mysql+keepalived 双主热备高能用
发布时间：2022-04-04 10:47:12 所属栏目：MySql教程 来源：互联网
导读：理论介绍： 我们通常说的双机热备是指两台机器都在运行，但并不是两台机器都同时在提供服务。当提供服务的一台出现故障的时候，另外一台会马上自动接管并且提供服务，而且切换的时间非常短。MySQL双主复制，即互为Master-Slave(只有一个Master提供写操作)
|
理论介绍： 我们通常说的双机热备是指两台机器都在运行，但并不是两台机器都同时在提供服务。当提供服务的一台出现故障的时候，另外一台会马上自动接管并且提供服务，而且切换的时间非常短。MySQL双主复制，即互为Master-Slave(只有一个Master提供写操作)，可以实现数据库服务器的热备，但是一个Master宕机后不能实现动态切换。使用Keepalived，可以通过虚拟IP，实现双主对外的统一接口以及自动检查、失败切换机制，从而实现MySQL数据库的高可用方案。 环境说明： 版本 version 5.6.25-log 主1库ip: 10.219.24.25 主2库ip：10.219.24.22 vip: 10.219.24.30 os 版本： centos 6.7 已安装热备软件：xtrabackup 防火墙已关 补充： 主从复制原理： http://blog.csdn.NET/zhang123456456/article/details/72972701 MySQL 5.6安装 ：http://blog.csdn.net/zhang123456456/article/details/53608554 xtrabackup 安装： http://blog.csdn.Net/zhang123456456/article/details/72836184 全程：先搭建一主一从，然后反过来搭建，就是双主复制，最后配置 Mysql+Keepalived 故障转移的高可用环境 一、 先一主一从搭建 ，再反向搭建，配置双主复制 1、 主库参数调整 -- 停止主库mysql [root@mysql02 ~]# service mysql stop [root@mysql02 ~]# netstat -nltp|grep mysql|grep 3606 -- 主库创建relay log目录 [root@mysql02 full]# mkdir -p /data/mysql/relaylog/ [root@mysql02 full]# chown -R mysql:mysql /data/mysql/relaylog -- 调整 my.cnf 参数 [root@mysql02 ~]# cat /etc/my.cnf [client] password = Oracle port = 3306 socket = /data/mysql/mysql.sock [mysqld] server-id=25 port = 3306 socket = /data/mysql/mysql.sock character_set_server = utf8 character_set_client = utf8 collation-server=utf8_general_ci lower_case_table_names = 1 max_connections = 1000 datadir = /data/mysql log_bin = /data/mysql/binarylog/binlog log_bin_index = /data/mysql/binarylog/binlog relay-log = /data/mysql/relaylog/relay relay-log-index = /data/mysql/relaylog/relay relay_log_purge = on binlog_format = mixed innodb_data_file_path = ibdata1:12M:autoextend auto_increment_increment = 10 auto_increment_offset = 1 [mysql] default-character-set = utf8 说明： a、主库必须配置的参数 server-id （主从的server-id必须不同）、log_bin、binlog_format、relay-log、relay-log-index、relay_log_purge auto-increment-offset、auto-increment-increment b、log-slave-updates 意思是，中继日志执行之后，这些变化是否需要计入自己的binarylog。 当你的B服务器需要作为另外一个服务器的主服务器的时候需要打开。 就是双主互相备份，或者多主循环备份。 我们这里需要， 所以打开。 c、auto-increment-offset、auto-increment-increment 两个参数用于在 双主（多主循环）互相备份。 因为每台数据库服务器都可能在同一个表中插入数据，如果表有一个自动增长的主键，那么就会在多服务器上出现主键冲突。 解决这个问题的办法就是让每个数据库的自增主键不连续。 上图说是， 我假设需要将来可能需要10台服务器做备份， 所以auto-increment-increment 设为10. 而 auto-increment-offset=1 表示这台服务器的序号。 从1开始， 不超过auto-increment-increment。这样做之后， 我在这台服务器上插入的第一个id就是 1， 第二行的id就是 11了， 而不是2.（同理，在第二台服务器上插入的第一个id就是2， 第二行就是12， 这个后面再介绍） 这样就不会出现主键冲突了。 后面我们会演示这个id的效果。 -- 启动主库 [root@mysql02 ~]# mysqld_safe --defaults-file=/etc/my.cnf & 2、 从库参数调整 -- 停止从库mysql [root@mysql01 ~]# service mysql stop [root@mysql01 ~]# netstat -nltp|grep mysql|grep 3606 -- 调整 my.cnf 参数 [root@mysql01 ~]# cat /etc/my.cnf [client] password = oracle port = 3306 socket = /data/mysql/mysql.sock [mysqld] server-id=22 port = 3306 socket = /data/mysql/mysql.sock character_set_server = utf8 character_set_client = utf8 collation-server=utf8_general_ci lower_case_table_names = 1 max_connections = 1000 datadir = /data/mysql log_bin = /data/mysql/binarylog/binlog log_bin_index = /data/mysql/binarylog/binlog relay-log = /data/mysql/relaylog/relay relay-log-index = /data/mysql/relaylog/relay relay_log_purge = on auto_increment_increment = 10 auto_increment_offset = 2 [mysql] default-character-set = utf8 说明：从库必须配置的参数 server-id、log_bin、relay-log、relay_log_purge、auto-increment-offset、auto-increment-increment、 3、 主库备份 -- 主库备份目录 [root@mysql02 full]# pwd /xtrabackup/full -- 主库 innobackupex 备份 [root@mysql02 ~]# innobackupex --user=root --password=oracle --port=3606 /xtrabackup/full/ 170610 17:50:23 Backup created in directory '/xtrabackup/full/2017-06-10_17-50-19/' MySQL binlog position: filename 'binlog.000010', position '120' .... 170610 17:50:23 completed OK! -- 查看备份 binlog编号 与 截止 position [root@mysql02 2017-06-10_17-50-19]# cat xtrabackup_binlog_info binlog.000010 120 4、 从库创建与主库相同的备份目录 [root@mysql01 ~]# mkdir -p /xtrabackup/full [root@mysql01 ~]# chown -R mysql:mysql /xtrabackup/full/ 5、 主库将备份 scp 到从库 [root@mysql02 full]# pwd /xtrabackup/full [root@mysql02 full]# scp -r 2017-06-10_17-50-19 10.219.24.22:/xtrabackup/full 6、 从库查看scp过来的备份 [root@mysql01 ~]# cd /xtrabackup/full/2017-06-10_17-50-19/ [root@mysql01 2017-06-10_17-50-19]# ll total 12320 -rw-r-----. 1 root root 419 Jun 10 18:01 backup-my.cnf -rw-r-----. 1 root root 12582912 Jun 10 18:01 ibdata1 drwxr-x---. 2 root root 4096 Jun 10 18:01 mysql drwxr-x---. 2 root root 4096 Jun 10 18:01 performance_schema drwxr-x---. 2 root root 4096 Jun 10 18:01 test -rw-r-----. 1 root root 18 Jun 10 18:01 xtrabackup_binlog_info -rw-r-----. 1 root root 113 Jun 10 18:01 xtrabackup_checkpoints -rw-r-----. 1 root root 482 Jun 10 18:01 xtrabackup_info -rw-r-----. 1 root root 2560 Jun 10 18:01 xtrabackup_logfile 7、 主库创建同步用户 mysql> GRANT replication slave ON *.* TO 'slave25'@'%' IDENTIFIED BY 'oracle'; Query OK, 0 rows affected (0.05 sec) 8、 从库恢复主库数据 -- 从库将原有datadir文件夹重命名到新位置，并创建原文件夹 [root@mysql01 ~]# mv /data/mysql /data/mysqlbak [root@mysql01 ~]# mkdir -p /data/mysql -- innobackupex apply-log [root@mysql01 ~]# innobackupex --apply-log --user=oracle --password=oracle --port=3606 /xtrabackup/full/2017-06-10_17-50-19/ -- innobackupex copy 恢复的文件到原来的数据位置 [root@mysql01 mysql]# innobackupex --defaults-file=/etc/my.cnf --user=root --copy-back /xtrabackup/full/2017-06-10_17-50-19/ 170610 18:25:11 completed OK! -- 创建binlog目录与 relaylog 目录并赋权 [root@mysql01 ~]# mkdir -p /data/mysql/binarylog [root@mysql01 ~]# mkdir -p /data/mysql/relaylog/ [root@mysql01 mysql]# chown -R mysql:mysql /data/mysql 9、 从库配置与检测 -- 从库启动 [root@mysql01 mysql]# mysqld_safe --defaults-file=/etc/my.cnf & -- 从库指定与主库同步的基本信息 mysql> change master to master_host='10.219.24.25', master_port=3306, master_user='slave25', master_password='oracle', master_log_file='binlog.000010', master_log_pos=120; Query OK, 0 rows affected, 2 warnings (0.04 sec) 参数解释： MASTER_HOST : 设置要连接的主服务器的ip地址 MASTER_USER : 设置要连接的主服务器的用户名 MASTER_PASSWORD : 设置要连接的主服务器的密码 MASTER_LOG_FILE : 设置要连接的主服务器的bin日志的日志名称 MASTER_LOG_POS : 设置要连接的主服务器的bin日志的记录位置 -- 启动slave 状态(开始监听msater的变化) mysql> start slave; Query OK, 0 rows affected, 1 warning (0.00 sec) -- 查看slave的状态. mysql> show slave status G; *************************** 1. row *************************** Slave_IO_State: Waiting for master to send event Master_Host: 10.219.24.25 #主库 IP Master_User: slave25 # 主库复制的用户 Master_Port: 3306 # 主库 mysqld Connect_Retry: 60 Master_Log_File: binlog.000010 #io_thread 读取主库 master_log_file Read_Master_Log_Pos: 717 # io_thread 读取主库 master_log_pos Relay_Log_File: relay.000002 Relay_Log_Pos: 877 Relay_Master_Log_File: binlog.000010 #sql_thread 执行主库的 master_log_file Slave_IO_Running: Yes # 关键的，io_thread 是否 running Slave_SQL_Running: Yes # 关键的，sql_thread 是否 running Replicate_Do_DB: Replicate_Ignore_DB: Replicate_Do_Table: Replicate_Ignore_Table: Replicate_Wild_Do_Table: Replicate_Wild_Ignore_Table: Last_Errno: 0 Last_Error: Skip_Counter: 0 Exec_Master_Log_Pos: 717 #sql_thread 执行主库的 master_log_pos Relay_Log_Space: 1040 Until_Condition: None Until_Log_File: Until_Log_Pos: 0 Master_SSL_Allowed: No Master_SSL_CA_File: Master_SSL_CA_Path: Master_SSL_Cert: Master_SSL_Cipher: Master_SSL_Key: Seconds_Behind_Master: 0 # 从库 的延迟 Master_SSL_Verify_Server_Cert: No Last_IO_Errno: 0 Last_IO_Error: Last_SQL_Errno: 0 Last_SQL_Error: Replicate_Ignore_Server_Ids: Master_Server_Id: 25 Master_UUID: 29d68531-4cf9-11e7-8e1f-000c297c4100 Master_Info_File: /data/mysql/master.info SQL_Delay: 0 SQL_Remaining_Delay: NULL Slave_SQL_Running_State: Slave has read all relay log; waiting for the slave I/O thread to update it Master_Retry_Count: 86400 Master_Bind: Last_IO_Error_Timestamp: Last_SQL_Error_Timestamp: Master_SSL_Crl: Master_SSL_Crlpath: Retrieved_Gtid_Set: Executed_Gtid_Set: Auto_Position: 0 1 row in set (0.00 sec) ERROR: No query specified 10、 主从同步检查 -- 主库 mysql> create database repl; Query OK, 1 row affected (0.00 sec) mysql> use repl Database changed mysql> create table repl (id int); Query OK, 0 rows affected (0.02 sec) mysql> insert into repl values(1); Query OK, 1 row affected (0.00 sec) -- 从库 mysql> show databases; +--------------------+ | Database | +--------------------+ | information_schema | | binarylog | | mysql | | performance_schema | | relaylog | | repl | | test | +--------------------+ 7 rows in set (0.00 sec) mysql> use repl Reading table information for completion of table and column names You can turn off this feature to get a quicker startup with -A Database changed mysql> select * from repl; +------+ | id | +------+ | 1 | +------+ 1 row in set (0.00 sec) >一主一从同步成功！ ################################################## #至此A到B的复制已经配置完成，下面配置从B到A的复制。# ################################################## 声明> 下面操作中 新主库即为原从库（10.219.24.22） 新从库为原主库（10.219.24.25） 11、 新主库创建同步用户 mysql> GRANT replication slave ON *.* TO 'slave22'@'%' IDENTIFIED BY 'oracle'; Query OK, 0 rows affected (0.00 sec) 12、 新主库查看 binlog 文件号与 position 点 mysql> show master status; +---------------+----------+--------------+------------------+-------------------+ | File | Position | Binlog_Do_DB | Binlog_Ignore_DB | Executed_Gtid_Set | +---------------+----------+--------------+------------------+-------------------+ | binlog.000004 | 313 | | | | +---------------+----------+--------------+------------------+-------------------+ 1 row in set (0.00 sec) 13、 新从库指定与主库同步的基本信息 mysql> change master to master_host='10.219.24.22', master_port=3306, master_user='slave22', master_password='oracle', master_log_file='binlog.000004', master_log_pos=313; Query OK, 0 rows affected, 2 warnings (0.04 sec) 14、新从库打开 slave 复制功能 mysql> start slave; Query OK, 0 rows affected (0.00 sec) 15、 新从库检测同步复制状态 mysql> show slave status G; *************************** 1. row *************************** Slave_IO_State: Waiting for master to send event Master_Host: 10.219.24.22 Master_User: slave22 Master_Port: 3306 Connect_Retry: 60 Master_Log_File: binlog.000004 Read_Master_Log_Pos: 313 Relay_Log_File: relay.000002 Relay_Log_Pos: 280 Relay_Master_Log_File: binlog.000004 Slave_IO_Running: Yes Slave_SQL_Running: Yes Replicate_Do_DB: Replicate_Ignore_DB: Replicate_Do_Table: Replicate_Ignore_Table: Replicate_Wild_Do_Table: Replicate_Wild_Ignore_Table: Last_Errno: 0 Last_Error: Skip_Counter: 0 Exec_Master_Log_Pos: 313 Relay_Log_Space: 443 Until_Condition: None Until_Log_File: Until_Log_Pos: 0 Master_SSL_Allowed: No Master_SSL_CA_File: Master_SSL_CA_Path: Master_SSL_Cert: Master_SSL_Cipher: Master_SSL_Key: Seconds_Behind_Master: 0 Master_SSL_Verify_Server_Cert: No Last_IO_Errno: 0 Last_IO_Error: Last_SQL_Errno: 0 Last_SQL_Error: Replicate_Ignore_Server_Ids: Master_Server_Id: 22 Master_UUID: 70023652-4dc7-11e7-9360-000c2944297a Master_Info_File: /data/mysql/master.info SQL_Delay: 0 SQL_Remaining_Delay: NULL Slave_SQL_Running_State: Slave has read all relay log; waiting for the slave I/O thread to update it Master_Retry_Count: 86400 Master_Bind: Last_IO_Error_Timestamp: Last_SQL_Error_Timestamp: Master_SSL_Crl: Master_SSL_Crlpath: Retrieved_Gtid_Set: Executed_Gtid_Set: Auto_Position: 0 1 row in set (0.00 sec) ERROR: No query specified -- 新从库测试数据同步状态 mysql> create database mm_repl; Query OK, 1 row affected (0.00 sec) mysql> use mm_repl; Database changed mysql> create table mm_repl(id int auto_increment,name varchar(10), primary key(id)); Query OK, 0 rows affected (0.01 sec) mysql> insert into mm_repl(name) values("andy"),("taoYe"); Query OK, 1 row affected (0.00 sec) mysql> select * from mm_repl; +----+-------+ | id | name | +----+-------+ | 1 | andy | | 11 | taoYe | +----+-------+ 2 rows in set (0.00 sec) -- 新主库测试数据同步状态 mysql> select * from mm_repl; +----+-------+ | id | name | +----+-------+ | 1 | andy | | 11 | taoYe | +----+-------+ 2 rows in set (0.00 sec) mysql> insert into mm_repl(name) values("andy"),("taoYe"); Query OK, 2 rows affected (0.00 sec) Records: 2 Duplicates: 0 Warnings: 0 mysql> select * from mm_repl; +----+-------+ | id | name | +----+-------+ | 1 | andy | | 11 | taoYe | | 12 | andy | | 22 | taoYe | +----+-------+ 4 rows in set (0.00 sec) -- 新从库检查同步复制 mysql> select * from mm_repl; +----+-------+ | id | name | +----+-------+ | 1 | andy | | 11 | taoYe | | 12 | andy | | 22 | taoYe | +----+-------+ 4 rows in set (0.00 sec) >主主同步测试成功 二、 配置Mysql+Keepalived故障转移的高可用环境 安装keepalived （ master1 和 master2 两台机器上都安装） 二、1.1 安装 LVS 前提：已经提前配置好本地 Yum 源 配置过程可参考> http://blog.csdn.NET/zhang123456456/article/details/56690945 1、 检查一下系统内核 安装之前，首先检查一下系统内核是否支持LVS的IPVS模块；自Linux 2.6开始，系统内核完全内置了LVS的各个模块。 [root@mysql02 ~]# modprobe -l|grep ipvs kernel/net/netfilter/ipvs/ip_vs.ko kernel/net/netfilter/ipvs/ip_vs_rr.ko kernel/net/netfilter/ipvs/ip_vs_wrr.ko kernel/net/netfilter/ipvs/ip_vs_lc.ko .... kernel/net/netfilter/ipvs/ip_vs_pe_sip.ko 看到如上输出信息，则表明系统内核默认支持IPVS模块。 2、安装依赖包 由于是源码包安装，所以编译时需要打上如下依赖包。 # yum -y install kernel-headers glibc-headers glibc-devel libgomp gcc libstdc++-devel gcc-c++ 3、 yum安装 ipvsadm-1.26-4 [root@mysql02 ~]# yum -y install kernel-devel ipvsadm =================================================================================== Package Arch Version Repository Size =================================================================================== Installing: ipvsadm x86_64 1.26-4.el6 c6-media 42 k Transaction Summary =================================================================================== Install 1 Package(s) Total download size: 42 k Installed size: 78 k Downloading Packages: Running rpm_check_debug Running Transaction Test Transaction Test Succeeded Running Transaction Installing : ipvsadm-1.26-4.el6.x86_64 1/1 Verifying : ipvsadm-1.26-4.el6.x86_64 1/1 Installed: ipvsadm.x86_64 0:1.26-4.el6 Complete! 4、 做个 ln [root@mysql02 ~]# ln -sv /usr/src/kernels/2.6.32-573.el6.x86_64/ /usr/src/linux <--请自行修改 kernels 版本号 5、 查看 ipvsadm 版本信息 [root@mysql02 ~]# ipvsadm -version ipvsadm v1.26 2008/5/15 (compiled with popt and IPVS v1.2.1) #> 能看到帮助信息，则说明IPVS安装成功。 二、1.2 安装 keepalived-1.2.24 1、官网下载地址： [root@mysql02 ~]# wget http://www.keepalived.org/software/keepalived-1.2.24.tar.gz 2、解压安装包： [root@mysql02 ~]# ll keepalived-1.2.24.tar.gz -rw-r--r--. 1 root root 601873 Jun 16 23:49 keepalived-1.2.24.tar.gz [root@mysql02 ~]# tar -zxvf keepalived-1.2.24.tar.gz 3、进入安装目录： [root@mysql02 ~]# cd keepalived-1.2.24 4、安装keepalived: [root@mysql03 keepalived-1.2.24]# ./configure --prefix=/usr/local/keepalived --with-kernel-dir=/usr/src/kernels/2.6.32-573.el6.x86_64/ 说明：--with-kernel-dri ：指定使用内核源码中的头文件，即include目录，该参数很重要，只有使用LVS时才用此参数。 [root@mysql03 keepalived-1.2.24]# make && make install 5、复制脚本和文件： 5.1 拷贝执行文件 [root@mysql02 keepalived-1.2.24]# cp /usr/local/keepalived/sbin/keepalived /usr/sbin/ 5.2 将init.d文件拷贝到etc下,加入开机启动项 [root@mysql02 keepalived-1.2.24]# cp /usr/local/keepalived/etc/rc.d/init.d/keepalived /etc/init.d/ 5.3 将keepalived文件拷贝到etc下，加入网卡配置 [root@mysql02 keepalived-1.2.24]# cp /usr/local/keepalived/etc/sysconfig/keepalived /etc/sysconfig/ 5.4 创建keepalived文件夹 [root@mysql02 keepalived-1.2.24]#mkdir /etc/keepalived 5.5 将keepalived配置文件拷贝到etc下 [root@mysql02 keepalived-1.2.24]# cp /usr/local/keepalived/etc/keepalived/keepalived.conf /etc/keepalived/ 5.6 复制启动脚本 [root@mysql02 keepalived-1.2.24]# cp /usr/local/keepalived/sbin/keepalived /etc/keepalived/ 6、加入开机启动项 [root@mysql02 keepalived-1.2.24]# chkconfig --add keepalived #添加时必须保证/etc/init.d/keepalived存在 7、 启动 [root@mysql02 keepalived-1.2.24]# chmod 755 /etc/init.d/keepalived [root@mysql03 keepalived-1.2.24]# service keepalived start Starting keepalived: [ OK ] 8、 关闭 [root@mysql02 keepalived-1.2.24]# service keepalived stop Stopping keepalived: [ OK ] > OK，成功！ _________________________________________________________________________________________________________ 9、 绑定虚拟IP（VIP），只在master1机器上绑定。## 说明，这一步随便做不做，不影响正常使用，作为了解。 说明： 绑定虚拟IP（VIP），这个只在master1机器上绑定即可。如下配置好keepalived后，当master1机器出现故障时，VIP资源就会被master2机器自动接管！（这一步其实可以不用这么直接在外部使用ifconfig绑定，VIP就直接在 Keepalived 的配置文件里配置就好，使用命令 ip addr 就能看出 vip；只有LVS+Keepalived 四层负载均衡的高可用环境中， vip是要在外面单独设置的（即ifconfig eth0:0 ....的方式创建vip），通过 ifconfig 可以查看出来 vip。） [root@mysql02 keepalived-1.2.24]# ifconfig eth0 Link encap:Ethernet HWaddr 00:0C:29:7C:41:00 inet addr:10.219.24.25 Bcast:10.255.255.255 Mask:255.0.0.0 inet6 addr: fe80::20c:29ff:fe7c:4100/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:12156 errors:0 dropped:0 overruns:0 frame:0 TX packets:2700 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1389340 (1.3 MiB) TX bytes:901117 (879.9 KiB) [root@mysql02 keepalived-1.2.24]# ifconfig eth0:0 10.219.24.30 broadcast 10.255.255.255 netmask 255.0.0.0 up [root@mysql02 keepalived-1.2.24]# route add -host 10.219.24.30 dev eth0:0 [root@mysql02 keepalived-1.2.24]# ifconfig eth0 Link encap:Ethernet HWaddr 00:0C:29:7C:41:00 inet addr:10.219.24.25 Bcast:10.255.255.255 Mask:255.0.0.0 inet6 addr: fe80::20c:29ff:fe7c:4100/64 Scope:Link UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 RX packets:12397 errors:0 dropped:0 overruns:0 frame:0 TX packets:2748 errors:0 dropped:0 overruns:0 carrier:0 collisions:0 txqueuelen:1000 RX bytes:1405326 (1.3 MiB) TX bytes:906805 (885.5 KiB) eth0:0 Link encap:Ethernet HWaddr 00:0C:29:7C:41:00 inet addr:10.219.24.30 Bcast:10.255.255.255 Mask:255.0.0.0 UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1 补充： 删除虚拟ip命令：ip addr del 10.219.24.30 dev eth0 以及 route del -host 10.219.24.30 dev eth0:0 _____________________________________________________________________________________________________________________ 9、 master1 机器上的keepalived.conf配置。（下面配置中没有使用lvs的负载均衡功能，所以不需要配置虚拟服务器virtual server） [root@mysql02 keepalived-1.2.24]# cp /etc/keepalived/keepalived.conf /etc/keepalived/keepalived.conf.bak -- 清空默认内容 [root@mysql02 keepalived-1.2.24]# > /etc/keepalived/keepalived.conf -- 采用如下配置 [root@mysql02 keepalived-1.2.24]# cat /etc/keepalived/keepalived.conf ! Configuration File for keepalived global_defs { router_id MYSQL-MASTER-HA } vrrp_script chk_mysql_port { script "/data/chk_mysql.sh" # mysql 健康检查脚本 interval 2 #脚本执行间隔，每2s检测一次 } vrrp_instance VI_2 { state BACKUP interface eth0 #指定虚拟ip的网卡接口 virtual_router_id 54 #路由器标识，MASTER和BACKUP必须是一致的 priority 100 advert_int 1 nopreempt authentication { auth_type PASS auth_pass 1111 } virtual_ipaddress { 10.219.24.30 # vip } track_script { chk_mysql_port } } 10、 master2机器上的keepalived配置。 master2机器上的keepalived.conf文件只修改priority为90、nopreempt不设置、real_server设置本地IP。 [root@mysql01 keepalived-1.2.24]# cp /etc/keepalived/keepalived.conf /etc/keepalived/keepalived.conf.bak [root@mysql01 keepalived-1.2.24]# >/etc/keepalived/keepalived.conf [root@mysql01 keepalived-1.2.24]# vim /etc/keepalived/keepalived.conf [root@mysql01 keepalived-1.2.24]# cat /etc/keepalived/keepalived.conf ! Configuration File for keepalived global_defs { router_id MYSQL-MASTER-HA } vrrp_script chk_mysql_port { script "/data/chk_mysql.sh" interval 2 } vrrp_instance VI_2 { state BACKUP interface eth0 virtual_router_id 54 priority 90 advert_int 1 nopreempt authentication { auth_type PASS auth_pass 1111 } virtual_ipaddress { 10.219.24.30 } track_script { chk_mysql_port } } 11、 编写切换脚本。 （master1 与 master2 都操作） KeepAlived做心跳检测，如果Master的MySQL服务挂了(3306端口挂了),那么它就会选择自杀。Slave的KeepAlived通过心跳检测发现这个情况，就会将VIP的请求接管。 -- 检查nc能否正常使用 [root@mysql02 keepalived-1.2.24]# nc -bash: nc: command not found -- 未安装nc包 ，yum安装 [root@mysql02 keepalived-1.2.24]# yum -y install nc [root@mysql02 keepalived-1.2.24]# vim /data/chk_mysql.sh [root@mysql02 ~]# cat /data/chk_mysql.sh #!/bin/bash # check mysql server status # mysql端口 PORTS="3306" function check_ports { for port in $PORTS;do nc -z 127.0.0.1 $port | grep -q succeeded [ "${PIPESTATUS[1]}" -eq 0 ] && mark=${mark}1 done # 如果mark值为空说明端口都不通。 # 如果mark等于1，说明有端口是通的。 echo $mark } ret1=$(check_ports) # 如果mysql端口不通，会尝试重启一次mysql if [ "$ret1" != 1 ];then service mysql stop mysqld_safe --defaults-file=/etc/my.cnf & q! #无用命令，只是为了跳出上面的命令 sleep 1 ret2=$(check_ports) # 如果还是有端口不通，表示mysql服务不正常，则停掉keepalived，使VIP发生切换 [ "$ret2" != 1 ] && /etc/init.d/keepalived stop fi 12、 脚本赋权 [root@mysql02 keepalived-1.2.24]# chmod 755 /data/chk_mysql.sh 13、 master1 与 master2 开启 keepalived 进程 -- master1 [root@mysql02 keepalived-1.2.24]# service keepalived start -- master2 [root@mysql01 keepalived-1.2.24]# service keepalived start 14、 测试 14.1 查看 vip 情况 ，发现 vip 只在 master1 上， tcpdump 抓包，发现只有 master1 10.219.24.25 多播包 这也符合 VRRP 工作机制 -- master1 [root@mysql02 keepalived-1.2.24]# ip a l 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:29:7c:41:00 brd ff:ff:ff:ff:ff:ff inet 10.219.24.25/8 brd 10.255.255.255 scope global eth0 inet 10.219.24.30/32 scope global eth0 inet6 fe80::20c:29ff:fe7c:4100/64 scope link valid_lft forever preferred_lft forever [root@mysql02 keepalived-1.2.24]# tcpdump -n 'host 224.0.0.18' tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on eth0, link-type EN10MB (Ethernet), capture size 65535 bytes 00:58:35.883861 IP 10.219.24.25 > 224.0.0.18: VRRPv2, Advertisement, vrid 54, prio 100, authtype simple, intvl 1s, length 20 00:58:36.885543 IP 10.219.24.25 > 224.0.0.18: VRRPv2, Advertisement, vrid 54, prio 100, authtype simple, intvl 1s, length 20 -- master2 [root@mysql01 keepalived-1.2.24]# ip a l 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:29:44:29:7a brd ff:ff:ff:ff:ff:ff inet 10.219.24.22/8 brd 10.255.255.255 scope global eth0 inet6 fe80::20c:29ff:fe44:297a/64 scope link valid_lft forever preferred_lft forever 14.2 master1 中 keepalived 关闭 ， 发现 vip 漂到了 master2 上 -- master1 [root@mysql02 keepalived-1.2.24]# service keepalived stop Stopping keepalived: [ OK ] -- master2 [root@mysql01 keepalived-1.2.24]# ip a l 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:29:44:29:7a brd ff:ff:ff:ff:ff:ff inet 10.219.24.22/8 brd 10.255.255.255 scope global eth0 inet 10.219.24.30/32 scope global eth0 inet6 fe80::20c:29ff:fe44:297a/64 scope link valid_lft forever preferred_lft forever 14.3 再将之前关闭的 keepalived 开启，发现 vip 未漂回来， 正好符合主备为 非抢占式。 [root@mysql02 keepalived-1.2.24]# service keepalived start Starting keepalived: [ OK ] [root@mysql02 keepalived-1.2.24]# ip a l 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:29:7c:41:00 brd ff:ff:ff:ff:ff:ff inet 10.219.24.25/8 brd 10.255.255.255 scope global eth0 inet6 fe80::20c:29ff:fe7c:4100/64 scope link valid_lft forever preferred_lft forever 14.4 将 master1 上的 mysql 关闭， 发现 master1 上的 keepalived 进程也被脚本 kill 了， 顺利的完成了 vip 漂回 master2 上 -- master1 [root@mysql02 keepalived-1.2.24]# ip a l 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:29:7c:41:00 brd ff:ff:ff:ff:ff:ff inet 10.219.24.25/8 brd 10.255.255.255 scope global eth0 inet6 fe80::20c:29ff:fe7c:4100/64 scope link valid_lft forever preferred_lft forever -- master2 [root@mysql01 keepalived-1.2.24]# ip a l 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:29:44:29:7a brd ff:ff:ff:ff:ff:ff inet 10.219.24.22/8 brd 10.255.255.255 scope global eth0 inet 10.219.24.30/32 scope global eth0 inet6 fe80::20c:29ff:fe44:297a/64 scope link valid_lft forever preferred_lft forever （编辑：锡盟站长网） 【声明】本站内容均来自网络，其相关言论仅代表作者个人观点，不代表本站立场。若无意侵犯到您的权利，请及时与联系站长删除相关内容! |
站长推荐