将mysql-connector-java.jar复制到/var/lib/ambari-server/resources目录下

cp mysql-connector-java-5.1.40.jar /var/lib/ambari-server/resources/mysql-jdbc-driver.jar

5、编辑/etc/ambari-server/conf/ambari.properties，添加如下内容

1. server.jdbc.driver.path=/usr/share/java/mysql-connector-java.jar

2. #【可选】修改默认8080端口

3. #client.api.port=18080

6、ambaria初始化

1. #设置mysql驱动

2. ambari-server setup --jdbc-db=mysql --jdbc-driver=/usr/share/java/mysql-connector-java.jar

3. ambari-server setup

完整初始化如下：

1. [root@hdc-data1 ~]# ambari-server setup

2. Using python /usr/bin/python

3. Setup ambari-server

4. Checking SELinux...

5. SELinux status is 'disabled'

6. Customize user account for ambari-server daemon [y/n] (n)? y

7. Enter user account for ambari-server daemon (root):

8. Adjusting ambari-server permissions and ownership...

9. Checking firewall status...

10. Checking JDK...

11. [1] Oracle JDK 1.8 + Java Cryptography Extension (JCE) Policy Files 8

12. [2] Oracle JDK 1.7 + Java Cryptography Extension (JCE) Policy Files 7

13. [3] Custom JDK

14. ==============================================================================

15. Enter choice (1): 3

16. WARNING: JDK must be installed on all hosts and JAVA_HOME must be valid on all hosts.

17. 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.

18. Path to JAVA_HOME: /opt/java/jdk1.8.0_181

19. Validating JDK on Ambari Server...done.

20. Checking GPL software agreement...

21. GPL License for LZO: https://www.gnu.org/licenses/old-licenses/gpl-2.0.en.html

22. Enable Ambari Server to download and install GPL Licensed LZO packages [y/n] (n)? y

23. Completing setup...

24. Configuring database...

25. Enter advanced database configuration [y/n] (n)? y

26. Configuring database...

27. ==============================================================================

28. Choose one of the following options:

29. [1] - PostgreSQL (Embedded)

30. [2] - Oracle

31. [3] - MySQL / MariaDB

32. [4] - PostgreSQL

33. [5] - Microsoft SQL Server (Tech Preview)

34. [6] - SQL Anywhere

35. [7] - BDB

36. ==============================================================================

37. Enter choice (1): 3

38. Hostname (localhost):

39. Port (3306):

40. Database name (ambari):

41. Username (ambari):

42. Enter Database Password (bigdata):

43. Re-enter password:

44. Configuring ambari database...

45. Configuring remote database connection properties...

46. WARNING: Before starting Ambari Server, you must run the following DDL against the database to create the schema: /var/lib/ambari-server/resources/Ambari-DDL-MySQL-CREATE.sql

47. Proceed with configuring remote database connection properties [y/n] (y)? y

48. Extracting system views...

49. .....ambari-admin-2.6.2.0.155.jar

50. ......

51. Adjusting ambari-server permissions and ownership...

52. Ambari Server 'setup' completed successfully.

说明：
1：提示是否自定义设置。输入：y 或 按回车继续
2：ambari-server 账号。使用root（推荐）用户直接回车 若使用ambari用户输入：ambari
3：设置JDK,输入数字3，选客户端安装jdk，如果服务器可以访问外网，则可以选1，自动下载jdk1.8，默认下载的安装目录是/usr/java/default
4：如果上面选择3自定义JDK,则需要设置JAVA_HOME
5：输入：y 或 按回车继续
6：输入“y“，进行数据库的高级配置
7：输入“3“，选择mysql/mariadb数据库
8：设置数据库的配置参数：host,port,database,username,password。根据实际情况输入，如果和括号内相同，则可以直接回车。若端口改6033则输入port:6033。
9：输入：y 或 按回车继续
（若设置更改，可以再次执行ambari-server setup进行设置）

7、将Ambari数据库脚本导入到数据库

1. #用Ambari用户（上面设置的用户）登录mysql

2. mysql -u ambari -p

3. use ambari;

4. source /var/lib/ambari-server/resources/Ambari-DDL-MySQL-CREATE.sql

image.png

8、启动Ambari

ambari-server start

 

image.png

浏览器访问： http://hdc-data1:8080/ 默认登录用户：admin，密码：admin

 

9、WEBUI配置

image.png

 

 

创建集群名称

image.png

 

 

选择HDP版本和选择本地仓库

image.png

配置HDP的Repository

image.png

输入集群节点host(FQDN)和Ambari节点SSH的私钥

image.png

 

 

等待ambari-agents注册

image.png

若报错

image.png

 

image.png

【遇到问题 Confirm Hosts】
注册ambari-agents时failed
NetUtil.py:96 - EOF occurred in violation of protocol (_ssl.c:579)
SSLError: Failed to connect. Please check openssl library versions.
解决：如下第三个方案，/etc/ambari-agent/conf/ambari-agent.ini文件只有执行该步骤进行注册才生成。修改后点击“Retry Failed”按钮，等待注册成功进入下一步。

1. # 1、yum upgrade openssl 已是最新本版，未解决

2. # 2、vi /etc/python/cert-verification.cfg 修改 verify=disable，未解决

3. [https]

4. #verify=platform_default

5. verify=disable

6. # 3、最后解决方案 在ambari-agent的配置文件/etc/ambari-agent/conf/ambari-agent.ini

7. 在 ［security］标签下面增加一项

8. [security]

9. force_https_protocol=PROTOCOL_TLSv1_2

image.png

 

 

选择安装组件，（最小化安装HDFS+YARN+MAPREDUCE2+Ambari Metrics+SmartSense+ZooKeeper+Hbase）不必要的组件可以暂时不安装节省安装时间，后续可以再添加安装组件

image.png

 

image.png

 

 

实际部署时需按规划设计进行分配（注意：Hbase HA 的话在Hbase Master 点击后面的小绿色+号，添加standby master）：

image.png

工作节点角色分配(生产环境全勾了)

image.png

组件配置信息修改
通常数据目录修改到/data/目录下，有红色信息提示的组件表示要输入相应的账号密码。以下示例，可根据自己的实际情况配置。

 

HDFS

image.png

 

image.png

HIVE,若数据库端口修改按实际填写

image.png

Oozie

image.png

Zookeeper

image.png

 

image.png

 

image.png

 

image.png

暂时修改了以下属性，具体配置信息可根据实际情况后续相应修改

端口属性	修改值
dfs.namenode.http-address	server1.hadoop:5070 （def:50070）
yarn.resourcemanager.webapp.address	server2.hadoop:8888 （def:8088）
yarn.resourcemanager.webapp.https.address	server2.hadoop:8890(def:8090)
mapreduce.jobhistory.webapp.address	server2.hadoop:18888 (def:19888)

属性	修改值
HDFS
NameNode	/data/hadoop/hdfs/namenode
DataNode	/data/hadoop/hdfs/data
SecondaryNameNode Checkpoint directories	/data/hadoop/hdfs/namesecondary
Hadoop PID Dir Prefix	/data/var/run/hadoop
Hadoop Log Dir Prefix	/data/var/log/hadoop
dfs.journalnode.edits.dir	/data/hadoop/hdfs/journalnode
Yarn
yarn.nodemanager.local-dirs	/data/hadoop/yarn/local
yarn.nodemanager.log-dirs	/data/hadoop/yarn/log
yarn.timeline-service.leveldb-state-store.path	/data/hadoop/yarn/timeline
yarn.timeline-service.leveldb-timeline-store.path	/data/hadoop/yarn/timeline
YARN Log Dir Prefix	/data/var/log/hadoop-yarn
YARN PID Dir Prefix	/data/var/run/hadoop-yarn
Mapreduce
Mapreduce Log Dir Prefix	/data/var/log/hadoop-mapreduce
Mapreduce PID Dir Prefix	/data/var/run/hadoop-mapreduce
mapreduce.jobhistory.recovery.store.leveldb.path	/data/hadoop/mapreduce/jhs
Hive
Hive Log Dir	/data/var/log/hive
Hive PID Dir	/data/var/run/hive
HBase
HBase Log Dir Prefix	/data/var/log/hbase
HBase PID Dir	/data/var/run/hbase
Oozie
Oozie Data Dir	/data/hadoop/oozie/data
Oozie Log Dir	/data/var/log/oozie
Oozie PID Dir	/data/var/run/oozie
zookeeper
ZooKeeper directory	/data/hadoop/zookeeper
ZooKeeper Log Dir	/data/var/log/zookeeper
ZooKeeper PID Dir	/data/var/run/zookeeper
ambari-infra
Metrics Collector log dir	/data/var/log/ambari-metrics-collector
Metrics Collector pid dir	/data/var/run/ambari-metrics-collector
Metrics Monitor log dir	/data/var/log/ambari-metrics-monitor
Metrics Monitor pid dir	/data/var/run/ambari-metrics-monitor
Aggregator checkpoint directory	/data/var/lib/ambari-metrics-collector/checkpoint
Metrics Grafana data dir	/data/var/lib/ambari-metrics-grafana
Metrics Grafana log dir	/data/var/log/ambari-metrics-grafana
Metrics Grafana pid dir	/data/var/run/ambari-metrics-grafana
hbase_log_dir	/data/var/log/ambari-metrics-collector
hbase_pid_dir	/data/var/run/ambari-metrics-collector/
hbase.tmp.dir	/data/var/lib/ambari-metrics-collector/hbase-tmp
ambari-infra
Infra Solr Client log dir	/data/var/log/ambari-infra-solr-client
Infra Solr log dir	/data/var/log/ambari-infra-solr
Infra Solr pid dir	/data/var/run/ambari-infra-solr
spark
livy2_log_dir	/data/var/log/livy2
livy2_pid_dir	/data/var/run/livy2
spark_log_dir	/data/var/log/spark2
spark_pid_dir	/data/var/run/spark2

 

 

等待安装完成

image.png

 

 

直到所有节点成功安装完成才能进行下一步操作

image.png

最终结果界面类似如下：

image.png

安装完成！

 

三、后续操作

操作前最好先将集群打个快照备份

1、HDFS启用HA(高可用)

参考http://www.louisvv.com/archives/1490.html

step1：关闭hbase、hive相关服务

image.png

step2:在HDFS选择Enable NameNode HA

image.png

step3:输入namenode ha serverid

image.png

step4:默认，进入下一步

image.png

step5:预览，直接进入下一步

image.png

step6:创建checkpoint，根据提示执行下列命令

image.png

根据提示在server1执行下列两条命令

image.png

命令执行完成方可进入下一步

image.png

step7:等待配置进程完成

image.png

step8:Journalnode初始化

image.png

需要在原NameNode节点执行下列命令

 

sudo su hdfs -l -c 'hdfs namenode -initializeSharedEdits'

 

image.png

step9: 启动组件，进入下一步

image.png

step10:初始化元数据，

 

sudo su hdfs -l -c 'hdfs zkfc -formatZK'

 

image.png

注意在 新添加的NameNode节点上操作

 

sudo su hdfs -l -c 'hdfs namenode -bootstrapStandby'

 

image.png

 

image.png

step11:等待所有服务重启，完成

image.png

 

image.png