【Ambari】Docker 安装Ambari 大数据单机版本

目录

一、前期准备

1.1 部署 docker

1.2 部署 docker-compose

1.3 版本说明

二 、镜像构建启动

2.1 系统镜像构建

2.2 安装包源镜像构建

2.3 kdc镜像构建

2.4 集群安装

2.5 容器导出为镜像

三、ubuntu环境安装测试

3.1 环境准备

3.2 集群容器启动

一、前期准备

1.1 部署 docker

# 安装yum-config-manager配置工具
yum -y install yum-utils

# 建议使用阿里云yum源：（推荐）
#yum-config-manager --add-repo https://download.docker.com/linux/centos/docker-ce.repo
yum-config-manager --add-repo http://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo

# 安装docker-ce版本
yum install -y docker-ce

# 启动并开机启动
systemctl enable --now docker
docker --version
docker info

1.2 部署 docker-compose

curl -sl https://github.com/docker/compose/releases/download/v2.16.0/docker-compose-linux-x86_64 -o /usr/local/bin/docker-compose

chmod +x /usr/local/bin/docker-compose
docker-compose --version

创建网络

docker network create hadoop-network

1.3 版本说明

如下安装只做测试。

二 、镜像构建启动

2.1 系统镜像构建

# 拉取 centos 镜像
from centos:7.6.1810

# 作者信息
label  kangll 

# 安装基本工具
run yum update -y && yum install -y \
    net-tools \
    vim \
    wget \
    curl \
    lrzsz \ 
    nc \
    openssl \
    sed \
    telnet \
    openssh-server \
    httpd \
    chronyd \
    zip unzip \
    openssh-clients \
    expect \
    krb5-server \
    krb5-libs \ 
    krb5-workstation \
    && yum clean all

# jdk 
run mkdir -p /usr/java 
add ./config/jdk-8u162-linux-x64.tar.gz /usr/java/
env java_home=/usr/java/jdk1.8.0_162
env classpath=.:$java_home/lib/dt.jar:$java_home/lib/tools.jar
env path=$java_home/bin:$path


# 添加一个名为myuser的用户
run useradd -m winner_spark
# 设置myuser用户的密码，这里设置为mypassword
run echo 'winner_spark:123456' | chpasswd
# 设置myuser用户为sudoer
run echo 'winner_spark all=(all) nopasswd:all' >> /etc/sudoers


# ssh
run sed -i '/^hostkey/'d /etc/ssh/sshd_config
run echo 'hostkey /etc/ssh/ssh_host_rsa_key' >> /etc/ssh/sshd_config
run mkdir /var/run/sshd/
run echo 'usedns no' >> /etc/ssh/sshd_config
run sed -i -e '/pam_loginuid.so/d' /etc/pam.d/sshd

copy ./config/hosts    /etc/

#set default password
run echo 'root:123456' | chpasswd
run /usr/bin/ssh-keygen -a

copy ./config/krb5.conf   /etc

#声明22端口 
expose 22

run mkdir -p /opt && mkdir -p /var/log/supervisord
run echo '/usr/sbin/sshd -d' >> /opt/run.sh
run chmod +x /opt/run.sh
cmd ["/opt/run.sh"]
 
#
workdir /opt/

系统镜像构建

docker build -f  dockerfile-centos -t centos-jdk:1.0  .

2.2 安装包源镜像构建

dockerfile-httpd

from centos-jdk:1.0

# centos7.6 jdk1.8
label kangll

run yum -y install httpd
run echo "crushlinux" >/var/www/html/index.html
run ln -sf /usr/share/zoneinfo/asia/shanghai /etc/localtime


# 添加yum 源
copy ./repo/ambari.repo   /etc/yum.repos.d/
copy ./repo/hdp.repo      /etc/yum.repos.d/
copy ./repo/hdp.gpl.repo  /etc/yum.repos.d/


run mkdir -p /hadoop/hdp
add ./hdp/ambari.tar.gz    /var/www/html
add ./hdp/hdp-gpl.tar.gz   /var/www/html
add ./hdp/hdp.tar.gz       /var/www/html
add ./hdp/hdp-utils.tar.gz /var/www/html

 
expose 80
 
cmd ["httpd","-dforeground"]

源镜像构建

docker build -t ambari:2.7.4  -f dockerfile-ambari  . 

2.3 kdc镜像构建

dockerfile-kerberos

from centos:7.6.1810

run yum install -y krb5-server krb5-libs krb5-auth-dialog krb5-workstation

run yum install -y net-tools telnet

copy ./config/krb5.conf   /etc
copy ./config/kdc.conf    /var/kerberos/krb5kdc/
copy ./config/kadm5.acl   /var/kerberos/krb5kdc/


cmd ["/usr/sbin/init"]

服务启动

# 进入容器 启动服务
systemctl restart krb5kdc
systemctl restart kadmin

docker-compose-base.yml 启动基础环境

version: '3'
services:
  mysql:
    image: registry.cn-hangzhou.aliyuncs.com/bigdata_cloudnative/mysql-test:5.7
    restart: always
    networks:
      - hadoop-network
    container_name: mysql
    hostname: mysql-test
    environment:
      mysql_root_password: winner001
      tz: asia/shanghai
    ports:
      - 3306:3306
    volumes:
      -  /data:/var/lib/mysql
      - ./conf/my.cnf:/etc/mysql/my.cnf
    command:
      --max_connections=1000
      --character-set-server=utf8mb4
      --collation-server=utf8mb4_general_ci
      --default-authentication-plugin=mysql_native_password
    healthcheck:
      test: ["cmd-shell", "curl --silent localhost:3306 >/dev/null || exit 1"]
      interval: 10s
      timeout: 10s
  httpd:
    container_name: httpd
    hostname: httpd
    image: httpd:1.0
    networks:
      - hadoop-network
    ports:
      - "80:80"
    privileged: true
    restart: always
    volumes:
      - "/etc/hosts:/etc/hosts"
  kdc:
    image: kerberos:1.0
    restart: always
    networks:
      - hadoop-network
    container_name: kerberos
    hostname: kerberos.winner
    ports:
      - 88:88
      - 749:749
      - 750:750
    volumes:
      - /etc/hosts:/etc/hosts
      - /hadoop/kerberos/krb5kdc:/var/kerberos/krb5kdc
      - /hadoop/kerberos/keytabs:/etc/security/keytabs
    command:
      - /bin/sh
      - -c
      - |
        /opt/apache/kerberosstart.sh

networks:
  hadoop-network:
    external:
      name: hadoop-network

ambari启动

# 启动基础环境
docker-compose -f docker-compose-base.yaml up -d

# 启动ambari , 在此容器中安装集群
docker-compose -f docker-compose-ambari.yaml up -d

2.4 集群安装

repo地址

节点和秘钥

agent安装

服务安装

由于是单机直接下一步

安装组件

输入密码

数据目录确认

用户确认下一步

服务参数确定

发布

安装完成

2.5 容器导出为像

docker ps -a
docker commit  容器id ambari-1.0.tar.gz

三、ubuntu环境安装测试

3.1 环境准备

我们准备ubuntu 环境，在ubuntu环境安装ambari单机，相当于在centos服务器安装的docker化的ambari 导出为镜像 然后在ubutntu 环境尝试启动。 

准备ubuntu 环境，版本：ubuntu 20.04.5 lts

将镜像 ambari-1.0.tar.gz 加载到ubuntu 环境

docker load -i ambari-1.0.tar.gz

mysql元数据库备份还原（ambari、hive、ranger）

hdfs、kerberos 备份的数据拷贝到ubuntu服务器并保持数据文件路径不变。

将在centos 安装完成的集群备份元数据库写入到新的库中 , 也就是ubuntu环境要使用的元数据库。

# 备份的元数据写入新的数据库中
mysql -hmysql-test -uroot -pwinner001 ambari < ambari.sql
mysql -hmysql-test -uroot -pwinner001 ranger < ranger.sql
mysql -hmysql-test -uroot -pwinner001 hive   < hive.sql

备份的kerberos和安装包解压

# 解压到 /usr 路径下，保存的是集群安装的数据
hdp.tar.gz

# 在根路径解压保留的kerberos 和hdfs 初始化的元数据
hadoop.tar.gz

基础环境启动docker-compose-base.yaml

version: '3'
services:
  mysql:
    image: registry.cn-hangzhou.aliyuncs.com/bigdata_cloudnative/mysql-test:5.7
    restart: always
    networks:
      - hadoop-network
    container_name: mysql
    hostname: mysql-test
    environment:
      mysql_root_password: winner001
      tz: asia/shanghai
    ports:
      - 3306:3306
    volumes:
      -  /data:/var/lib/mysql
      - ./conf/my.cnf:/etc/mysql/my.cnf
    command:
      --max_connections=1000
      --character-set-server=utf8mb4
      --collation-server=utf8mb4_general_ci
      --default-authentication-plugin=mysql_native_password
    healthcheck:
      test: ["cmd-shell", "curl --silent localhost:3306 >/dev/null || exit 1"]
      interval: 10s
      timeout: 10s
  httpd:
    container_name: httpd
    hostname: httpd
    privileged: true
    image: httpd:1.0
    networks:
      - hadoop-network
    ports:
      - "80:80"
    privileged: true
    restart: always
    volumes:
      - "/etc/hosts:/etc/hosts"
  kdc:
    image: kerberos:1.0
    restart: always
    privileged: true
    networks:
      - hadoop-network
    container_name: kerberos
    hostname: kerberos.winner
    ports:
      - 888:88
      - 8749:749
      - 8750:750
    volumes:
      - /etc/hosts:/etc/hosts
      - /hadoop/kerberos/krb5kdc:/var/kerberos/krb5kdc
      - /hadoop/kerberos/keytabs:/etc/security/keytabs

networks:
  hadoop-network:
    external:
      name: hadoop-network

启动

# 启动
docker-compose -f docker-compose-base.yaml up -d

# 停止
docker-compose -f docker-compose-base.yaml down 

3.2 集群容器启动

镜像

启动容器

version: '3'
services:
  ambari:
    image: ambari:2.7.4  # 修改为新的容器
    container_name: ambari
    hostname: ambari-server
    restart: always
    privileged: true
    networks:
      - hadoop-network
    ports:
      - 8080:8080
      - 8088:8088
      - 50070:50070
      - 50075:50075
      - 8188:8188
      - 8042:8042
      - 10000:10000
      - 9083:9083
      - 6080:6080
      - 16010:16010
    volumes:
      - /etc/hosts:/etc/hosts
      - /usr/hdp:/usr/hdp
      - /hadoop/hdfs:/hadoop/hdfs
      - /hadoop/etc/hadoop/:/etc/hadoop/
      - /hadoop/etc/hbase/:/etc/hbase/
      - /hadoop/etc/hive/:/etc/hive/
      - /hadoop/etc/livy2/:/etc/livy2/
      - /hadoop/etc/spark2/:/etc/spark2/
      - /hadoop/etc/tez/:/etc/tez/
      - /hadoop/etc/zookeeper/:/etc/zookeeper/
      - /hadoop/yarn:/hadoop/yarn
      - /hadoop/kerberos/krb5kdc:/var/kerberos/krb5kdc
      - /hadoop/kerberos/keytabs:/etc/security/keytabs
    command:
      - /bin/sh
      - -c
      - |
        /opt/apache/bootstrap.sh
networks:
  hadoop-network:
    external:
      name: hadoop-network

启动

docker-compose -f docker-compose-ambari.yaml up -d

下次安装直接加载镜像，然后执行yaml文件即可

docker load -i ambari-1.0.tar.gz
docker load -i kerberos.tar.gz
docker load -i mysql.tar.gz
docker load -i hdp-repo.tar.gz

脚本

bootstrap.sh

#!/bin/bash

set -x

ambari_db=ambari
myurl=mysql-test
myuser=root
mypwd=winner001
myport=3306
mydb=ipvacloud
java_home=/usr/java/jdk1.8.0_162
mysql_driver=/usr/share/java/mysql-connector-java.jar


basedir=$(cd "$(dirname "$0")"; pwd)

wait_for() {
    echo waiting for $1 to listen on $2...
    while ! nc -z $1 $2; do echo waiting...; sleep 1s; done
}

setup_ambari() {
  # 强制替换文件
  cp -f /tmp/ambari.properties /etc/ambari-server/conf/ 
  cp -f /tmp/password.dat     /etc/ambari-server/conf/
  cp -f /tmp/ambari-agent.ini    /etc/ambari-agent/conf/
}


# wait for mysql and httpd
#wait_for windp-aio 3306
#wait_for httpd 80

# 开始ambari安装
yum -y install ambari-server
yum -y install ambari-agent

re=`mysql -h${myurl} -u${myuser} -p${mypwd} -e "show databases"|grep ambari|wc -l`
if [ $re -eq 0 ];then
   # sql 初始化
   mysql -h${myurl} -u${myuser} -p${mypwd} < /tmp/init_db.sql 
   mysql -h${myurl} -u${myuser} -p${mypwd} ambari  < /tmp/ambari-ddl-mysql-create.sql
fi

# 配置
setup_ambari

# 启动ambari服务
/usr/sbin/ambari-server start
/usr/sbin/ambari-agent start

/usr/sbin/sshd -d

tail -f /var/log/ambari-server/ambari-server.log

参考文章：https://www.cnblogs.com/liugp/p/17489616.html

ubuntu 环境安装docker: 【云计算】ubuntu多种安装docker方式_ubuntu安装docker-csdn博客