nba买球:Keepalived HAProxy实现MySQL高可用负载均衡的

作者:计算机教程

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.

测试

vip 10.1.6.173

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

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

root@10.1.6.203:~# apt-get install keepalivedroot@10.1.6.203:~# 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 }

2台mysqlcluster 10.1.6.203 master 10.1.6.205 backup

haproxy需要对每一个mysqlcluster服务器进行健康检查

root@10.1.6.205:~# 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 }

1.在2台主机分别配置haproxy.cfg

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。

下面再介绍下haproxy

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

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

通过vip,haproxy查看各节点状态

HAProxy

root@10.1.6.203: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

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.

子进程:VRRP子进程

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

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

Keepalived

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)

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.

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

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。

2.安装xinetd

VRRPD配置包括三个类:

root@10.1.6.203: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/start_ha.logfiroot@10.1.6.203: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 root@10.1.6.203:scripts# cat stop_haproxy.sh #!/bin/bash pid=`pidof haproxy`echo "`date  %c` stop haproxy"  /etc/keepalived/scripts/stop_ha.logkill -9 $pid
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}

测试:

Layer3:Keepalived会定期向服务器群中的服务器.发送一个ICMP的数据包,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。

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

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

同理配置10.1.6.205

3.在每个节点添加xinetd服务脚本和mysqlchk端口号

root@10.1.6.203: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  #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  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root@10.1.6.205:~$ 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  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  stats refresh 5s stats uri /status stats realm Haproxy Manager stats auth admin:p@a1SZs24
root@10.1.6.203:~# ls -l /opt/mysqlchk -rwxr--r-- 1 nobody root 1994 2013-09-17 11:27 /opt/mysqlchkroot@10.1.6.203:~# cat /opt/mysqlchk #!/bin/bash## This script checks if a mysql server is healthy running on localhost. It will# return:# "HTTP/1.x 200 OKr" (if mysql is running smoothly)# - OR -# "HTTP/1.x 500 Internal Server Errorr" (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 Unavailablern" echo -e "Content-Type: Content-Type: text/plainrn" echo -e "rn" echo -e "Cannot write to $Irn" echo -e "rn" exit 1 fi fi done} return_ok(){ echo -e "HTTP/1.1 200 OKrn" echo -e "Content-Type: text/htmlrn" echo -e "Content-Length: 43rn" echo -e "rn" echo -e "htmlbodyMySQL is running./body/htmlrn" echo -e "rn" rm $ERR_FILE $TMP_FILE exit 0}return_fail(){ echo -e "HTTP/1.1 503 Service Unavailablern" echo -e "Content-Type: text/htmlrn" echo -e "Content-Length: 42rn" echo -e "rn" echo -e "htmlbodyMySQL is *down*./body/htmlrn" sed -e 's/n$/rn/' $ERR_FILE echo -e "rn" 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;
wget -O/tmp/haproxy-1.4.22.tar.gz  xvfz /tmp/haproxy-1.4.22.tar.gz -C /tmp/cd /tmp/haproxy-1.4.22make TARGET=linux26make install

这里使用了 VRRP实例, VRRP脚本

注意配置选项:

root@10.1.6.203:~# apt-get install xinetd
 track_script { chk_haproxy }

父进程:内存管理,子进程管理等等

子进程:Healthchecking 子进程实例

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

root@10.1.6.203:~# 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} root@10.1.6.203:~# vim /etc/services mysqlchk 9222/tcp # mysqlchk
root@10.1.6.203:~# 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:ffroot@10.1.6.203:~# 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 root@10.1.6.203:~# netstat -tunlp | grep xinetcp 0 0 10.1.6.203:9222 0.0.0.0:* LISTEN 30897/xinetd root@10.1.6.203:~# ps -ef | grep haproxyroot 1042 1 0 Sep17  00:00:00 /usr/local/sbin/haproxy -f /etc/haproxy/haproxy.cfg

安装haproxy

4.编写mysqlchk监控服务脚本

Software Design

本文由nba买球发布,转载请注明来源

关键词: