4.5萬字手把手教你實現MySQL TB級資料儲存！！

冰河團隊發表於2021-01-10
MySql

寫在前面

業界對系統的高可用有著基本的要求，簡單的說，這些要求可以總結為如下所示。

  • 系統架構中不存在單點問題。
  • 可以最大限度的保障服務的可用性。

一般情況下系統的高可用可以用幾個9來評估。所謂的幾個9就是系統可以保證對外提供的服務的時間達到總時間的百分比。例如如果需要達到99.99的高可用，則系統全年發生故障的總時間不能超過52分鐘。

系統高可用架構

我們既然需要實現系統的高可用架構，那麼，我們到底需要搭建一個什麼樣的系統架構呢？我們可以將需要搭建的系統架構簡化成下圖所示。

伺服器規劃

由於我電腦資源有限，我這裡在4臺伺服器上搭建高可用環境，大家可以按照本文將環境擴充套件到更多的伺服器，搭建步驟都是一樣的。

主機名 IP地址 安裝的服務
binghe151 192.168.175.151 Mycat、Zookeeper、MySQL、HAProxy、Keepalived、Xinetd
binghe152 192.168.175.152 Zookeeper、MySQL
binghe153 192.168.175.153 Zookeeper、MySQL
binghe154 192.168.175.154 Mycat、MySQL、HAProxy、Keepalived、Xinetd
binghe155 192.168.175.155 MySQL

注意：HAProxy和Keepalived最好和Mycat部署在同一臺伺服器上。

安裝MySQL

小夥伴們可以關注【冰河技術】微信公眾號，參考《MySQL之——原始碼編譯MySQL8.x+升級gcc+升級cmake（親測完整版）》

安裝JDK

由於Mycat和Zookeeper的執行需要JDK環境的支援，所有我們需要在每臺伺服器上安裝JDK環境。

這裡，我以在binghe151伺服器上安裝JDK為例，其他伺服器的安裝方式與在binghe151伺服器上的安裝方式相同。安裝步驟如下所示。

（1）到JDK官網下載JDK 1.8版本，JDK1.8的下載地址為：https://www.oracle.com/technetwork/java/javase/downloads/jdk8-downloads-2133151.html。

注：我下載的JDK安裝包版本為：jdk-8u212-linux-x64.tar.gz，如果JDK版本已更新，大家下載對應的版本即可。

（2）將下載的jdk-8u212-linux-x64.tar.gz安裝包上傳到binghe151伺服器的/usr/local/src目錄下。

（3）解壓jdk-8u212-linux-x64.tar.gz檔案，如下所示。

tar -zxvf jdk-8u212-linux-x64.tar.gz

（4）將解壓的jdk1.8.0_212目錄移動到binghe151伺服器下的/usr/local目錄下，如下所示。

mv jdk1.8.0_212/ /usr/local/src/

（5）配置JDK系統環境變數，如下所示。

vim /etc/profile
JAVA_HOME=/usr/local/jdk1.8.0_212
CLASS_PATH=.:$JAVA_HOME/lib
PATH=$JAVA_HOME/bin:$PATH
export JAVA_HOME CLASS_PATH PATH

使系統環境變數生效，如下所示。

source /etc/profile

（6）檢視JDK版本，如下所示。

[root@binghe151 ~]# java -version
java version "1.8.0_212"
Java(TM) SE Runtime Environment (build 1.8.0_212-b10)
Java HotSpot(TM) 64-Bit Server VM (build 25.212-b10, mixed mode)

結果顯示，正確輸出了JDK的版本資訊，說明JDK安裝成功。

安裝Mycat

下載Mycat 1.6.7.4 Release版本，解壓到伺服器的/usr/local/mycat目錄下，並配置Mycat的系統環境變數，隨後，配置Mycat的配置檔案，Mycat的最終結果配置如下所示。

  • schema.xml
<?xml version="1.0"?>
<!DOCTYPE mycat:schema SYSTEM "schema.dtd">
<mycat:schema xmlns:mycat="http://io.mycat/">

	<schema name="shop" checkSQLschema="false" sqlMaxLimit="1000">
		<!--<table name="order_master" primaryKey="order_id" dataNode = "ordb"/>-->
		<table name="order_master" primaryKey="order_id" dataNode = "orderdb01,orderdb02,orderdb03,orderdb04" rule="order_master" autoIncrement="true">
			<childTable name="order_detail" primaryKey="order_detail_id" joinKey="order_id" parentKey="order_id" autoIncrement="true"/>
		</table>
		<table name="order_cart" primaryKey="cart_id" dataNode = "ordb"/>
		<table name="order_customer_addr" primaryKey="customer_addr_id" dataNode = "ordb"/>
		<table name="region_info" primaryKey="region_id" dataNode = "ordb,prodb,custdb" type="global"/>
		<table name="serial" primaryKey="id" dataNode = "ordb"/>
		<table name="shipping_info" primaryKey="ship_id" dataNode = "ordb"/>
		<table name="warehouse_info" primaryKey="w_id" dataNode = "ordb"/>
		<table name="warehouse_proudct" primaryKey="wp_id" dataNode = "ordb"/>
		
		<table name="product_brand_info" primaryKey="brand_id" dataNode = "prodb"/>
		<table name="product_category" primaryKey="category_id" dataNode = "prodb"/>
		<table name="product_comment" primaryKey="comment_id" dataNode = "prodb"/>
		<table name="product_info" primaryKey="product_id" dataNode = "prodb"/>
		<table name="product_pic_info" primaryKey="product_pic_id" dataNode = "prodb"/>
		<table name="product_supplier_info" primaryKey="supplier_id" dataNode = "prodb"/>
		
		<table name="customer_balance_log" primaryKey="balance_id" dataNode = "custdb"/>
		<table name="customer_inf" primaryKey="customer_inf_id" dataNode = "custdb"/>
		<table name="customer_level_inf" primaryKey="customer_level" dataNode = "custdb"/>
		<table name="customer_login" primaryKey="customer_id" dataNode = "custdb"/>
		<table name="customer_login_log" primaryKey="login_id" dataNode = "custdb"/>
		<table name="customer_point_log" primaryKey="point_id" dataNode = "custdb"/>
		
	</schema>
	
	<dataNode name="mycat" dataHost="binghe151" database="mycat" />
	 
	<dataNode name="ordb" dataHost="binghe152" database="order_db" />
	<dataNode name="prodb" dataHost="binghe153" database="product_db" />
	<dataNode name="custdb" dataHost="binghe154" database="customer_db" />
	
	<dataNode name="orderdb01" dataHost="binghe152" database="orderdb01" />
	<dataNode name="orderdb02" dataHost="binghe152" database="orderdb02" />
	<dataNode name="orderdb03" dataHost="binghe153" database="orderdb03" />
	<dataNode name="orderdb04" dataHost="binghe153" database="orderdb04" />
	
	<dataHost name="binghe151" maxCon="1000" minCon="10" balance="1"
			  writeType="0" dbType="mysql" dbDriver="native" switchType="1"  slaveThreshold="100">
		<heartbeat>select user()</heartbeat>
		<writeHost host="binghe51" url="192.168.175.151:3306" user="mycat" password="mycat"/>
	</dataHost>
	
	<dataHost name="binghe152" maxCon="1000" minCon="10" balance="1"
			  writeType="0" dbType="mysql" dbDriver="native" switchType="1"  slaveThreshold="100">
		<heartbeat>select user()</heartbeat>
		<writeHost host="binghe52" url="192.168.175.152:3306" user="mycat" password="mycat"/>
	</dataHost>
	
	<dataHost name="binghe153" maxCon="1000" minCon="10" balance="1"
			  writeType="0" dbType="mysql" dbDriver="native" switchType="1"  slaveThreshold="100">
		<heartbeat>select user()</heartbeat>
		<writeHost host="binghe53" url="192.168.175.153:3306" user="mycat" password="mycat"/>
	</dataHost>
	
	<dataHost name="binghe154" maxCon="1000" minCon="10" balance="1"
			  writeType="0" dbType="mysql" dbDriver="native" switchType="1"  slaveThreshold="100">
		<heartbeat>select user()</heartbeat>
		<writeHost host="binghe54" url="192.168.175.154:3306" user="mycat" password="mycat"/>
	</dataHost>
	
</mycat:schema>
  • server.xml
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE mycat:server SYSTEM "server.dtd">
<mycat:server xmlns:mycat="http://io.mycat/">
	<system>
		<property name="useHandshakeV10">1</property>
        <property name="defaultSqlParser">druidparser</property>
		<property name="serverPort">3307</property>
		<property name="managerPort">3308</property>
		<property name="nonePasswordLogin">0</property>
		<property name="bindIp">0.0.0.0</property>
		<property name="charset">utf8mb4</property>
		<property name="frontWriteQueueSize">2048</property>
		<property name="txIsolation">2</property>
		<property name="processors">2</property>
		<property name="idleTimeout">1800000</property>
		<property name="sqlExecuteTimeout">300</property>
		<property name="useSqlStat">0</property>
		<property name="useGlobleTableCheck">0</property>
		<property name="sequenceHandlerType">1</property>
		<property name="defaultMaxLimit">1000</property>
		<property name="maxPacketSize">104857600</property>
		
		<property name="sqlInterceptor">
			io.mycat.server.interceptor.impl.StatisticsSqlInterceptor
		</property>
		<property name="sqlInterceptorType">
			UPDATE,DELETE,INSERT
		</property>
		<property name="sqlInterceptorFile">/tmp/sql.txt</property>
	</system>
	
	<firewall>
		<whitehost>
			<host user="mycat" host="192.168.175.151"></host>
		</whitehost>
		<blacklist check="true">
			<property name="noneBaseStatementAllow">true</property>
			<property name="deleteWhereNoneCheck">true</property>
		</blacklist>
	</firewall>
	
	<user name="mycat" defaultAccount="true">
		<property name="usingDecrypt">1</property>
		<property name="password">cTwf23RrpBCEmalp/nx0BAKenNhvNs2NSr9nYiMzHADeEDEfwVWlI6hBDccJjNBJqJxnunHFp5ae63PPnMfGYA==</property>
		<property name="schemas">shop</property>
	</user>

</mycat:server>
  • rule.xml
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE mycat:rule SYSTEM "rule.dtd">
<mycat:rule xmlns:mycat="http://io.mycat/">
	<tableRule name="order_master">
		<rule>
			<columns>customer_id</columns>
			<algorithm>mod-long</algorithm>
		</rule>
	</tableRule>
	
	<function name="mod-long" class="io.mycat.route.function.PartitionByMod">
		<property name="count">4</property>
	</function>
</mycat:rule>
  • sequence_db_conf.properties
#sequence stored in datanode
GLOBAL=mycat
ORDER_MASTER=mycat
ORDER_DETAIL=mycat

關於Mycat的配置，僅供大家參考，大家不一定非要按照我這裡配置，根據自身業務需要配置即可。本文的重點是實現Mycat的高可用環境搭建。

在MySQL中建立Mycat連線MySQL的賬戶，如下所示。

CREATE USER 'mycat'@'192.168.175.%' IDENTIFIED BY 'mycat';
ALTER USER 'mycat'@'192.168.175.%' IDENTIFIED WITH mysql_native_password BY 'mycat'; 
GRANT SELECT, INSERT, UPDATE, DELETE,EXECUTE  ON *.* TO 'mycat'@'192.168.175.%';
FLUSH PRIVILEGES;

安裝Zookeeper叢集

安裝配置完JDK後，就需要搭建Zookeeper叢集了，根據對伺服器的規劃，現將Zookeeper叢集搭建在“binghe151”、“binghe152”、“binghe153”三臺伺服器上。

1.下載Zookeeper

到Apache官網去下載Zookeeper的安裝包，Zookeeper的安裝包下載地址為：https://mirrors.tuna.tsinghua.edu.cn/apache/zookeeper/。具體如下圖所示。

也可以在binghe151伺服器上執行如下命令直接下載zookeeper-3.5.5。

wget https://mirrors.tuna.tsinghua.edu.cn/apache/zookeeper/zookeeper-3.5.5/apache-zookeeper-3.5.5-bin.tar.gz

執行上述命令就可以直接把apache-zookeeper-3.5.5-bin.tar.gz安裝包下載到binghe151伺服器上。

2.安裝並配置Zookeeper

注意：（1）、（2）、（3）步都是在binghe152伺服器上執行的。

（1）解壓Zookeeper安裝包

在binghe151伺服器上執行如下命令，將Zookeeper解壓到“/usr/local/”目錄下，並將Zookeeper目錄修改為zookeeper-3.5.5。

tar -zxvf apache-zookeeper-3.5.5-bin.tar.gz
mv apache-zookeeper-3.5.5-bin zookeeper-3.5.5

（2）配置Zookeeper系統環境變數

同樣，需要在/etc/profile檔案中配置Zookeeper系統環境變數，如下：

ZOOKEEPER_HOME=/usr/local/zookeeper-3.5.5
PATH=$ZOOKEEPER_HOME/bin:$PATH
export ZOOKEEPER_HOME PATH

結合之前配置的JDK系統環境變數，/etc/profile，總體配置如下：

MYSQL_HOME=/usr/local/mysql
JAVA_HOME=/usr/local/jdk1.8.0_212
MYCAT_HOME=/usr/local/mycat
ZOOKEEPER_HOME=/usr/local/zookeeper-3.5.5
MPC_HOME=/usr/local/mpc-1.1.0
GMP_HOME=/usr/local/gmp-6.1.2
MPFR_HOME=/usr/local/mpfr-4.0.2
CLASS_PATH=.:$JAVA_HOME/lib
LD_LIBRARY_PATH=$MPC_LIB_HOME/lib:$GMP_HOME/lib:$MPFR_HOME/lib:$LD_LIBRARY_PATH
PATH=$MYSQL_HOME/bin:$JAVA_HOME/bin:$ZOOKEEPER_HOME/bin:$MYCAT_HOME/bin:$PATH
export JAVA_HOME ZOOKEEPER_HOME MYCAT_HOME CLASS_PATH MYSQL_HOME MPC_LIB_HOME GMP_HOME MPFR_HOME LD_LIBRARY_PATH PATH

（3）配置Zookeeper

首先，需要將$ZOOKEEPER_HOME/conf（$ZOOKEEPER_HOME為Zookeeper的安裝目錄）目錄下的zoo_sample.cfg檔案修改為zoo.cfg檔案。具體命令如下：

cd /usr/local/zookeeper-3.5.5/conf/
mv zoo_sample.cfg zoo.cfg

接下來修改zoo.cfg檔案，修改後的具體內容如下：

tickTime=2000
initLimit=10
syncLimit=5
dataDir=/usr/local/zookeeper-3.5.5/data
dataLogDir=/usr/local/zookeeper-3.5.5/dataLog
clientPort=2181
server.1=binghe151:2888:3888
server.2=binghe152:2888:3888
server.3=binghe153:2888:3888

在Zookeeper的安裝目錄下建立data和dataLog兩個資料夾。

mkdir -p /usr/local/zookeeper-3.5.5/data
mkdir -p /usr/local/zookeeper-3.5.5/dataLog

切換到新建的data目錄下，建立myid檔案，具體內容為數字1，如下所示：

cd /usr/local/zookeeper-3.5.5/data
vim myid

將數字1寫入到檔案myid。

3.將Zookeeper和系統環境變數檔案複製到其他伺服器

注意：（1）、（2）步是在binghe151伺服器上執行的。

（1）複製Zookeeper到其他伺服器

根據對伺服器的規劃，現將Zookeeper複製到binghe152和binghe53伺服器，具體執行操作如下所示：

scp -r /usr/local/zookeeper-3.5.5/ binghe152:/usr/local/
scp -r /usr/local/zookeeper-3.5.5/ binghe153:/usr/local/

（2）複製系統環境變數檔案到其他伺服器

根據對伺服器的規劃，現將系統環境變數檔案/etc/profile複製到binghe152、binghe153伺服器，具體執行操作如下所示：

scp /etc/profile binghe152:/etc/
scp /etc/profile binghe153:/etc/

上述操作可能會要求輸入密碼，根據提示輸入密碼即可。

4.修改其他伺服器上的myid檔案

修改binghe152伺服器上Zookeeper的myid檔案內容為數字2，同時修改binghe153伺服器上Zookeeper的myid檔案內容為數字3。具體如下：

在binghe152伺服器上執行如下操作：

echo "2" > /usr/local/zookeeper-3.5.5/data/myid
cat /usr/local/zookeeper-3.5.5/data/myid
2

在binghe153伺服器上執行如下操作：

echo "3" > /usr/local/zookeeper-3.5.5/data/myid
cat /usr/local/zookeeper-3.5.5/data/myid
3

5.使環境變數生效

分別在binghe151、binghe152、binghe153上執行如下操作，使系統環境變數生效。

source /etc/profile

6.啟動Zookeeper叢集

分別在binghe151、binghe152、binghe153上執行如下操作，啟動Zookeeper叢集。

zkServer.sh start

7.檢視Zookeeper叢集的啟動狀態

  • binghe151伺服器
[root@binghe151 ~]# zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /usr/local/zookeeper-3.5.5/bin/../conf/zoo.cfg
Client port found: 2181. Client address: localhost.
Mode: follower
  • binghe152伺服器
[root@binghe152 local]# zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /usr/local/zookeeper-3.5.5/bin/../conf/zoo.cfg
Client port found: 2181. Client address: localhost.
Mode: leader
  • binghe153伺服器
[root@binghe153 ~]# zkServer.sh status
ZooKeeper JMX enabled by default
Using config: /usr/local/zookeeper-3.5.5/bin/../conf/zoo.cfg
Client port found: 2181. Client address: localhost.
Mode: follower

可以看到，binghe151和binghe153伺服器上的Zookeeper角色為follower，binghe152伺服器上的Zookeeper角色為leader。

初始化Mycat配置到Zookeeper叢集

注意：初始化Zookeeper中的資料，是在binghe151伺服器上進行的，原因是之前我們已經在binghe151伺服器上安裝了Mycat。

1.檢視初始化指令碼

在Mycat安裝目錄下的bin目錄中提供了一個init_zk_data.sh指令碼檔案，如下所示。

[root@binghe151 ~]# ll /usr/local/mycat/bin/
total 384
-rwxr-xr-x 1 root root   3658 Feb 26 17:10 dataMigrate.sh
-rwxr-xr-x 1 root root   1272 Feb 26 17:10 init_zk_data.sh
-rwxr-xr-x 1 root root  15701 Feb 28 20:51 mycat
-rwxr-xr-x 1 root root   2986 Feb 26 17:10 rehash.sh
-rwxr-xr-x 1 root root   2526 Feb 26 17:10 startup_nowrap.sh
-rwxr-xr-x 1 root root 140198 Feb 28 20:51 wrapper-linux-ppc-64
-rwxr-xr-x 1 root root  99401 Feb 28 20:51 wrapper-linux-x86-32
-rwxr-xr-x 1 root root 111027 Feb 28 20:51 wrapper-linux-x86-64

init_zk_data.sh指令碼檔案就是用來向Zookeeper中初始化Mycat的配置的，這個檔案會通過讀取Mycat安裝目錄下的conf目錄下的配置檔案，將其初始化到Zookeeper叢集中。

2.複製Mycat配置檔案

首先，我們檢視下Mycat安裝目錄下的conf目錄下的檔案資訊，如下所示。

[root@binghe151 ~]# cd /usr/local/mycat/conf/
[root@binghe151 conf]# ll
total 108
-rwxrwxrwx 1 root root   92 Feb 26 17:10 autopartition-long.txt
-rwxrwxrwx 1 root root   51 Feb 26 17:10 auto-sharding-long.txt
-rwxrwxrwx 1 root root   67 Feb 26 17:10 auto-sharding-rang-mod.txt
-rwxrwxrwx 1 root root  340 Feb 26 17:10 cacheservice.properties
-rwxrwxrwx 1 root root 3338 Feb 26 17:10 dbseq.sql
-rwxrwxrwx 1 root root 3532 Feb 26 17:10 dbseq - utf8mb4.sql
-rw-r--r-- 1 root root   86 Mar  1 22:37 dnindex.properties
-rwxrwxrwx 1 root root  446 Feb 26 17:10 ehcache.xml
-rwxrwxrwx 1 root root 2454 Feb 26 17:10 index_to_charset.properties
-rwxrwxrwx 1 root root 1285 Feb 26 17:10 log4j2.xml
-rwxrwxrwx 1 root root  183 Feb 26 17:10 migrateTables.properties
-rwxrwxrwx 1 root root  271 Feb 26 17:10 myid.properties
-rwxrwxrwx 1 root root   16 Feb 26 17:10 partition-hash-int.txt
-rwxrwxrwx 1 root root  108 Feb 26 17:10 partition-range-mod.txt
-rwxrwxrwx 1 root root  988 Mar  1 16:59 rule.xml
-rwxrwxrwx 1 root root 3883 Mar  3 23:59 schema.xml
-rwxrwxrwx 1 root root  440 Feb 26 17:10 sequence_conf.properties
-rwxrwxrwx 1 root root   84 Mar  3 23:52 sequence_db_conf.properties
-rwxrwxrwx 1 root root   29 Feb 26 17:10 sequence_distributed_conf.properties
-rwxrwxrwx 1 root root   28 Feb 26 17:10 sequence_http_conf.properties
-rwxrwxrwx 1 root root   53 Feb 26 17:10 sequence_time_conf.properties
-rwxrwxrwx 1 root root 2420 Mar  4 15:14 server.xml
-rwxrwxrwx 1 root root   18 Feb 26 17:10 sharding-by-enum.txt
-rwxrwxrwx 1 root root 4251 Feb 28 20:51 wrapper.conf
drwxrwxrwx 2 root root 4096 Feb 28 21:17 zkconf
drwxrwxrwx 2 root root 4096 Feb 28 21:17 zkdownload

接下來，將Mycat安裝目錄下的conf目錄下的schema.xml檔案、server.xml檔案、rule.xml檔案和sequence_db_conf.properties檔案複製到conf目錄下的zkconf目錄下，如下所示。

cp schema.xml server.xml rule.xml sequence_db_conf.properties zkconf/

3.將Mycat配置資訊寫入Zookeeper叢集

執行init_zk_data.sh指令碼檔案，向Zookeeper叢集中初始化配置資訊，如下所示。

[root@binghe151 bin]# /usr/local/mycat/bin/init_zk_data.sh  
o2020-03-08 20:03:13 INFO JAVA_CMD=/usr/local/jdk1.8.0_212/bin/java
o2020-03-08 20:03:13 INFO Start to initialize /mycat of ZooKeeper
o2020-03-08 20:03:14 INFO Done

根據以上資訊得知，Mycat向Zookeeper寫入初始化配置資訊成功。

4.驗證Mycat配置資訊是否成功寫入Mycat

我們可以使用Zookeeper的客戶端命令zkCli.sh 登入Zookeeper來驗證Mycat的配置資訊是否成功寫入Mycat。

首先，登入Zookeeper，如下所示。

[root@binghe151 ~]# zkCli.sh 
Connecting to localhost:2181
###################此處省略N行輸出######################
Welcome to ZooKeeper!

WATCHER::

WatchedEvent state:SyncConnected type:None path:null
[zk: localhost:2181(CONNECTED) 0]

接下來，在Zookeeper命令列檢視mycat的資訊，如下所示。

[zk: localhost:2181(CONNECTED) 0] ls /
[mycat, zookeeper]
[zk: localhost:2181(CONNECTED) 1] ls /mycat
[mycat-cluster-1]
[zk: localhost:2181(CONNECTED) 2] ls /mycat/mycat-cluster-1
[cache, line, rules, schema, sequences, server]
[zk: localhost:2181(CONNECTED) 3]

可以看到，在/mycat/mycat-cluster-1下存在6個目錄，接下來，檢視下schema目錄下的資訊，如下所示。

[zk: localhost:2181(CONNECTED) 3] ls /mycat/mycat-cluster-1/schema
[dataHost, dataNode, schema]

接下來，我們檢視下dataHost的配置，如下所示。

[zk: localhost:2181(CONNECTED) 4] get /mycat/mycat-cluster-1/schema/dataHost
[{"balance":1,"maxCon":1000,"minCon":10,"name":"binghe151","writeType":0,"switchType":1,"slaveThreshold":100,"dbType":"mysql","dbDriver":"native","heartbeat":"select user()","writeHost":[{"host":"binghe51","url":"192.168.175.151:3306","password":"root","user":"root"}]},{"balance":1,"maxCon":1000,"minCon":10,"name":"binghe152","writeType":0,"switchType":1,"slaveThreshold":100,"dbType":"mysql","dbDriver":"native","heartbeat":"select user()","writeHost":[{"host":"binghe52","url":"192.168.175.152:3306","password":"root","user":"root"}]},{"balance":1,"maxCon":1000,"minCon":10,"name":"binghe153","writeType":0,"switchType":1,"slaveThreshold":100,"dbType":"mysql","dbDriver":"native","heartbeat":"select user()","writeHost":[{"host":"binghe53","url":"192.168.175.153:3306","password":"root","user":"root"}]},{"balance":1,"maxCon":1000,"minCon":10,"name":"binghe154","writeType":0,"switchType":1,"slaveThreshold":100,"dbType":"mysql","dbDriver":"native","heartbeat":"select user()","writeHost":[{"host":"binghe54","url":"192.168.175.154:3306","password":"root","user":"root"}]}]

上面的輸出資訊格式比較亂，但可以看出是Json格式的資訊，我們可以將輸出資訊進行格式化，格式化後的結果如下所示。

[
    {
        "balance": 1,
        "maxCon": 1000,
        "minCon": 10,
        "name": "binghe151",
        "writeType": 0,
        "switchType": 1,
        "slaveThreshold": 100,
        "dbType": "mysql",
        "dbDriver": "native",
        "heartbeat": "select user()",
        "writeHost": [
            {
                "host": "binghe51",
                "url": "192.168.175.151:3306",
                "password": "root",
                "user": "root"
            }
        ]
    },
    {
        "balance": 1,
        "maxCon": 1000,
        "minCon": 10,
        "name": "binghe152",
        "writeType": 0,
        "switchType": 1,
        "slaveThreshold": 100,
        "dbType": "mysql",
        "dbDriver": "native",
        "heartbeat": "select user()",
        "writeHost": [
            {
                "host": "binghe52",
                "url": "192.168.175.152:3306",
                "password": "root",
                "user": "root"
            }
        ]
    },
    {
        "balance": 1,
        "maxCon": 1000,
        "minCon": 10,
        "name": "binghe153",
        "writeType": 0,
        "switchType": 1,
        "slaveThreshold": 100,
        "dbType": "mysql",
        "dbDriver": "native",
        "heartbeat": "select user()",
        "writeHost": [
            {
                "host": "binghe53",
                "url": "192.168.175.153:3306",
                "password": "root",
                "user": "root"
            }
        ]
    },
    {
        "balance": 1,
        "maxCon": 1000,
        "minCon": 10,
        "name": "binghe154",
        "writeType": 0,
        "switchType": 1,
        "slaveThreshold": 100,
        "dbType": "mysql",
        "dbDriver": "native",
        "heartbeat": "select user()",
        "writeHost": [
            {
                "host": "binghe54",
                "url": "192.168.175.154:3306",
                "password": "root",
                "user": "root"
            }
        ]
    }
]

可以看到，我們在Mycat的schema.xml檔案中配置的dataHost節點的資訊，成功寫入到Zookeeper中了。

為了驗證Mycat的配置資訊，是否已經同步到Zookeeper的其他節點上，我們也可以在binghe152和binghe153伺服器上登入Zookeeper，檢視Mycat配置資訊是否寫入成功。

  • binghe152伺服器
[root@binghe152 ~]# zkCli.sh 
Connecting to localhost:2181
#################省略N行輸出資訊################
[zk: localhost:2181(CONNECTED) 0] get /mycat/mycat-cluster-1/schema/dataHost
[{"balance":1,"maxCon":1000,"minCon":10,"name":"binghe151","writeType":0,"switchType":1,"slaveThreshold":100,"dbType":"mysql","dbDriver":"native","heartbeat":"select user()","writeHost":[{"host":"binghe51","url":"192.168.175.151:3306","password":"root","user":"root"}]},{"balance":1,"maxCon":1000,"minCon":10,"name":"binghe152","writeType":0,"switchType":1,"slaveThreshold":100,"dbType":"mysql","dbDriver":"native","heartbeat":"select user()","writeHost":[{"host":"binghe52","url":"192.168.175.152:3306","password":"root","user":"root"}]},{"balance":1,"maxCon":1000,"minCon":10,"name":"binghe153","writeType":0,"switchType":1,"slaveThreshold":100,"dbType":"mysql","dbDriver":"native","heartbeat":"select user()","writeHost":[{"host":"binghe53","url":"192.168.175.153:3306","password":"root","user":"root"}]},{"balance":1,"maxCon":1000,"minCon":10,"name":"binghe154","writeType":0,"switchType":1,"slaveThreshold":100,"dbType":"mysql","dbDriver":"native","heartbeat":"select user()","writeHost":[{"host":"binghe54","url":"192.168.175.154:3306","password":"root","user":"root"}]}]

可以看到，Mycat的配置資訊成功同步到了binghe152伺服器上的Zookeeper中。

  • binghe153伺服器
[root@binghe153 ~]# zkCli.sh 
Connecting to localhost:2181
#####################此處省略N行輸出資訊#####################
[zk: localhost:2181(CONNECTED) 0] get /mycat/mycat-cluster-1/schema/dataHost
[{"balance":1,"maxCon":1000,"minCon":10,"name":"binghe151","writeType":0,"switchType":1,"slaveThreshold":100,"dbType":"mysql","dbDriver":"native","heartbeat":"select user()","writeHost":[{"host":"binghe51","url":"192.168.175.151:3306","password":"root","user":"root"}]},{"balance":1,"maxCon":1000,"minCon":10,"name":"binghe152","writeType":0,"switchType":1,"slaveThreshold":100,"dbType":"mysql","dbDriver":"native","heartbeat":"select user()","writeHost":[{"host":"binghe52","url":"192.168.175.152:3306","password":"root","user":"root"}]},{"balance":1,"maxCon":1000,"minCon":10,"name":"binghe153","writeType":0,"switchType":1,"slaveThreshold":100,"dbType":"mysql","dbDriver":"native","heartbeat":"select user()","writeHost":[{"host":"binghe53","url":"192.168.175.153:3306","password":"root","user":"root"}]},{"balance":1,"maxCon":1000,"minCon":10,"name":"binghe154","writeType":0,"switchType":1,"slaveThreshold":100,"dbType":"mysql","dbDriver":"native","heartbeat":"select user()","writeHost":[{"host":"binghe54","url":"192.168.175.154:3306","password":"root","user":"root"}]}]

可以看到，Mycat的配置資訊成功同步到了binghe153伺服器上的Zookeeper中。

配置Mycat支援Zookeeper啟動

1.在binghe151伺服器上配置Mycat

在binghe151伺服器上進入Mycat安裝目錄的conf目錄下，檢視檔案資訊，如下所示。

[root@binghe151 ~]# cd /usr/local/mycat/conf/
[root@binghe151 conf]# ll
total 108
-rwxrwxrwx 1 root root   92 Feb 26 17:10 autopartition-long.txt
-rwxrwxrwx 1 root root   51 Feb 26 17:10 auto-sharding-long.txt
-rwxrwxrwx 1 root root   67 Feb 26 17:10 auto-sharding-rang-mod.txt
-rwxrwxrwx 1 root root  340 Feb 26 17:10 cacheservice.properties
-rwxrwxrwx 1 root root 3338 Feb 26 17:10 dbseq.sql
-rwxrwxrwx 1 root root 3532 Feb 26 17:10 dbseq - utf8mb4.sql
-rw-r--r-- 1 root root   86 Mar  1 22:37 dnindex.properties
-rwxrwxrwx 1 root root  446 Feb 26 17:10 ehcache.xml
-rwxrwxrwx 1 root root 2454 Feb 26 17:10 index_to_charset.properties
-rwxrwxrwx 1 root root 1285 Feb 26 17:10 log4j2.xml
-rwxrwxrwx 1 root root  183 Feb 26 17:10 migrateTables.properties
-rwxrwxrwx 1 root root  271 Feb 26 17:10 myid.properties
-rwxrwxrwx 1 root root   16 Feb 26 17:10 partition-hash-int.txt
-rwxrwxrwx 1 root root  108 Feb 26 17:10 partition-range-mod.txt
-rwxrwxrwx 1 root root  988 Mar  1 16:59 rule.xml
-rwxrwxrwx 1 root root 3883 Mar  3 23:59 schema.xml
-rwxrwxrwx 1 root root  440 Feb 26 17:10 sequence_conf.properties
-rwxrwxrwx 1 root root   84 Mar  3 23:52 sequence_db_conf.properties
-rwxrwxrwx 1 root root   29 Feb 26 17:10 sequence_distributed_conf.properties
-rwxrwxrwx 1 root root   28 Feb 26 17:10 sequence_http_conf.properties
-rwxrwxrwx 1 root root   53 Feb 26 17:10 sequence_time_conf.properties
-rwxrwxrwx 1 root root 2420 Mar  4 15:14 server.xml
-rwxrwxrwx 1 root root   18 Feb 26 17:10 sharding-by-enum.txt
-rwxrwxrwx 1 root root 4251 Feb 28 20:51 wrapper.conf
drwxrwxrwx 2 root root 4096 Feb 28 21:17 zkconf
drwxrwxrwx 2 root root 4096 Feb 28 21:17 zkdownload

可以看到，在Mycat的conf目錄下，存在一個myid.properties檔案，接下來，使用vim編輯器編輯這個檔案，如下所示。

vim myid.properties

編輯後的myid.properties檔案的內容如下所示。

loadZk=true
zkURL=192.168.175.151:2181,192.168.175.152:2181,192.168.175.153:2181
clusterId=mycat-cluster-1
myid=mycat_151
clusterSize=2
clusterNodes=mycat_151,mycat_154
#server  booster  ;   booster install on db same server,will reset all minCon to 2
type=server
boosterDataHosts=dataHost1

其中幾個重要的引數說明如下所示。

  • loadZk：表示是否載入Zookeeper配置。true：是； false：否；
  • zkURL：Zookeeper的連線地址，多個Zookeeper連線地址以逗號隔開；
  • clusterId：當前Mycat叢集的Id標識，此標識需要與Zookeeper中/mycat目錄下的目錄名稱相同，如下所示。
[zk: localhost:2181(CONNECTED) 1] ls /mycat
[mycat-cluster-1]
  • myid：當前Mycat節點的id，這裡我的命名方式為mycat_字首加上IP地址的最後三位；
  • clusterSize：表示Mycat叢集中的Mycat節點個數，這裡，我們在binghe151和binghe154節點上部署Mycat，所以Mycat節點的個數為2。
  • clusterNodes：Mycat叢集中，所有的Mycat節點，此處的節點需要配置myid中配置的Mycat節點id，多個節點之前以逗號分隔。這裡我配置的節點為：mycat_151,mycat_154。

2.在binghe154伺服器上安裝全新的Mycat

在binghe154伺服器上下載並安裝和binghe151伺服器上相同版本的Mycat，並將其解壓到binghe154伺服器上的/usr/local/mycat目錄下。

也可以在binghe151伺服器上直接輸入如下命令將Mycat的安裝目錄複製到binghe154伺服器上。

[root@binghe151 ~]# scp -r /usr/local/mycat binghe154:/usr/local

注意：別忘了在binghe154伺服器上配置Mycat的系統環境變數。

3.修改binghe154伺服器上的Mycat配置

在binghe154伺服器上修改Mycat安裝目錄下的conf目錄中的myid.properties檔案，如下所示。

vim /usr/local/mycat/conf/myid.properties

修改後的myid.properties檔案的內容如下所示。

loadZk=true
zkURL=192.168.175.151:2181,192.168.175.152:2181,192.168.175.153:2181
clusterId=mycat-cluster-1
myid=mycat_154
clusterSize=2
clusterNodes=mycat_151,mycat_154
#server  booster  ;   booster install on db same server,will reset all minCon to 2
type=server
boosterDataHosts=dataHost1

4.重啟Mycat

分別重啟binghe151伺服器和binghe154伺服器上的Mycat，如下所示。

注意：先重啟

  • binghe151伺服器
[root@binghe151 ~]# mycat restart
Stopping Mycat-server...
Stopped Mycat-server.
Starting Mycat-server...
  • binghe154伺服器
[root@binghe154 ~]# mycat restart
Stopping Mycat-server...
Stopped Mycat-server.
Starting Mycat-server...

在binghe151和binghe154伺服器上分別檢視Mycat的啟動日誌，如下所示。

STATUS | wrapper  | 2020/03/08 21:08:15 | <-- Wrapper Stopped
STATUS | wrapper  | 2020/03/08 21:08:15 | --> Wrapper Started as Daemon
STATUS | wrapper  | 2020/03/08 21:08:15 | Launching a JVM...
INFO   | jvm 1    | 2020/03/08 21:08:16 | Wrapper (Version 3.2.3) http://wrapper.tanukisoftware.org
INFO   | jvm 1    | 2020/03/08 21:08:16 |   Copyright 1999-2006 Tanuki Software, Inc.  All Rights Reserved.
INFO   | jvm 1    | 2020/03/08 21:08:16 | 
INFO   | jvm 1    | 2020/03/08 21:08:28 | MyCAT Server startup successfully. see logs in logs/mycat.log

從日誌的輸出結果可以看出，Mycat重啟成功。

此時，先重啟binghe151伺服器上的Mycat，再重啟binghe154伺服器上的Mycat之後，我們會發現binghe154伺服器上的Mycat的conf目錄下的schema.xml、server.xml、rule.xml和sequence_db_conf.properties檔案與binghe151伺服器上Mycat的配置檔案相同，這就是binghe154伺服器上的Mycat從Zookeeper上讀取配置檔案的結果。

以後，我們只需要修改Zookeeper中有關Mycat的配置，這些配置就會自動同步到Mycat中，這樣可以保證多個Mycat節點的配置是一致的。

配置虛擬IP

分別在binghe151和binghe154伺服器上配置虛擬IP，如下所示。

ifconfig eth0:1 192.168.175.110 broadcast 192.168.175.255 netmask 255.255.255.0 up
route add -host 192.168.175.110 dev eth0:1

配置完虛擬IP的效果如下所示，以binghe151伺服器為例。

[root@binghe151 ~]# ifconfig
eth0      Link encap:Ethernet  HWaddr 00:0C:29:10:A1:45  
          inet addr:192.168.175.151  Bcast:192.168.175.255  Mask:255.255.255.0
          inet6 addr: fe80::20c:29ff:fe10:a145/64 Scope:Link
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1
          RX packets:116766 errors:0 dropped:0 overruns:0 frame:0
          TX packets:85230 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:1000 
          RX bytes:25559422 (24.3 MiB)  TX bytes:55997016 (53.4 MiB)

eth0:1    Link encap:Ethernet  HWaddr 00:0C:29:10:A1:45  
          inet addr:192.168.175.110  Bcast:192.168.175.255  Mask:255.255.255.0
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1

lo        Link encap:Local Loopback  
          inet addr:127.0.0.1  Mask:255.0.0.0
          inet6 addr: ::1/128 Scope:Host
          UP LOOPBACK RUNNING  MTU:65536  Metric:1
          RX packets:51102 errors:0 dropped:0 overruns:0 frame:0
          TX packets:51102 errors:0 dropped:0 overruns:0 carrier:0
          collisions:0 txqueuelen:0 
          RX bytes:2934009 (2.7 MiB)  TX bytes:2934009 (2.7 MiB)

注意：在命令列新增VIP後，當伺服器重啟後，VIP資訊會消失，所以，最好是將建立VIP的命令寫到一個指令碼檔案中，例如，將命令寫到/usr/local/script/vip.sh檔案中，如下所示。

mkdir /usr/local/script
vim /usr/local/script/vip.sh

檔案的內容如下所示。

ifconfig eth0:1 192.168.175.110 broadcast 192.168.175.255 netmask 255.255.255.0 up
route add -host 192.168.175.110 dev eth0:1

接下來，將/usr/local/script/vip.sh檔案新增到伺服器開機啟動項中，如下所示。

echo /usr/local/script/vip.sh >> /etc/rc.d/rc.local

配置IP轉發

在binghe151和binghe154伺服器上配置系統核心IP轉發功能，編輯/etc/sysctl.conf檔案，如下所示。

vim /etc/sysctl.conf

找到如下一行程式碼。

net.ipv4.ip_forward = 0

將其修改成如下所示的程式碼。

net.ipv4.ip_forward = 1

儲存並退出vim編輯器，並執行如下命令使配置生效。

sysctl -p

安裝並配置xinetd服務

我們需要在安裝HAProxy的伺服器上，也就是在binghe151和binghe154伺服器上安裝xinetd服務來開啟48700埠。

（1）在伺服器命令列執行如下命令安裝xinetd服務，如下所示。

yum install xinetd -y

（2）編輯/etc/xinetd.conf檔案，如下所示。

vim /etc/xinetd.conf

檢查檔案中是否存在如下配置。

includedir /etc/xinetd.d

如果/etc/xinetd.conf檔案中沒有以上配置，則在/etc/xinetd.conf檔案中新增以上配置；如果存在以上配置，則不用修改。

（3）建立/etc/xinetd.d目錄，如下所示。

mkdir /etc/xinetd.d

注意：如果/etc/xinetd.d目錄已經存在，建立目錄時會報如下錯誤。

mkdir: cannot create directory `/etc/xinetd.d': File exists

大家可不必理會此錯誤資訊。

（4）在/etc/xinetd.d目錄下新增Mycat狀態檢測伺服器的配置檔案mycat_status，如下所示。

touch /etc/xinetd.d/mycat_status

（5）編輯mycat_status檔案，如下所示。

vim /etc/xinetd.d/mycat_status

編輯後的mycat_status檔案中的內容如下所示。

service mycat_status
{
flags = REUSE
socket_type = stream
port = 48700
wait = no
user = root
server =/usr/local/bin/mycat_check.sh
log_on_failure += USERID
disable = no
}

部分xinetd配置引數說明如下所示。

  • socket_type：表示封包處理方式，Stream為TCP資料包。
  • port：表示xinetd服務監聽的埠號。
  • wait：表示不需等待，即服務將以多執行緒的方式執行。
  • user：執行xinted服務的使用者。
  • server：需要啟動的服務指令碼。
  • log_on_failure：記錄失敗的日誌內容。
  • disable：需要啟動xinted服務時，需要將此配置項設定為no。

（6）在/usr/local/bin目錄下新增mycat_check.sh服務指令碼，如下所示。

touch /usr/local/bin/mycat_check.sh

（7）編輯/usr/local/bin/mycat_check.sh檔案，如下所示。

vim /usr/local/bin/mycat_check.sh

編輯後的檔案內容如下所示。

#!/bin/bash
mycat=`/usr/local/mycat/bin/mycat status | grep 'not running' | wc -l`
if [ "$mycat" = "0" ]; then
/bin/echo -e "HTTP/1.1 200 OK\r\n"
else
/bin/echo -e "HTTP/1.1 503 Service Unavailable\r\n"
/usr/local/mycat/bin/mycat start
fi

為mycat_check.sh檔案賦予可執行許可權，如下所示。

chmod a+x /usr/local/bin/mycat_check.sh

（8）編輯/etc/services檔案，如下所示。

vim /etc/services

在檔案末尾新增如下所示的內容。

mycat_status  48700/tcp        # mycat_status

其中，埠號需要與在/etc/xinetd.d/mycat_status檔案中配置的埠號相同。

（9）重啟xinetd服務，如下所示。

service xinetd restart

（10）檢視mycat_status服務是否成功啟動，如下所示。

  • binghe151伺服器
[root@binghe151 ~]# netstat -antup|grep 48700
tcp    0   0 :::48700          :::*      LISTEN   2776/xinetd
  • binghe154伺服器
[root@binghe154 ~]# netstat -antup|grep 48700
tcp    0   0 :::48700          :::*      LISTEN   6654/xinetd

結果顯示，兩臺伺服器上的mycat_status伺服器啟動成功。

至此，xinetd服務安裝並配置成功，即Mycat狀態檢查服務安裝成功。

安裝並配置HAProxy

我們直接在binghe151和binghe154伺服器上使用如下命令安裝HAProxy。

yum install haproxy -y

安裝完成後，我們需要對HAProxy進行配置，HAProxy的配置檔案目錄為/etc/haproxy，我們檢視這個目錄下的檔案資訊，如下所示。

[root@binghe151 ~]# ll /etc/haproxy/
total 4
-rw-r--r-- 1 root root 3142 Oct 21  2016 haproxy.cfg

發現/etc/haproxy/目錄下存在一個haproxy.cfg檔案。接下來，我們就修改haproxy.cfg檔案，修改後的haproxy.cfg檔案的內容如下所示。

global
    log         127.0.0.1 local2

    chroot      /var/lib/haproxy
    pidfile     /var/run/haproxy.pid
    maxconn     4000
    user        haproxy
    group       haproxy
    daemon
    stats socket /var/lib/haproxy/stats

defaults
    mode                    http
    log                     global
    option                  httplog
    option                  dontlognull
    option http-server-close
    option                  redispatch
    retries                 3
    timeout http-request    10s
    timeout queue           1m
    timeout connect         10s
    timeout client          1m
    timeout server          1m
    timeout http-keep-alive 10s
    timeout check           10s
    maxconn                 3000

listen  admin_status
      bind 0.0.0.0:48800
      stats uri /admin-status
      stats auth  admin:admin
listen    allmycat_service
      bind 0.0.0.0:3366
      mode tcp
      option tcplog
	  option httpchk OPTIONS * HTTP/1.1\r\nHost:\ www
      balance    roundrobin
      server    mycat_151 192.168.175.151:3307 check port 48700 inter 5s rise 2 fall 3
      server    mycat_154 192.168.175.154:3307 check port 48700 inter 5s rise 2 fall 3
listen    allmycat_admin
      bind 0.0.0.0:3377
      mode tcp
      option tcplog
	  option httpchk OPTIONS * HTTP/1.1\r\nHost:\ www
      balance    roundrobin
      server    mycat_151 192.168.175.151:3308 check port 48700 inter 5s rise 2 fall 3
      server    mycat_154 192.168.175.154:3308 check port 48700 inter 5s rise 2 fall 3

接下來，在binghe151伺服器和binghe154伺服器上啟動HAProxy，如下所示。

haproxy -f /etc/haproxy/haproxy.cfg

接下來，我們使用mysql命令連線HAProxy監聽的虛擬IP和埠來連線Mycat，如下所示。

[root@binghe151 ~]# mysql -umycat -pmycat -h192.168.175.110 -P3366 --default-auth=mysql_native_password  
mysql: [Warning] Using a password on the command line interface can be insecure.
Welcome to the MySQL monitor.  Commands end with ; or \g.
Your MySQL connection id is 2
Server version: 5.6.29-mycat-1.6.7.4-release-20200228205020 MyCat Server (OpenCloudDB)

Copyright (c) 2000, 2019, Oracle and/or its affiliates. All rights reserved.

Oracle is a registered trademark of Oracle Corporation and/or its
affiliates. Other names may be trademarks of their respective
owners.

Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.

mysql>

可以看到，連線Mycat成功。

安裝Keepalived

1.安裝並配置Keepalived

直接在binghe151和binghe154伺服器上輸入如下命令安裝Keepalived。

yum install keepalived -y

安裝成功後，會在/etc目錄下生成一個keepalived目錄，接下來，我們在/etc/keepalived目錄下配置keepalived.conf檔案，如下所示。

vim /etc/keepalived/keepalived.conf
  • binghe151伺服器配置
! Configuration Fileforkeepalived
vrrp_script chk_http_port {
  script "/etc/keepalived/check_haproxy.sh"
  interval 2
  weight 2
}
vrrp_instance VI_1 {
  state MASTER 
  interface eth0 
  virtual_router_id 51 
  priority 150 
  advert_int 1 
  authentication {
  auth_type PASS
  auth_pass 1111
}
track_script {
  chk_http_port
}
virtual_ipaddress { 
  192.168.175.110 dev eth0 scope global
}
}
  • binghe154伺服器配置
! Configuration Fileforkeepalived
vrrp_script chk_http_port {
  script "/etc/keepalived/check_haproxy.sh"
  interval 2
  weight 2
}
vrrp_instance VI_1 {
  state SLAVE 
  interface eth0 
  virtual_router_id 51 
  priority 120
  advert_int 1 
  authentication {
  auth_type PASS
  auth_pass 1111
}
track_script {
  chk_http_port
}
virtual_ipaddress { 
  192.168.175.110 dev eth0 scope global
}
}

2.編寫檢測HAProxy的指令碼

接下來，需要分別在binghe151和binghe154伺服器上的/etc/keepalived目錄下建立check_haproxy.sh指令碼，指令碼內容如下所示。

#!/bin/bash
STARTHAPROXY="/usr/sbin/haproxy -f /etc/haproxy/haproxy.cfg"
STOPKEEPALIVED="/etc/init.d/keepalived stop"
#STOPKEEPALIVED="/usr/bin/systemctl stop keepalived"
LOGFILE="/var/log/keepalived-haproxy-state.log"
echo "[check_haproxy status]" >> $LOGFILE
A=`ps -C haproxy --no-header |wc -l`
echo "[check_haproxy status]" >> $LOGFILE
date >> $LOGFILE
if [ $A -eq 0 ];then
   echo $STARTHAPROXY >> $LOGFILE
   $STARTHAPROXY >> $LOGFILE 2>&1
   sleep 5
fi
if [ `ps -C haproxy --no-header |wc -l` -eq 0 ];then
   exit 0
else
   exit 1
fi

使用如下命令為check_haproxy.sh指令碼授予可執行許可權。

chmod a+x /etc/keepalived/check_haproxy.sh

3.啟動Keepalived

配置完成後，我們就可以啟動Keepalived了，分別在binghe151和binghe154伺服器上啟動Keepalived，如下所示。

/etc/init.d/keepalived start

檢視Keepalived是否啟動成功，如下所示。

  • binghe151伺服器
[root@binghe151 ~]# ps -ef | grep keepalived
root       1221      1  0 20:06 ?        00:00:00 keepalived -D
root       1222   1221  0 20:06 ?        00:00:00 keepalived -D
root       1223   1221  0 20:06 ?        00:00:02 keepalived -D
root      93290   3787  0 21:42 pts/0    00:00:00 grep keepalived
  • binghe154伺服器
[root@binghe154 ~]# ps -ef | grep keepalived
root       1224      1  0 20:06 ?        00:00:00 keepalived -D
root       1225   1224  0 20:06 ?        00:00:00 keepalived -D
root       1226   1224  0 20:06 ?        00:00:02 keepalived -D
root      94636   3798  0 21:43 pts/0    00:00:00 grep keepalived

可以看到，兩臺伺服器上的Keepalived服務啟動成功。

4.驗證Keepalived繫結的虛擬IP

接下來，我們分別檢視兩臺伺服器上的Keepalived是否繫結了虛擬IP。

  • binghe151伺服器
[root@binghe151 ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:10:a1:45 brd ff:ff:ff:ff:ff:ff
    inet 192.168.175.151/24 brd 192.168.175.255 scope global eth0
    inet 192.168.175.110/32 scope global eth0
    inet 192.168.175.110/24 brd 192.168.175.255 scope global secondary eth0:1
    inet6 fe80::20c:29ff:fe10:a145/64 scope link 
       valid_lft forever preferred_lft forever

可以看到如下一行程式碼。

inet 192.168.175.110/32 scope global eth0

說明binghe151伺服器上的Keepalived繫結了虛擬IP 192.168.175.110。

  • binghe154伺服器
[root@binghe154 ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:50:56:22:2a:75 brd ff:ff:ff:ff:ff:ff
    inet 192.168.175.154/24 brd 192.168.175.255 scope global eth0
    inet 192.168.175.110/24 brd 192.168.175.255 scope global secondary eth0:1
    inet6 fe80::250:56ff:fe22:2a75/64 scope link 
       valid_lft forever preferred_lft forever

可以看到binghe154伺服器上的Keepalived並沒有繫結虛擬IP。

5.測試虛擬IP的漂移

如何測試虛擬IP的漂移呢？首先，我們停止binghe151伺服器上的Keepalived，如下所示。

/etc/init.d/keepalived stop

接下來，檢視binghe154伺服器上Keepalived繫結虛擬IP的情況，如下所示。

[root@binghe154 ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:50:56:22:2a:75 brd ff:ff:ff:ff:ff:ff
    inet 192.168.175.154/24 brd 192.168.175.255 scope global eth0
    inet 192.168.175.110/32 scope global eth0
    inet 192.168.175.110/24 brd 192.168.175.255 scope global secondary eth0:1
    inet6 fe80::250:56ff:fe22:2a75/64 scope link 
       valid_lft forever preferred_lft forever

可以看到，在輸出的結果資訊中，存在如下一行資訊。

 inet 192.168.175.110/32 scope global eth0

說明binghe154伺服器上的Keepalived繫結了虛擬IP 192.168.175.110，虛擬IP漂移到了binghe154伺服器上。

6.binghe151伺服器上的Keepalived搶佔虛擬IP

接下來，我們啟動binghe151伺服器上的Keepalived，如下所示。

/etc/init.d/keepalived start

啟動成功後，我們再次檢視虛擬IP的繫結情況，如下所示。

  • binghe151伺服器
[root@binghe151 ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:0c:29:10:a1:45 brd ff:ff:ff:ff:ff:ff
    inet 192.168.175.151/24 brd 192.168.175.255 scope global eth0
    inet 192.168.175.110/32 scope global eth0
    inet 192.168.175.110/24 brd 192.168.175.255 scope global secondary eth0:1
    inet6 fe80::20c:29ff:fe10:a145/64 scope link 
       valid_lft forever preferred_lft forever
  • binghe154伺服器
[root@binghe154 ~]# ip addr
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN 
    link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
    inet 127.0.0.1/8 scope host lo
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever
2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000
    link/ether 00:50:56:22:2a:75 brd ff:ff:ff:ff:ff:ff
    inet 192.168.175.154/24 brd 192.168.175.255 scope global eth0
    inet 192.168.175.110/24 brd 192.168.175.255 scope global secondary eth0:1
    inet6 fe80::250:56ff:fe22:2a75/64 scope link 
       valid_lft forever preferred_lft forever

由於binghe151伺服器上配置的Keepalived優先順序要高於binghe154伺服器上的Keepalived，所以，再次啟動binghe151伺服器上的Keepalived後，binghe151伺服器上的Keepalived會搶佔虛擬IP。

配置MySQL主從複製

這裡，為了簡單，我將binghe154和binghe155伺服器上的MySQL配置成主從複製，大家也可以根據實際情況，自行配置其他伺服器上MySQL的主從複製（注意：我這裡配置的是一主一從模式）。

1.編輯my.cnf檔案

  • binghe154伺服器
server_id = 154
log_bin = /data/mysql/log/bin_log/mysql-bin
binlog-ignore-db=mysql
binlog_format= mixed
sync_binlog=100
log_slave_updates = 1
binlog_cache_size=32m
max_binlog_cache_size=64m
max_binlog_size=512m
lower_case_table_names = 1
relay_log = /data/mysql/log/bin_log/relay-bin
relay_log_index = /data/mysql/log/bin_log/relay-bin.index
master_info_repository=TABLE
relay-log-info-repository=TABLE
relay-log-recovery
  • binghe155伺服器
server_id = 155
log_bin = /data/mysql/log/bin_log/mysql-bin
binlog-ignore-db=mysql
binlog_format= mixed
sync_binlog=100
log_slave_updates = 1
binlog_cache_size=32m
max_binlog_cache_size=64m
max_binlog_size=512m
lower_case_table_names = 1
relay_log = /data/mysql/log/bin_log/relay-bin
relay_log_index = /data/mysql/log/bin_log/relay-bin.index
master_info_repository=TABLE
relay-log-info-repository=TABLE
relay-log-recovery

2.同步兩臺伺服器上MySQL的資料

在binghe154伺服器上只有一個customer_db資料庫，我們使用mysqldump命令匯出customer_db資料庫，如下所示。

[root@binghe154 ~]# mysqldump --master-data=2 --single-transaction -uroot -p --databases customer_db > binghe154.sql
Enter password:

接下來，我們檢視binghe154.sql檔案。

more binghe154.sql

在檔案中，我們可以找到如下資訊。

CHANGE MASTER TO MASTER_LOG_FILE='mysql-bin.000042', MASTER_LOG_POS=995;

說明當前MySQL的二進位制日誌檔案為mysql-bin.000042，二進位制日誌檔案的位置為995。

接下來，我們將binghe154.sql檔案複製到binghe155伺服器上，如下所示。

scp binghe154.sql 192.168.175.155:/usr/local/src

在binghe155伺服器上，將binghe154.sql指令碼匯入到MySQL中，如下所示。

mysql -uroot -p < /usr/local/src/binghe154.sql

此時，完成了資料的初始化。

3.建立主從複製賬號

在binghe154伺服器的MySQL中，建立用於主從複製的MySQL賬號，如下所示。

mysql> CREATE USER 'repl'@'192.168.175.%' IDENTIFIED BY 'repl123456';
Query OK, 0 rows affected (0.01 sec)

mysql> ALTER USER 'repl'@'192.168.175.%' IDENTIFIED WITH mysql_native_password BY 'repl123456';                           
Query OK, 0 rows affected (0.00 sec)

mysql> GRANT REPLICATION SLAVE ON *.* TO 'repl'@'192.168.175.%';
Query OK, 0 rows affected (0.00 sec)

mysql> FLUSH PRIVILEGES;
Query OK, 0 rows affected (0.00 sec)

4.配置複製鏈路

登入binghe155伺服器上的MySQL，並使用如下命令配置複製鏈路。

mysql> change master to 
     > master_host='192.168.175.154',
     > master_port=3306,
     > master_user='repl',
     > master_password='repl123456',
     > MASTER_LOG_FILE='mysql-bin.000042',
     > MASTER_LOG_POS=995;

其中，MASTER_LOG_FILE='mysql-bin.000042', MASTER_LOG_POS=995 就是在binghe154.sql檔案中找到的。

5.啟動從庫

在binghe155伺服器的MySQL命令列啟動從庫，如下所示。

mysql> start slave;

檢視從庫是否啟動成功，如下所示。

mysql> SHOW slave STATUS \G
*************************** 1. row ***************************
               Slave_IO_State: Waiting for master to send event
                  Master_Host: 192.168.175.151
                  Master_User: binghe152
                  Master_Port: 3306
                Connect_Retry: 60
              Master_Log_File: mysql-bin.000007
          Read_Master_Log_Pos: 1360
               Relay_Log_File: relay-bin.000003
                Relay_Log_Pos: 322
        Relay_Master_Log_File: mysql-bin.000007
             Slave_IO_Running: Yes
            Slave_SQL_Running: Yes
#################省略部分輸出結果資訊##################

結果顯示Slave_IO_Running選項和Slave_SQL_Running選項的值均為Yes，說明MySQL主從複製環境搭建成功。

最後，別忘了在binghe155伺服器的MySQL中建立Mycat連線MySQL的使用者，如下所示。

CREATE USER 'mycat'@'192.168.175.%' IDENTIFIED BY 'mycat';
ALTER USER 'mycat'@'192.168.175.%' IDENTIFIED WITH mysql_native_password BY 'mycat'; 
GRANT SELECT, INSERT, UPDATE, DELETE,EXECUTE  ON *.* TO 'mycat'@'192.168.175.%';
FLUSH PRIVILEGES;

配置Mycat讀寫分離

修改Mycatd的schema.xml檔案，實現binghe154和binghe155伺服器上的MySQL讀寫分離。在Mycat安裝目錄的conf/zkconf目錄下，修改schema.xml檔案，修改後的schema.xml檔案如下所示。

<!DOCTYPE mycat:schema SYSTEM "schema.dtd">
<mycat:schema xmlns:mycat="http://io.mycat/">
    <schema name="shop" checkSQLschema="true" sqlMaxLimit="1000">
        <table name="order_master" dataNode="orderdb01,orderdb02,orderdb03,orderdb04" rule="order_master" primaryKey="order_id" autoIncrement="true">
            <childTable name="order_detail" joinKey="order_id" parentKey="order_id" primaryKey="order_detail_id" autoIncrement="true"/>
        </table>
        <table name="order_cart" dataNode="ordb" primaryKey="cart_id"/>
        <table name="order_customer_addr" dataNode="ordb" primaryKey="customer_addr_id"/>
        <table name="region_info" dataNode="ordb,prodb,custdb" primaryKey="region_id" type="global"/>
        <table name="serial" dataNode="ordb" primaryKey="id"/>
        <table name="shipping_info" dataNode="ordb" primaryKey="ship_id"/>
        <table name="warehouse_info" dataNode="ordb" primaryKey="w_id"/>
        <table name="warehouse_proudct" dataNode="ordb" primaryKey="wp_id"/>
        <table name="product_brand_info" dataNode="prodb" primaryKey="brand_id"/>
        <table name="product_category" dataNode="prodb" primaryKey="category_id"/>
        <table name="product_comment" dataNode="prodb" primaryKey="comment_id"/>
        <table name="product_info" dataNode="prodb" primaryKey="product_id"/>
        <table name="product_pic_info" dataNode="prodb" primaryKey="product_pic_id"/>
        <table name="product_supplier_info" dataNode="prodb" primaryKey="supplier_id"/>
        <table name="customer_balance_log" dataNode="custdb" primaryKey="balance_id"/>
        <table name="customer_inf" dataNode="custdb" primaryKey="customer_inf_id"/>
        <table name="customer_level_inf" dataNode="custdb" primaryKey="customer_level"/>
        <table name="customer_login" dataNode="custdb" primaryKey="customer_id"/>
        <table name="customer_login_log" dataNode="custdb" primaryKey="login_id"/>
        <table name="customer_point_log" dataNode="custdb" primaryKey="point_id"/>
    </schema>
	
    <dataNode name="mycat" dataHost="binghe151" database="mycat"/>
    <dataNode name="ordb" dataHost="binghe152" database="order_db"/>
    <dataNode name="prodb" dataHost="binghe153" database="product_db"/>
    <dataNode name="custdb" dataHost="binghe154" database="customer_db"/>
    <dataNode name="orderdb01" dataHost="binghe152" database="orderdb01"/>
    <dataNode name="orderdb02" dataHost="binghe152" database="orderdb02"/>
    <dataNode name="orderdb03" dataHost="binghe153" database="orderdb03"/>
    <dataNode name="orderdb04" dataHost="binghe153" database="orderdb04"/>
	
    <dataHost balance="1" maxCon="1000" minCon="10" name="binghe151" writeType="0" switchType="1" slaveThreshold="100" dbType="mysql" dbDriver="native">
        <heartbeat>select user()</heartbeat>
        <writeHost host="binghe51" url="192.168.175.151:3306" password="mycat" user="mycat"/>
    </dataHost>
    <dataHost balance="1" maxCon="1000" minCon="10" name="binghe152" writeType="0" switchType="1" slaveThreshold="100" dbType="mysql" dbDriver="native">
        <heartbeat>select user()</heartbeat>
        <writeHost host="binghe52" url="192.168.175.152:3306" password="mycat" user="mycat"/>
    </dataHost>
    <dataHost balance="1" maxCon="1000" minCon="10" name="binghe153" writeType="0" switchType="1" slaveThreshold="100" dbType="mysql" dbDriver="native">
        <heartbeat>select user()</heartbeat>
        <writeHost host="binghe53" url="192.168.175.153:3306" password="mycat" user="mycat"/>
    </dataHost>
    <dataHost balance="1" maxCon="1000" minCon="10" name="binghe154" writeType="0" switchTymycate="1" slaveThreshold="100" dbType="mysql" dbDriver="native">
        <heartbeat>select user()</heartbeat>
        <writeHost host="binghe54" url="192.168.175.154:3306" password="mycat" user="mycat">
			<readHost host="binghe55", url="192.168.175.155:3306" user="mycat" password="mycat"/>
		</writeHost>
         <writeHost host="binghe55" url="192.168.175.155:3306" password="mycat" user="mycat"/>
    </dataHost>
</mycat:schema>

儲存並退出vim編輯器，接下來，初始化Zookeeper中的資料，如下所示。

/usr/local/mycat/bin/init_zk_data.sh

上述命令執行成功後，會自動將配置同步到binghe151和binghe154伺服器上的Mycat的安裝目錄下的conf目錄下的schema.xml中。

接下來，分別啟動binghe151和binghe154伺服器上的Mycat服務。

mycat restart

如何訪問高可用環境

此時，整個高可用環境配置完成，上層應用連線高可用環境時，需要連線HAProxy監聽的IP和埠。比如使用mysql命令連線高可用環境如下所示。

[root@binghe151 ~]# mysql -umycat -pmycat -h192.168.175.110 -P3366 --default-auth=mysql_native_password
mysql: [Warning] Using a password on the command line interface can be insecure.
Welcome to the MySQL monitor.  Commands end with ; or \g.
Your MySQL connection id is 2
Server version: 5.6.29-mycat-1.6.7.4-release-20200228205020 MyCat Server (OpenCloudDB)

Copyright (c) 2000, 2019, Oracle and/or its affiliates. All rights reserved.

Oracle is a registered trademark of Oracle Corporation and/or its
affiliates. Other names may be trademarks of their respective
owners.

Type 'help;' or '\h' for help. Type '\c' to clear the current input statement.

mysql> show databases;
+----------+
| DATABASE |
+----------+
| shop     |
+----------+
1 row in set (0.10 sec)

mysql> use shop;
Database changed
mysql> show tables;
+-----------------------+
| Tables in shop        |
+-----------------------+
| customer_balance_log  |
| customer_inf          |
| customer_level_inf    |
| customer_login        |
| customer_login_log    |
| customer_point_log    |
| order_cart            |
| order_customer_addr   |
| order_detail          |
| order_master          |
| product_brand_info    |
| product_category      |
| product_comment       |
| product_info          |
| product_pic_info      |
| product_supplier_info |
| region_info           |
| serial                |
| shipping_info         |
| warehouse_info        |
| warehouse_proudct     |
+-----------------------+
21 rows in set (0.00 sec)

這裡，我只是對binghe154伺服器上的MySQL擴充套件了讀寫分離環境，大家也可以根據實際情況對其他伺服器的MySQL實現主從複製和讀寫分離，這樣，整個高可用環境就實現了HAProxy的高可用、Mycat的高可用、MySQL的高可用、Zookeeper的高可用和Keepalived的高可用。

好了，今天就到這兒吧，我是冰河，大家有啥問題可以在下方留言，也可以加我微信，一起交流技術，一起進階，一起牛逼~~

相關文章