elasticjob-cloud-scheduler 3节点部署（Docker-compose部署）

资源清单

依赖 ZooKeeper 集群，需要先部署 ZooKeeper

ZooKeeper集群部署

依赖 Mesos 集群，需要先部署 Mesos

Mesos集群部署

主机 IP
elasticjob节点1 10.0.0.1
elasticjob节点2 10.0.0.2
elasticjob节点3 10.0.0.3
数据库节点 10.0.0.4
软件 版本
docker 20.10.12
docker-compose 1.23.1
mesos镜像 mesosphere/mesos-master:1.7.1
elasticjob elasticjob-3.0.1
服务 端口
elasticjob 8899

一、Docker 安装

1. 使用国内 yum 源

# yum install -y yum-utils device-mapper-persistent-data lvm2
# yum-config-manager --add-repo https://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo

2. 卸载旧版本的 docker

## 如果主机上已经有docker存在且不是想要安装的版本，需要先进行卸载。
# yum remove -y docker \
              docker-client \
              docker-client-latest \
              docker-common \
              docker-latest \
              docker-latest-logrotate \
              docker-logrotate \
              docker-selinux \
              docker-engine-selinux \
              docker-engine \
              container*

3. 安装 Docker20.10 版本

# yum -y install docker-ce-20.10.12-3.el7 docker-ce-cli-20.10.12-3.el7 vim

4. 设置镜像加速

# mkdir /etc/docker
# vi /etc/docker/daemon.json

{
  "registry-mirrors": ["https://xxxxxxxxx.mirror.aliyuncs.com"]
}

5. 启动 docker

# systemctl start docker
# systemctl enable docker
# systemctl status docker

二、Docker-compose 安装

1. Docker-compose 安装

## github.com 可能访问超时，可以使用下面的获取下载下来后上传服务器即可
# curl -L "https://github.com/docker/compose/releases/download/1.29.2/docker-compose-$(uname -s)-$(uname -m)" -o /usr/local/bin/docker-compose

# curl -k "https://dl.cactifans.com/zabbix_docker/docker-compose" -o /usr/bin/docker-compose

# chmod a+x /usr/bin/docker-compose

2. 查看 docker-compose 版本

# docker-compose version

三、elasticjob-cloud-scheduler 安装

elasticjob-cloud-scheduler 启动依赖 mesos 底层库，而 mesos使用 Docker部署，而需要使用含有 mesos 底层库的镜像，制作 elasticjob 镜像

1. 详细的 Dockerfile 文件信息

FROM mesosphere/mesos:1.7.1
MAINTAINER evescn

RUN apt-get update && \
  apt-get -y install openjdk-8-jdk && \
  java -version

ADD apache-shardingsphere-elasticjob-3.0.1-cloud-scheduler-bin.tar.gz /
COPY mysql-connector-java-8.0.29.jar  /apache-shardingsphere-elasticjob-3.0.1-cloud-scheduler-bin/lib/

ENTRYPOINT ["sh", "/apache-shardingsphere-elasticjob-3.0.1-cloud-scheduler-bin/bin/start.sh"]

2. 构建镜像

# mkidr /root/elasticjob-cloud-scheduler -pv
# cd /root/elasticjob-cloud-scheduler

## 下载 elasticjob 文件
# wget https://archive.apache.org/dist/shardingsphere/elasticjob-3.0.1/apache-shardingsphere-elasticjob-3.0.1-cloud-scheduler-bin.tar.gz

## 下载 mysql-connector-java-8.0.29.jar 文件
# wget https://repo1.maven.org/maven2/mysql/mysql-connector-java/8.0.29/mysql-connector-java-8.0.29.jar

## 准备 dockerfile 文件
# cat << EOF >> Dockerfile

FROM mesosphere/mesos:1.7.1
MAINTAINER evescn

RUN apt-get update && \
  apt-get -y install openjdk-8-jdk && \
  java -version

ADD apache-shardingsphere-elasticjob-3.0.1-cloud-scheduler-bin.tar.gz /
COPY mysql-connector-java-8.0.29.jar  /apache-shardingsphere-elasticjob-3.0.1-cloud-scheduler-bin/lib/

ENTRYPOINT ["sh", "/apache-shardingsphere-elasticjob-3.0.1-cloud-scheduler-bin/bin/start.sh"]

EOF

## 构建镜像
# docker build -t elasticjob:v3.0.1_20220428 .

3. 分发镜像到节点

# docker save elasticjob:v3.0.1_20220411 > elasticjob_v3.0.1_20220411.tar

## 连接到3台主机分别执行下面的操作
# scp elasticjob_v3.0.1_20220411.tar 10.0.0.1:/root
# ssh 10.0.0.1
# docker load < elasticjob_v3.0.1_20220411.tar

4. 创建 Event trace 日志记录数据库

mysql 数据库部署在 10.0.0.4 主机，如何部署 mysql 服务，请自行百度

## 连接数据库，执行以下 SQL 语句

create database elastic_job_cloud_log default character set utf8mb4 collate utf8mb4_unicode_ci;

CREATE TABLE `job_execution_log` (
  `id` varchar(40) NOT NULL,
  `job_name` varchar(100) NOT NULL,
  `task_id` varchar(255) NOT NULL,
  `hostname` varchar(255) NOT NULL,
  `ip` varchar(50) NOT NULL,
  `sharding_item` int(11) NOT NULL,
  `execution_source` varchar(20) NOT NULL,
  `failure_cause` varchar(4000) DEFAULT NULL,
  `is_success` int(11) NOT NULL,
  `start_time` timestamp NULL DEFAULT NULL,
  `complete_time` timestamp NULL DEFAULT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;


CREATE TABLE `job_status_trace_log` (
  `id` varchar(40) NOT NULL,
  `job_name` varchar(100) NOT NULL,
  `original_task_id` varchar(255) NOT NULL,
  `task_id` varchar(255) NOT NULL,
  `slave_id` varchar(50) NOT NULL,
  `source` varchar(50) NOT NULL,
  `execution_type` varchar(20) NOT NULL,
  `sharding_item` varchar(100) NOT NULL,
  `state` varchar(20) NOT NULL,
  `message` varchar(4000) DEFAULT NULL,
  `creation_time` timestamp NULL DEFAULT NULL,
  PRIMARY KEY (`id`),
  KEY `TASK_ID_STATE_INDEX` (`task_id`,`state`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

5. 启动服务

3节点 均需要执行以下步骤

a | 生成 docker-compose.yml 文件

# mkidr /data/elasticjob-cloud-scheduler/{bin,conf} -pv
# cd /data/elasticjob-cloud-scheduler

# cat << EOF >> docker-compose.yml
version: '3'
services:
  elasticjob-cloud-scheduler:
    image: elasticjob:v3.0.1_20220411
    restart: always
    network_mode: "host"
    hostname: elasticjob-cloud-scheduler-1
    container_name: elasticjob-cloud-scheduler-1
    volumes: # 挂载数据
      - /data/elasticjob-cloud-scheduler/bin/start.sh:/apache-shardingsphere-elasticjob-3.0.1-cloud-scheduler-bin/bin/start.sh
      - /data/elasticjob-cloud-scheduler/conf/elasticjob-cloud-scheduler.properties:/apache-shardingsphere-elasticjob-3.0.1-cloud-scheduler-bin/conf/elasticjob-cloud-scheduler.properties
    extra_hosts:  # 设置容器 hosts
      - "zk1:10.0.0.1"
      - "zk2:10.0.0.3"
      - "zk3:10.0.0.2"
EOF

b | 解压服务包，获取配置文件

解压 步骤3中 tar.gz 包，负责软件 bin/ 和 conf 下配置文件到 /data/elasticjob-cloud-scheduler/

# cd /root/elasticjob-cloud-scheduler

# tar -xvf apache-shardingsphere-elasticjob-3.0.1-cloud-scheduler-bin.tar.gz
# cp apache-shardingsphere-elasticjob-3.0.1-cloud-scheduler-bin/bin/start.sh /data/elasticjob-cloud-scheduler/bin
# cp apache-shardingsphere-elasticjob-3.0.1-cloud-scheduler-bin/conf/elasticjob-cloud-scheduler.properties /data/elasticjob-cloud-scheduler/conf/

c | 修改启动脚本

# cat bin/start.sh

## 容器运行 source 报错，修改为 .
......
#source ${CONF_DIR}/elasticjob-cloud-scheduler.properties
. ${CONF_DIR}/elasticjob-cloud-scheduler.properties
......

d | 修改配置文件

# cat conf/elasticjob-cloud-scheduler.properties

# Routable IP address
hostname=10.0.0.1           ## 其他服务器，修改为对应IP即可

# Username for mesos framework
user=

# Mesos zookeeper address
mesos_url=zk://javaDev:edocyunjavaDev@zk1:2181,zk2:2181,zk3:2181/mesos

# Role for mesos framework

#mesos_role=

# ElasticJob-Cloud's zookeeper address
zk_servers=zk1:2181,zk2:2181,zk3:2181

# ElasticJob-Cloud's zookeeper namespace
zk_namespace=elasticjob-cloud

# ElasticJob-Cloud's zookeeper digest
zk_digest=user:123456

# Job rest API port
http_port=8899

# Max size of job accumulated
job_state_queue_size=10000

# Event trace rdb config

event_trace_rdb_driver=com.mysql.jdbc.Driver
event_trace_rdb_url=jdbc:mysql://10.0.0.4:3306/elastic_job_cloud_log
event_trace_rdb_username=root
event_trace_rdb_password=root

# Task reconciliation interval

#reconcile_interval_minutes=-1

# Enable/Disable mesos partition aware feature

# enable_partition_aware=false

# Auth username
auth_username=root

# Auth password
auth_password=pwd
EOF

e | 启动服务

# docker-compose up -d

6. 查看服务

登陆 Mesos web 界面中，查看 frameworks 页面

访问任意节点url地址：http://10.0.0.1:5050/ ，会自动跳转到 Leader节点，再访问 http://master:5050/#/frameworks

posted @ 2022-04-28 16:31  evescn  阅读(553)  评论(0)  编辑  收藏  举报