Keepalived+HAProxy实现MySQL高可用负载均衡的配置

数据库  /  管理员 发布于 6年前   243

 Keepalived

由于在生产环境使用了mysqlcluster,需要实现高可用负载均衡,这里提供了keepalived+haproxy来实现.

      keepalived主要功能是实现真实机器的故障隔离及负载均衡器间的失败切换.可在第3,4,5层交换.它通过VRRPv2(Virtual Router Redundancy Protocol) stack实现的.

      Layer3:Keepalived会定期向服务器群中的服务器.发送一个ICMP的数据包（既我们平时用的Ping程序）,如果发现某台服务的IP地址没有激活，Keepalived便报告这台服务器失效，并将它从服务器群中剔除，这种情况的典型例子是某台服务器被非法关机。Layer3的方式是以服务器的IP地址是否有效作为服务器工作正常与否的标准。

     Layer4:主要以TCP端口的状态来决定服务器工作正常与否。如web server的服务端口一般是80，如果Keepalived检测到80端口没有启动，则Keepalived将把这台服务器从服务器群中剔除。

     Layer5:在网络上占用的带宽也要大一些。Keepalived将根据用户的设定检查服务器程序的运行是否正常，如果与用户的设定不相符，则Keepalived将把服务器从服务器群中剔除。

Software Design

keepalived启动后会有单个进程

8352 ?    Ss   0:00 /usr/sbin/keepalived8353 ?    S   0:00 \_ /usr/sbin/keepalived8356 ?    S   0:01 \_ /usr/sbin/keepalived

父进程:内存管理,子进程管理等等

子进程:VRRP子进程

子进程:Healthchecking 子进程
实例

2台mysqlcluster 10.1.6.203 master  10.1.6.205 backup

vip 10.1.6.173

目的访问10.1.6.173 3366端口 分别轮询通过haproxy转发到10.1.6.203 3306 和10.1.6.205 3306

mysqlcluster搭建参照之前博客,这里在2台机上安装keepalived

[email protected]:~# apt-get install [email protected]:~# cat /etc/keepalived/keepalived.conf vrrp_script chk_haproxy {    script "killall -0 haproxy"  # verify the pid existance    interval 2          # check every 2 seconds    weight -2          # add 2 points of prio if OK} vrrp_instance VI_1 {    interface eth1        # interface to monitor    state MASTER virtual_router_id 51     # Assign one ID for this route    priority 101         # 101 on master, 100 on backup    nopreempt    debug     virtual_ipaddress {        10.1.6.173    }     track_script {    #注意大括号空格        chk_haproxy    }     notify_master /etc/keepalived/scripts/start_haproxy.sh #表示当切换到master状态时,要执行的脚本    notify_fault /etc/keepalived/scripts/stop_keepalived.sh #故障时执行的脚本    notify_stop  /etc/keepalived/scripts/stop_haproxy.sh #keepalived停止运行前运行notify_stop指定的脚本 }

VRRPD配置包括三个类:

• VRRP同步组(synchroization group)
• VRRP实例(VRRP Instance)
• VRRP脚本

这里使用了 VRRP实例, VRRP脚本

注意配置选项:

stat:指定instance(Initial)的初始状态，就是说在配置好后，这台服务器的初始状态就是这里指定的，但这里指定的不算，还是得要通过竞选通过优先级来确定，里如果这里设置为master，但如若他的优先级不及另外一台，那么这台在发送通告时，会发送自己的优先级，另外一台发现优先级不如自己的高，那么他会就回抢占为master

• interface：实例绑定的网卡，因为在配置虚拟IP的时候必须是在已有的网卡上添加的
• priority 101：设置本节点的优先级，优先级高的为master
• debug：debug级别
• nopreempt:设置为不抢占

vrrp_script chk_haproxy {    script "killall -0 haproxy"  # verify the pid existance    interval 2          # check every 2 seconds 脚本执行间隔    weight -2          # add 2 points of prio if OK 脚本结果导致的优先级变更：2表示优先级+2；-2则表示优先级-2}

然后在实例(vrrp_instance)里面引用，有点类似脚本里面的函数引用一样：先定义，后引用函数名

    track_script {        chk_haproxy    }

注意：VRRP脚本(vrrp_script)和VRRP实例(vrrp_instance)属于同一个级别

[email protected]:scripts# cat start_haproxy.sh #!/bin/bash sleep 5get=`ip addr |grep 10.1.6.173 |wc -l`echo $get >> /etc/keepalived/scripts/start_ha.log if [ $get -eq 1 ]then    echo "`date +%c` success to get vip" >> /etc/keepalived/scripts/start_ha.log    /usr/local/sbin/haproxy -f /etc/haproxy/haproxy.cfgelse    echo "`date +%c` can not get vip" >> /etc/keepalived/scripts/[email protected]:scripts# cat stop_keepalived.sh #!/bin/bash pid=`pidof keepalived`if [ $pid == "" ]then  echo "`date +%c` no keepalived process id" >> /etc/keepalived/scripts/stop_keep.logelse  echo "`date +%c` will stop keepalived " >> /etc/keepalived/scripts/stop_keep.log  /etc/init.d/keepalived stopfi /etc/init.d/keepalived stop [email protected]:scripts# cat stop_haproxy.sh #!/bin/bash pid=`pidof haproxy`echo "`date +%c` stop haproxy" >> /etc/keepalived/scripts/stop_ha.logkill -9 $pid

同理配置10.1.6.205

[email protected]:~# cat /etc/keepalived/keepalived.conf vrrp_script chk_haproxy {  script "killall -0 haproxy"  # verify the pid existance  interval 2          # check every 2 seconds  weight 2           # add 2 points of prio if OK} vrrp_instance VI_1 {  interface eth1        # interface to monitor  state BACKUP  virtual_router_id 51     # Assign one ID for this route  priority 100         # 101 on master, 100 on backup  virtual_ipaddress {    10.1.6.173  }   track_script {    chk_haproxy  } notify_master /etc/keepalived/scripts/start_haproxy.shnotify_fault /etc/keepalived/scripts/stop_keepalived.shnotify_stop /etc/keepalived/scripts/stop_haproxy.sh }

 HAProxy

下面再介绍下haproxy

       HAProxy是一款基于TCP(第四层)和HTTP(第七层)应用的代理软件,它也可作为负载均衡器.可以支持数以万计的并发连接.同时可以保护服务器不暴露到网络上,通过端口映射.它还自带监控服务器状态的页面.

      安装haproxy

wget -O/tmp/haproxy-1.4.22.tar.gz http://haproxy.1wt.eu/download/1.4/src/haproxy-1.4.22.tar.gztar xvfz /tmp/haproxy-1.4.22.tar.gz -C /tmp/cd /tmp/haproxy-1.4.22make TARGET=linux26make install

      haproxy需要对每一个mysqlcluster服务器进行健康检查

1.在2台主机分别配置haproxy.cfg

[email protected]:scripts# cat /etc/haproxy/haproxy.cfg global    maxconn 51200 #默认最大连接数     #uid 99    #gid 99    daemon    #以后台形式运行haproxy    #quiet    nbproc 1   #进程数量(可以设置多个进程提高性能)     pidfile /etc/haproxy/haproxy.pid #haproxy的pid存放路径,启动进程的用户必须有权限访问此文件  defaults    mode tcp      #所处理的类别 (#7层 http;4层tcp )     option redispatch  #serverId对应的服务器挂掉后,强制定向到其他健康的服务器     option abortonclose #当服务器负载很高的时候，自动结束掉当前队列处理比较久的连接     timeout connect 5000s  #连接超时    timeout client 50000s #客户端超时    timeout server 50000s  #服务器超时    log 127.0.0.1 local0  #错误日志记录    balance roundrobin  #默认的负载均衡的方式,轮询方式  listen proxy    bind 10.1.6.173:3366  #监听端口     mode tcp        #http的7层模式    option httpchk    #心跳检测的文件    server db1 10.1.6.203:3306 weight 1 check port 9222 inter 12000 rise 3 fall 3   #服务器定义,check inter 12000是检测心跳频率 rise 3是3次正确认为服务器可用， fall 3是3次失败认为服务器不可用，weight代表权重     server db2 10.1.6.205:3306 weight 1 check port 9222 inter 12000 rise 3 fall 3 listen haproxy_stats    mode http    bind 10.1.6.173:8888    option httplog    stats refresh 5s      stats uri /status #网站健康检测URL，用来检测HAProxy管理的网站是否可以用，正常返回200，不正常返回503     stats realm Haproxy Manager    stats auth admin:p@a1SZs24 #账号密码[email protected]:~$ cat /etc/haproxy/haproxy.cfg global    maxconn 51200    #uid 99    #gid 99    daemon    #quiet    nbproc 1    pidfile /etc/haproxy/haproxy.pid defaults    mode tcp    option redispatch      option abortonclose    timeout connect 5000s    timeout client 50000s    timeout server 50000s    log 127.0.0.1 local0    balance roundrobin  listen proxy    bind 10.1.6.173:3366    mode tcp    option httpchk    server db1 10.1.6.203:3306 weight 1 check port 9222 inter 12000 rise 3 fall 3    server db2 10.1.6.205:3306 weight 1 check port 9222 inter 12000 rise 3 fall 3 listen haproxy_stats    mode http    bind 10.1.6.173:8888    option httplog    stats refresh 5s      stats uri /status     stats realm Haproxy Manager    stats auth admin:p@a1SZs24

2.安装xinetd

[email protected]:~# apt-get install xinetd

3.在每个节点添加xinetd服务脚本和mysqlchk端口号

[email protected]:~# vim /etc/xinetd.d/mysqlchk # default: on# description: mysqlchkservice mysqlchk         #需要在servive定义{    flags      = REUSE    socket_type   = stream    port      = 9222    wait      = no    user      = nobody    server     = /opt/mysqlchk     log_on_failure += USERID    disable     = no    per_source   = UNLIMITED    bind      = 10.1.6.173} [email protected]:~# vim /etc/services mysqlchk    9222/tcp# mysqlchk

4.编写mysqlchk监控服务脚本

[email protected]:~# ls -l /opt/mysqlchk -rwxr--r-- 1 nobody root 1994 2013-09-17 11:27 /opt/[email protected]:~# cat /opt/mysqlchk #!/bin/bash## This script checks if a mysql server is healthy running on localhost. It will# return:# "HTTP/1.x 200 OK\r" (if mysql is running smoothly)# - OR -# "HTTP/1.x 500 Internal Server Error\r" (else)## The purpose of this script is make haproxy capable of monitoring mysql properly# MYSQL_HOST="localhost"MYSQL_SOCKET="/var/run/mysqld/mysqld.sock"MYSQL_USERNAME="mysqlchkusr"   #该账户密码需要在mysql里添加MYSQL_PASSWORD="secret"MYSQL_OPTS="-N -q -A"TMP_FILE="/dev/shm/mysqlchk.$$.out"ERR_FILE="/dev/shm/mysqlchk.$$.err"FORCE_FAIL="/dev/shm/proxyoff"MYSQL_BIN="/opt/mysqlcluster/mysql-cluster-gpl-7.2.6-linux2.6-x86_64/bin/mysql"CHECK_QUERY="select 1" preflight_check(){  for I in "$TMP_FILE" "$ERR_FILE"; do    if [ -f "$I" ]; then      if [ ! -w $I ]; then        echo -e "HTTP/1.1 503 Service Unavailable\r\n"        echo -e "Content-Type: Content-Type: text/plain\r\n"        echo -e "\r\n"        echo -e "Cannot write to $I\r\n"        echo -e "\r\n"        exit 1      fi    fi  done} return_ok(){  echo -e "HTTP/1.1 200 OK\r\n"  echo -e "Content-Type: text/html\r\n"  echo -e "Content-Length: 43\r\n"  echo -e "\r\n"  echo -e "<html><body>MySQL is running.</body></html>\r\n"  echo -e "\r\n"  rm $ERR_FILE $TMP_FILE  exit 0}return_fail(){  echo -e "HTTP/1.1 503 Service Unavailable\r\n"  echo -e "Content-Type: text/html\r\n"  echo -e "Content-Length: 42\r\n"  echo -e "\r\n"  echo -e "<html><body>MySQL is *down*.</body></html>\r\n"  sed -e 's/\n$/\r\n/' $ERR_FILE  echo -e "\r\n"  rm $ERR_FILE $TMP_FILE  exit 1}preflight_checkif [ -f "$FORCE_FAIL" ]; then    echo "$FORCE_FAIL found" > $ERR_FILE    return_fail;fi$MYSQL_BIN $MYSQL_OPTS --host=$MYSQL_HOST --socket=$MYSQL_SOCKET --user=$MYSQL_USERNAME --password=$MYSQL_PASSWORD -e "$CHECK_QUERY" > $TMP_FILE 2> $ERR_FILEif [ $? -ne 0 ]; then    return_fail;fireturn_ok;

测试

2个节点开启keepalived(主节点会获得vip,自动拉起haproxy),xinetd

[email protected]:~# ip add1: lo: <LOOPBACK,UP,LOWER_UP> mtu 16436 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 lo2: eth0: <BROADCAST,MULTICAST> mtu 1500 qdisc pfifo_fast state DOWN qlen 1000  link/ether 00:26:b9:36:0f:81 brd ff:ff:ff:ff:ff:ff  inet 211.151.105.186/26 brd 211.151.105.191 scope global eth03: eth1: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP qlen 1000  link/ether 00:26:b9:36:0f:83 brd ff:ff:ff:ff:ff:ff  inet 10.1.6.203/24 brd 10.1.6.255 scope global eth1  inet 10.1.6.173/32 scope global eth14: eth2: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000  link/ether 00:26:b9:36:0f:85 brd ff:ff:ff:ff:ff:ff5: eth3: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN qlen 1000  link/ether 00:26:b9:36:0f:87 brd ff:ff:ff:ff:ff:[email protected]:~# netstat -tunlp | grep hatcp    0   0 10.1.6.173:3366   0.0.0.0:*        LISTEN   1042/haproxy  tcp    0   0 10.1.6.203:8888   0.0.0.0:*        LISTEN   1042/haproxy  udp    0   0 0.0.0.0:56562      0.0.0.0:*  1042/haproxy  [email protected]:~# netstat -tunlp | grep xinetcp    0   0 10.1.6.203:9222   0.0.0.0:*        LISTEN   30897/xinetd  [email protected]:~# ps -ef | grep haproxyroot   1042   1 0 Sep17 ?    00:00:00 /usr/local/sbin/haproxy -f /etc/haproxy/haproxy.cfg

测试:

通过vip10.1.6.173 3366访问cluster数据库(注意账户dave权限需要加3个ip10.1.6.203,10.1.6.205,10.1.6.173)

[email protected]:mgm# mysql -udave -p -h 10.1.6.173 -P 3366Enter password: Welcome to the MySQL monitor. Commands end with ; or \g.Your MySQL connection id is 1344316Server version: 5.5.22-ndb-7.2.6-gpl-log MySQL Cluster Community Server (GPL) Type 'help;' or '\h' for help. Type '\c' to clear the buffer. mysql> show databases;+--------------------+| Database      |+--------------------+| information_schema | | dave       | | test        | +--------------------+3 rows in set (0.01 sec) mysql>

手动分别使keepalive,haproxy,数据库挂掉.vip10.1.6.173会自动漂到10.1.6.205从上，并不影响vip的访问

通过vip,haproxy查看各节点状态

http://10.1.6.173:8888/status