demo提供本地化的Zebra使用演示, 通过spring+zebra-dao的方式进行数据库访问，分别可以执行 shard, group, single的功能, 执行入口位于test目录下。

2.环境配置说明

2.1应用名

在src/main/resources目录下创建app.properties作为应用名配置

app.name=zebra_ut

2.2机房信息

在src/main/resource目录下创建region目录,其中包含2个文件center.json idc.json center.json

{
  "regions" : [
    {"region" : "region1", "centers" : [
      {"center":"center1","priority":2, "idcs":[
        {"idc" : "idc1", "desc" : "机房1"},
        {"idc" : "idc2", "desc" : "机房2"},
        {"idc" : "idc3", "desc" : "机房3"}
      ]},
      {"center":"center2","priority":1,"idcs":[
        {"idc" : "idc4", "desc" : "机房4"}
      ]}
    ]},
    {"region" : "region2", "centers" : [
      {"center":"center3","priority":1,"idcs":[
        {"idc" : "idc5", "desc" : "机房5"},
        {"idc" : "idc6", "desc" : "机房6"}
      ]}
    ]}
  ]
}

idc.json

{
    "regions" : [
        {"region" : "region1", "idcs" : [
            {"idc" : "idc1", "net" : ["192.1", "192.2"], "desc" : "机房1"},
            {"idc" : "idc2", "net" : ["192.3", "192.4"], "desc" : "机房2"},
            {"idc" : "idc3", "net" : ["192.5"], "desc" : "机房3"},
            {"idc" : "idc4", "net" : ["192.6"], "desc" : "机房4"}
        ]},
        {"region" : "region2", "idcs" : [
            {"idc" : "idc5", "net" : ["192.10", "192.11"], "desc" : "机房5"},
            {"idc" : "idc6", "net" : ["192.20", "192.21"], "desc" : "机房6"}
        ]}
    ]
}

2.3Zebra系统配置

zebra一些系统配置项，包括限流的配置

zebra_sample.zebra.system.retryTimes=3

2.4.1 本地jdbcRef配置

数据新建完毕后，在src/main/resources目录下创建数据库的配置文件(zebra抽象为jdbcRef，并根据jdbcref找到数据库相关配置，访问数据库)，文件名必须以jdbcRef.properties格式命名，例如jdbcRef="sample",那么创建sample.properties配置文件

zebra.group.sample=\
<groupConfig>\
    <singleConfig>\
        <name>sample-n1</name>\
        <writeWeight>0</writeWeight>\
        <readWeight>-1</readWeight>\
    </singleConfig>\
    <singleConfig>\
        <name>sample-n2</name>\
        <writeWeight>-1</writeWeight>\
        <readWeight>1</readWeight>\
    </singleConfig>\
    <singleConfig>\
        <name>sample-n3</name>\
        <writeWeight>-1</writeWeight>\
        <readWeight>1</readWeight>\
    </singleConfig>\
</groupConfig>

#zebra.ds.sample-n1
zebra.ds.sample-n1={\
  "active" : "true",\
  "url" : "jdbc:mysql://127.0.0.1:3306/sample?characterEncoding=UTF8&socketTimeout=60000",\
  "username" : "root",\
  "password" : "123456",\
  "driverClass":"com.mysql.jdbc.Driver",\
  "properties" : "idleConnectionTestPeriod=80&acquireRetryAttempts=50&acquireRetryDelay=300&maxStatements=1"\
  }

#zebra.ds.sample-n2
zebra.ds.sample-n2={\
  "active" : "true",\
  "url" : "jdbc:mysql://127.0.0.1:3306/sample?characterEncoding=UTF8&socketTimeout=60000",\
  "username" : "root",\
  "password" : "123456",\
  "driverClass":"com.mysql.jdbc.Driver",\
  "properties" : "idleConnectionTestPeriod=80&acquireRetryAttempts=50&acquireRetryDelay=300&maxStatements=1"\
  }

#zebra.ds.sample-n3
zebra.ds.sample-n3 = {\
  "active" : "true",\
  "url" : "jdbc:mysql://127.0.0.1:3306/sample?characterEncoding=UTF8&socketTimeout=60000",\
  "username" : "root",\
  "password" : "123456",\
  "driverClass":"com.mysql.jdbc.Driver",\
  "properties" : "idleConnectionTestPeriod=80&acquireRetryAttempts=50&acquireRetryDelay=300&maxStatements=1"\
  }

2.4.2 zookeeper配置

zkserver=localhost:2181

2.5 使用

在src/test/java/com/dianping/zebra/sample目录下存在3个sample类 分别演示了如何使用spring+zebra的方式访问数据库
GroupDataSourceSample是读写分离的示例
ShardDataSourceSample是分库分表的示例
SingleDataSourceSample是单数据源示例

3.库表结构

3.1建库

create database zebra;

3.2建表

User表

CREATE TABLE `User` (
  `id` int(11) unsigned NOT NULL AUTO_INCREMENT,
  `Name` varchar(32) DEFAULT NULL,
  `Mis` varchar(20) NOT NULL ,
  `Email` varchar(128) DEFAULT NULL,
  `UpdateTime` timestamp NOT NULL DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
  `CreateTime` datetime DEFAULT NULL,
  PRIMARY KEY (`id`),
  UNIQUE KEY `IX_Mis` (`Mis`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;

Feed分表 表名为Feed0,Feed1,Feed2,Feed3

CREATE TABLE `Feed0` (
  `id` int(11) unsigned NOT NULL AUTO_INCREMENT,
  `Uid` varchar(30) NOT NULL DEFAULT '',
  `UpdateTime` timestamp NULL DEFAULT NULL,
  PRIMARY KEY (`id`)
) ENGINE=InnoDB DEFAULT CHARSET=utf8;