keepalived root权限必须用root用吗
点击联系发帖人
时间：2016-03-04 02:33
Linux开源集群架构之keepalived应用详解_图文_百度文库
两大类热门资源免费畅读
续费一年阅读会员，立省24元！
Linux开源集群架构之keepalived应用详解
上传于||文档简介
&&k​e​e​p​a​l​i​v​e​d​安​装​
​
​应​用​
​
​架​设​高​可​用​M​Y​S​Q​L​
​
​整​合​L​V​S
阅读已结束，如果下载本文需要使用
想免费下载本文？
下载文档到电脑，查找使用更方便
还剩83页未读，继续阅读
你可能喜欢使用keepalived 实现MYSQL 高可用-mysql-电脑编程网使用keepalived 实现MYSQL 高可用作者： 和相关&&使用keepalived 实现MYSQL 高可用
keepalived
下载地址：http://www.keepalived.org/download.html
master:192.168.215.80
slave :192.168.215.28
VIP
:192.168.215.34
安装 keepalived
#=======================================================================
1:#wget http://www.keepalived.org/software/keepalived-1.1.20.tar.gz
2:#tar zxvf keepalived-1.1.20.tar.gz
3:#cd keepalived-1.1.20
4:#./configure
在这一步，可能看到的是如下：
#========================================================================
Keepalived configuration
------------------------
Keepalived version
: 1.1.20
Compiler
: gcc
Compiler flags
: -g -O2
Extra Lib
: -lpopt -lssl -lcrypto
Use IPVS Framework
: No
IPVS sync daemon support : No
Use VRRP Framework
: Yes
Use Debug flags
: No
#==========================================================================
出现3个NO，一个YES。解决的办法是编译时带上内核文件夹重新编译
# ./configure --with-kernel-dir=/usr/src/kernels/2.6.18-194.26.1.el5-x86_64/
Keepalived configuration
------------------------
Keepalived version
: 1.1.20
Compiler
: gcc
Compiler flags
: -g -O2
Extra Lib
: -lpopt -lssl -lcrypto
Use IPVS Framework
: Yes
IPVS sync daemon support : Yes
Use VRRP Framework
: Yes
Use Debug flags
: No
#==========================================================================
5:#make && make install
#=======================================================================
到这里，安装完成。下面写配置文件。
配置 keepalived(master,192.168.215.80)
#=======================================================================
#mkdir /etc/keepalived
#默认的配置文件获取的路径
#vi /etc/keepalived/keepalived.conf
#配置文件的名称（把下面复制即可）
! Configuration File for keepalived
global_defs {
#全局定义
&&&&¬ification_email {
#告警邮件地址组（故障发邮件在这里没有测试，我把它放到健康检查脚本里面了）
&&&&&
#发生故障，发告警的邮件地址，一般不用，可以随便写
&&&&&}
&&&&¬ification_email_from
#邮件发送地址
&&&&&smtp_server 127.0.0.1
#SMTPIP
&&&&&smtp_connect_timeout 30
#邮件连接超时最长时间
&&&&&router_id MySQL-ha
#路由标识。在全局定义里只有这个是必须的，其他都可以不要。
&&&&&}
vrrp_script check__alived {
#检查MYSQL状态的函数（注意这里的花括号和名称之间有空格）
&&&&script "/root/-moniter.sh"
#这里定义脚本，注意本行需要缩进，我原来没注意，在这上面整了很长时间。可以在这个脚本里检查MYSQL是否还活着，服务器的负载是否过高等，以及什么条件做什么操作。
&&&&&&&&interval 1
#运行上面检查脚本的时间间隔，单位为妙。keepalive的守护进程执行这个脚本
&&&&&&&&weight 5
#本脚本的运行权重，权重越高的，越优先检查（只有一个检查脚本，就无所谓了）
}
vrrp_instance VI_1 {
&&&&&state BACKUP
#两台配置此处均是BACKUP，一台宕机，另外一台把VIP抢过去。
&&&&&interface eth1
#在这里要注意，虚拟IP用的是哪个网卡进行通讯。我这里机器上的eth1是内网的，eth0是外网的。
&&&&&virtual_router_id 51
#虚拟路由ID。这个标识是同一个vrrp实例使用唯一的标识。即同一个vrrp_stance,MASTER和BACKUP的virtual_router_id是一致的，同时在整个vrrp内是唯一的。
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&#keepalived通过发送和接受组广播包中的同一个virtual_router_id中的成员的存活，来确定对方的不可用，一旦检测到对方不可用，立即切换他的备份角色为主。
&&&&&priority 100
#权重，数值越大，权重越大，Master必须大于Slave。另一台改为90。
&&&&&advert_int 1
#Master和Slave负载均衡器之间同步检查的时间间隔，单位：秒
&&&&&#nopreempt
#抢占，在MASTER恢复后，把VIP抢过来。在SLAVE端也一样，如果MASTER出问题了，SLAVE就把VIP抢过来。
&&&&&authentication {
#Master和Slave之间认证的方式
&&&&&auth_type PASS
&&&&&auth_pass 1111
#认证的密码
&&&&&}
&&&&&track_script{
# 使用的VRRP脚本（检查健康脚步），上面是定义，这里是使用。
&&&&&&&&&check_mysql_alived
&&&&&}
&&&&&virtual_ipaddress {
#VIP的地址
&&&&&&&&&192.168.215.34
&&&
virtual_server 192.168.215.34 3306 {
&&&&&lb_algo wrr
&&&&&lb_kind DR
&&&&&persistence_timeout 60
#会话保持时间，单位为秒，当用户从远程帐号过行登陆网站时，有了这个会话保持功能，就能把用户的请求转发到同一个应用服务器
&&&&&protocol TCP
&&&&&real_server 192.168.215.80 3306 {
#真实IP（每台机器必须指定各自的真实IP）
&&&&&TCP_CHECK {
connect_timeout 10
#连接超时时间
nb_get_retry 3
delay_before_retry 3
#重连间隔时间
connect_port 3306
#健康检查端口
&&&&&&&&&&&&&}
&&&&&&&&&}
&&&&&}
#==========================================================================
MATER端配置到此结束。
编写健康服务检测的脚本
#==========================================================================
[root@l10ntermwiki bruce]# cat mysql-moniter.sh
#monitor mysql service status and server load
#!/bin/bash
function Send_Email()
#发邮件的函数。
{
ADMIN=""
MESSAGE_FILE="/tmp/monitor.log"
#邮件里面带上本机IP地址和时间，信息更明确。
&&&&&&&&echo "VIP has been changed AT `date`,IP:`cat /etc/hosts |grep -v "127.0.0.1"|awk '{print $1}'`,please check it. Message:$1" & $MESSAGE_FILE
&&&&&&&&/bin/mail -s "VIP has been changed" $ADMIN & $MESSAGE_FILE
rm -r $MESSAGE_FILE
}
CPU_NUMBER=`cat /proc/cpuinfo |grep processor |wc -l`
#获取CPU个数。
MAX_LOAD=`echo $CPU_NUMBER*2|bc`
#最大负载认为是CPU*2。设置成CPU个数更好。
if [ `service mysql status |awk '{print $2}'` = 'running' ] #检查mysql是否活着，这里用mysql作为服务名称，有些用mysqld作为服务名称。
then
&&&LOAD=`uptime`
&&&LOAD=${LOAD##* }
#
if [ `expr $LOAD \& $CPU_LOAD` = 1 ]
# another method,compare integer and decimal
&&&if [ `echo "$LOAD & $MAX_LOAD" | bc` = 1 ]
#如果负载过高。
&&&then
&&&&&&pkill keepalived
# if load is greater than max load then kill keepalived,take VIP point to another server.
&&&&&&Send_Email "Mysql is alived,Average load is busy,so kill keepalived"
&&&fi
else
#mysql服务挂了，VIP漂移走。
&&&pkill keepalived # if mysql service is not running ,then kill keepalived,take VIP point to another server.
&&&Send_Email "Mysql service is not alived,so kill keepalived."
fi
echo `date`&&/tmp/pk.log
#这里只是为了检查脚本是否被正常执行，实际使用删除掉。
#健康检查脚本完毕。
#chmod +x /home/bruce/MySQL.sh
#分配执行权限
#==========================================================================
启动keepalvice
#==========================================================================
[root@ldc2095 keepalived-1.1.20]# keepalived -D
#-D参数，会把日志记录到/var/log/message里面
[root@ldc2095 keepalived-1.1.20]# ps aux |grep keep
root
0:00 keepalived -D
root
0:00 keepalived -D
root
0:00 keepalived -D
root
0:00 grep keep
#===========================================================================
另外一个机器的配置和上面的一样。除了修改修改的几个地方。
健康检查脚本也一致。相关资料：｜｜｜｜｜｜｜使用keepalived 实现MYSQL 高可用来源网络，如有侵权请告知，即处理！编程Tags： & dngood 的BLOG
用户名：dngood
文章数：177
评论数：396
访问量：1754889
注册日期：
阅读量：24883
阅读量：235739
阅读量：980595
阅读量：140504
51CTO推荐博文
本文为& &haproxy + keepalived 一键安装包&& 的 第四篇
目的:帮助快速设置haproxy 与 keepalived 配置文档
(keepalived 为haproxy 做 HA 高可用）
第四篇 haproxy + keepalived &一键安装包&& 配置与测试HOWTO
一 haproxy.sh 与 keepalived.sh 服务脚本&
[root@four&cluster]#&ls&/usr/local/sbin/*.ah&haproxy.sh&&keepalived.sh&
二 设置自动启动 haproxy.sh 与 keepalived.sh
[root@four&cluster]#&cat&/etc/rc.local&&#!/bin/sh&#&#&This&script&will&be&executed&*after*&all&the&other&init&scripts.&#&You&can&put&your&own&initialization&stuff&in&here&if&you&don't&#&want&to&do&the&full&Sys&V&style&init&stuff.&&touch&/var/lock/subsys/local&ulimit&-SHn&65535&&/usr/local/sbin/haproxy.sh&start&/usr/local/sbin/keepalived.sh&start&
三 haproxy 相关配置
3.1&haroxy&程序目录&#ls&/usr/local/haproxy&doc&&haproxy.cfg&&html&&logs&&sbin&&share&&3.2&haproxy&配置文档&[root@four&cluster]#&file&/usr/local/haproxy/haproxy.cfg&&/usr/local/haproxy/haproxy.cfg:&UTF-8&Unicode&text&&3.3&编辑haproxy&配置文档&[root@four&cluster]#&vim&/usr/local/haproxy/haproxy.cfg&&第37&行&定义健康检查html文档，所以你需要在你的web服务上建立check.html文件用于haproxy&健康检查，也可以跟具需要自定义检查文档。&option&&httpchk&HEAD&/check.html&#用来做健康检查html文档&第38&39&行修10.0.1.252/253&替换成你的web&服务器ip地址&server&server1&10.0.1.252:80&cookie&server1&check&inter&2000&rise&3&&&&&fall&3&weight&3&server&server2&10.0.1.253:80&cookie&server2&check&inter&2000&rise&3&&&&&fall&3&maxconn&120&weight&3&&3.4&启动haproxy&输入如下命令&&[root@four&cluster]#&haproxy.sh&start&&3.5&检查是否有haproxy&进程&[root@four&cluster]#&ps&aux&|&grep&haproxy&|&grep&-v&'grep'&nobody&&&&&&0.2&&&&?&&&&&&&&Ss&&&20:35&&&0:00&/usr/local/haproxy/sbin/haproxy&-f&/usr/local/haproxy/haproxy.cfg&&3.6&检查haproxy&是否监听80&端口&[root@four&cluster]#&lsof&-i&:80&COMMAND&&PID&&&USER&&&FD&&&TYPE&DEVICE&SIZE&NODE&NAME&haproxy&4216&nobody&&&&4u&&IPv4&&&8229&&&&&&&TCP&*:http&(LISTEN)&&3.7&简单测试&haproxy&看看否返回&&200&OK&&[root@four&cluster]#&curl&-IL&http://10.0.1.254&(换成你的ip&)&HTTP/1.1&200&OK&Server:&nginx/1.0.5&Date:&Wed,&28&Dec&:30&GMT&Content-Type:&text/html&Content-Length:&159&Last-Modified:&Sat,&03&Dec&:10&GMT&Connection:&close&Accept-Ranges:&bytes&
四 keepalived 相关配置
4.1&keepalived&程序目录&[root@four&cluster]#&ls&/usr/local/keepalived&bin&&etc&&haproxy_check.sh&&keepalived.conf&&sbin&&share&&4.2&haproxy_check.sh&shell脚本监控程序,用于检查&80端口，来判断haproxy&是否可用&[root@four&cluster]#&file&/usr/local/keepalived/haproxy_check.sh&&/usr/local/keepalived/haproxy_check.sh:&Bourne-Again&shell&script&text&executable&&4.3&keepalived&配置文档&[root@four&cluster]#&file&/usr/local/keepalived/keepalived.conf&&/usr/local/keepalived/keepalived.conf:&ASCII&text&&4.4&编辑keepalived&配置文档&[root@four&cluster]#&vim&/usr/local/keepalived/keepalived.conf&&第16&行&interface&默认值ð1&,改成你的haproxy&外网卡(公网ip网卡)&interfaceð0&&第&34&行&默认值&10.0.1.250/24&devð1&，10.0.1.250&替换成你virtual&ip&地址,devð1&改成你的haproxy&外网卡(与第16行一致)&10.0.1.250/24&devð0&scope&global&&4.5&启动keepalived&输入如下命令&&[root@four&cluster]#&keepalived.sh&start&[root@four&cluster]#&4242&&&&&&&&&&&&&&&&#此处返回的是keepalived&进程id，敲下回车键即可&&4.6&检查是否有keepalived&进程&#&ps&aux&|&grep&keepalived&|&grep&-v&'grep'&&root&&&&&&&&0.0&&63844&&&464&pts/0&&&&S&&&&20:52&&&0:00&/bin/bash&/usr/local/sbin/keepalived.sh&start&root&&&&&&&&0.1&&35776&&&608&?&&&&&&&&Ss&&&20:52&&&0:00&/usr/local/keepalived/sbin/keepalived&-f&/usr/local/keepalived/keepalived.conf&root&&&&&&&&0.1&&3&?&&&&&&&&S&&&&20:52&&&0:00&/usr/local/keepalived/sbin/keepalived&-f&/usr/local/keepalived/keepalived.conf&root&&&&&&&&0.2&&6&pts/0&&&&S&&&&20:52&&&0:00&/bin/bash&/usr/local/keepalived/haproxy_check.sh&start&&4.7&查看keepalived&输出日志&[root@four&cluster]#&tail&/var/log/messages&&Dec&28&20:52:28&four&Keepalived_vrrp:&Registering&Kernel&netlink&reflector&Dec&28&20:52:28&four&Keepalived_vrrp:&Registering&Kernel&netlink&command&channel&Dec&28&20:52:28&four&Keepalived_vrrp:&Registering&gratutious&ARP&shared&channel&Dec&28&20:52:28&four&Keepalived:&Starting&VRRP&child&process,&pid=4241&Dec&28&20:53:08&four&Keepalived_vrrp:&Opening&file&'/usr/local/keepalived/keepalived.conf'.&&Dec&28&20:53:08&four&Keepalived_vrrp:&Configuration&is&using&:&63577&Bytes&Dec&28&20:53:08&four&Keepalived_vrrp:&Using&LinkWatch&kernel&netlink&reflector...&Dec&28&20:53:08&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&Dec&28&20:53:12&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Transition&to&MASTER&STATE&Dec&28&20:53:13&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&&&&&&&&&&&&&&#当前keepalived&状态为master&
五 haproxy 与 keepalived 整体测试
5.1&查看&haproxy&与&keepalived&进程树&[root@four&cluster]#&pstree&&init─┬─crond&&&&&&├─events/0&&&&&&├─events/1&&&&&&├─haproxy&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&#haproxy&进程&&&&&&├─keepalived───keepalived&&&&&&&&&&&&&&&&&&&&&&#keepalived&进程(2个)&&&&&&├─keepalived.sh───haproxy_check.s───sleep&&&&&&#keepalived.sh&调用&haproxy_check.sh&(监控脚本)&调用&sleep&（用于控检查间隔时间,2s）&&&5.2&模拟网络故障&haproxy&切换测试&&haproxy&1&日志&[root@one&~]#&tail&/var/log/messages&-f&Dec&28&21:09:43&one&kernel:&NFSD:&starting&90-second&grace&period&Dec&28&21:09:44&one&Keepalived:&Starting&Keepalived&v1.2.2&(12/25,2011)&&Dec&28&21:09:44&one&Keepalived_vrrp:&Registering&Kernel&netlink&reflector&Dec&28&21:09:44&one&Keepalived_vrrp:&Registering&Kernel&netlink&command&channel&Dec&28&21:09:44&one&Keepalived_vrrp:&Registering&gratutious&ARP&shared&channel&Dec&28&21:09:44&one&Keepalived:&Starting&VRRP&child&process,&pid=1993&Dec&28&21:10:24&one&Keepalived_vrrp:&Opening&file&'/usr/local/keepalived/keepalived.conf'.&&Dec&28&21:10:24&one&Keepalived_vrrp:&Configuration&is&using&:&63597&Bytes&Dec&28&21:10:24&one&Keepalived_vrrp:&Using&LinkWatch&kernel&netlink&reflector...&Dec&28&21:10:24&one&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&&haproxy&2&日志&[root@four&cluster]#&ifdownð0&[root@four&cluster]#&tail&/var/log/messages&&Dec&28&20:52:28&four&Keepalived:&Starting&VRRP&child&process,&pid=4241&Dec&28&20:53:08&four&Keepalived_vrrp:&Opening&file&'/usr/local/keepalived/keepalived.conf'.&&Dec&28&20:53:08&four&Keepalived_vrrp:&Configuration&is&using&:&63577&Bytes&Dec&28&20:53:08&four&Keepalived_vrrp:&Using&LinkWatch&kernel&netlink&reflector...&Dec&28&20:53:08&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&Dec&28&20:53:12&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Transition&to&MASTER&STATE&Dec&28&20:53:13&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&Dec&28&21:12:07&four&Keepalived_vrrp:&Kernel&is&reporting:&interfaceð0&DOWN&Dec&28&21:12:07&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&FAULT&STATE&Dec&28&21:12:07&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Now&in&FAULT&state&&&&&&&&&&&&&&&&&#keepalived&fault&状态&&haproxy&1&日志&[root@one&~]#&tail&/var/log/messages&-f&Dec&28&21:09:43&one&kernel:&NFSD:&starting&90-second&grace&period&Dec&28&21:09:44&one&Keepalived:&Starting&Keepalived&v1.2.2&(12/25,2011)&&Dec&28&21:09:44&one&Keepalived_vrrp:&Registering&Kernel&netlink&reflector&Dec&28&21:09:44&one&Keepalived_vrrp:&Registering&Kernel&netlink&command&channel&Dec&28&21:09:44&one&Keepalived_vrrp:&Registering&gratutious&ARP&shared&channel&Dec&28&21:09:44&one&Keepalived:&Starting&VRRP&child&process,&pid=1993&Dec&28&21:10:24&one&Keepalived_vrrp:&Opening&file&'/usr/local/keepalived/keepalived.conf'.&&Dec&28&21:10:24&one&Keepalived_vrrp:&Configuration&is&using&:&63597&Bytes&Dec&28&21:10:24&one&Keepalived_vrrp:&Using&LinkWatch&kernel&netlink&reflector...&Dec&28&21:10:24&one&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&Dec&28&21:12:08&one&Keepalived_vrrp:&VRRP_Instance(VI_1)&Transition&to&MASTER&STATE&Dec&28&21:12:09&one&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&&&&&&&&&&&&&#成功切换为&master&&&haproxy&1&检查vip&&[root@one&~]#&ip&add&1:&lo:&,UP,LOWER_UP&mtu&16436&qdisc&noqueue&&&&&&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:ð0:&,MULTICAST,UP,LOWER_UP&mtu&1500&qdisc&pfifo_fast&qlen&1000&&&&&link/ether&08:00:27:74:3c:05&brd&ff:ff:ff:ff:ff:ff&&&&&inet&10.0.2.251/24&brd&10.0.2.255&scope&globalð0&&&&&inet6&fe80::a00:27ff:fe74:3c05/64&scope&link&&&&&&&&&valid_lft&forever&preferred_lft&forever&3:ð1:&,MULTICAST,UP,LOWER_UP&mtu&1500&qdisc&pfifo_fast&qlen&1000&&&&&link/ether&08:00:27:3e:9a:2b&brd&ff:ff:ff:ff:ff:ff&&&&&inet&10.0.1.251/24&brd&10.0.1.255&scope&globalð1&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&& &&&&inet&10.0.1.250/24&scope&global&secondaryð1&
#virtual ip&&&&inet6&fe80::a00:27ff:fe3e:9a2b/64&scope&link&&&&&&&&&valid_lft&forever&preferred_lft&forever&4:&sit0:&&mtu&1480&qdisc&noop&&&&&&link/sit&0.0.0.0&brd&0.0.0.0&&5.3&模拟haproxy&80端口故障&keepalived&切换测试&&haproxy&2&恢复&[root@four&cluster]#&ifupð0&haproxy&2&日志&[root@four&cluster]#&tail&/var/log/messages&-f&Dec&28&20:53:08&four&Keepalived_vrrp:&Configuration&is&using&:&63577&Bytes&Dec&28&20:53:08&four&Keepalived_vrrp:&Using&LinkWatch&kernel&netlink&reflector...&Dec&28&20:53:08&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&Dec&28&20:53:12&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Transition&to&MASTER&STATE&Dec&28&20:53:13&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&Dec&28&21:12:07&four&Keepalived_vrrp:&Kernel&is&reporting:&interfaceð0&DOWN&Dec&28&21:12:07&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&FAULT&STATE&Dec&28&21:12:07&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Now&in&FAULT&state&Dec&28&21:15:34&four&kernel:&e1000:ð0&NIC&Link&is&Up&1000&Mbps&Full&Duplex,&Flow&Control:&RX&Dec&28&21:15:35&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&&&haproxy&1&进程&[root@one&~]#&pstree&init─┬─crond&&&&&&├─dbus-daemon&&&&&&├─events/0&&&&&&├─haproxy&&&&&&├─keepalived───keepalived&&&&&&├─keepalived.sh───haproxy_check.s───sleep&&停止&haproxy&进程&[root@one&~]#&killall&haproxy&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&#停止haproxy&进程&[root@one&~]#&tail&/usr/local/keepalived/fault.log&&&&&&&&&&&&&&&&&&&#fault&log&&_21:17:50&http://10.0.1.251:80&1&&&&&&fault&&_21:17:51&http://10.0.1.251:80&2&&&&&&fault&&_21:17:52&http://10.0.1.251:80&3&&&&&&fault&&_21:17:53&http://10.0.1.251:80&4&&&&&&fault&&_21:17:54&http://10.0.1.251:80&5&&&&&&fault&&_21:17:55&http://10.0.1.251:80&6&&&&&&fault&&_21:17:56&http://10.0.1.251:80&7&&&&&&fault&&_21:17:57&http://10.0.1.251:80&8&&&&&&fault&&haproxy&1&日志&[root@one&~]#&tail&/var/log/messages&&Dec&28&21:12:09&one&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&Dec&28&21:17:58&one&Keepalived:&Terminating&on&signal&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&Dec&28&21:17:58&one&Keepalived:&Stopping&Keepalived&v1.2.2&(12/25,2011)&&&&&&&&&&&&&&&&&&&#keepalived&被haproxy_check.sh&监控脚本关闭&Dec&28&21:17:58&one&Keepalived_vrrp:&Terminating&VRRP&child&process&on&signal&Dec&28&21:17:58&one&kernel:&ADDRCONF(NETDEV_UP):ð0:&link&is¬&ready&Dec&28&21:17:58&one&kernel:&e1000:ð0&NIC&Link&is&Up&1000&Mbps&Full&Duplex,&Flow&Control:&RX&Dec&28&21:17:58&one&kernel:&ADDRCONF(NETDEV_CHANGE):ð0:&link&becomes&ready&Dec&28&21:18:00&one&kernel:&ADDRCONF(NETDEV_UP):ð1:&link&is¬&ready&Dec&28&21:18:00&one&kernel:&e1000:ð1&NIC&Link&is&Up&1000&Mbps&Full&Duplex,&Flow&Control:&RX&Dec&28&21:18:00&one&kernel:&ADDRCONF(NETDEV_CHANGE):ð1:&link&becomes&ready&&&haproxy&2&日志&[root@four&cluster]#&tail&/var/log/messages&-f&Dec&28&20:53:08&four&Keepalived_vrrp:&Configuration&is&using&:&63577&Bytes&Dec&28&20:53:08&four&Keepalived_vrrp:&Using&LinkWatch&kernel&netlink&reflector...&Dec&28&20:53:08&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&Dec&28&20:53:12&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Transition&to&MASTER&STATE&Dec&28&20:53:13&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&Dec&28&21:12:07&four&Keepalived_vrrp:&Kernel&is&reporting:&interfaceð0&DOWN&Dec&28&21:12:07&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&FAULT&STATE&Dec&28&21:12:07&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Now&in&FAULT&state&Dec&28&21:15:34&four&kernel:&e1000:ð0&NIC&Link&is&Up&1000&Mbps&Full&Duplex,&Flow&Control:&RX&Dec&28&21:15:35&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&Dec&28&21:18:02&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Transition&to&MASTER&STATE&Dec&28&21:18:03&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&&&&&&&&&&#keepalived&状态为&master&&&&&&&&&&&&&[root@four&cluster]#&ip&add&&&&&&&&&&&&&&&&&1:&lo:&,UP,LOWER_UP&mtu&16436&qdisc&noqueue&&&&&&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:ð0:&,MULTICAST,UP,LOWER_UP&mtu&1500&qdisc&pfifo_fast&qlen&1000&&&&&link/ether&08:00:27:27:ef:a6&brd&ff:ff:ff:ff:ff:ff&&&&&inet&10.0.2.254/24&brd&10.0.2.255&scope&globalð0&&&&&inet6&fe80::a00:27ff:fe27:efa6/64&scope&link&&&&&&&&&valid_lft&forever&preferred_lft&forever&3:ð1:&,MULTICAST,UP,LOWER_UP&mtu&1500&qdisc&pfifo_fast&qlen&1000&&&&&link/ether&08:00:27:cd:f3:db&brd&ff:ff:ff:ff:ff:ff&&&&&inet&10.0.1.254/24&brd&10.0.1.255&scope&globalð1&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&& &&&&inet&10.0.1.250/24&scope&global&secondaryð1&
#virtual ip &&&&inet6&fe80::a00:27ff:fecd:f3db/64&scope&link&&&&&&&&&valid_lft&forever&preferred_lft&forever&4:&sit0:&&mtu&1480&qdisc&noop&&&&&&link/sit&0.0.0.0&brd&0.0.0.0&
脚本默认监控ð1&网卡，所以你需要根据自己的需要调整，如果你haproxy&virtual&ip&在eth0&网卡，那么你需要做如下更改:&6.1&使用&#&号注释掉&第14行&第16行，取消掉&第15行&第17行&#eth1_ip=$(/sbin/ifconfigð1&&|&sed&-n&'/inet&addr:/&s/inet&addr://pg'&|&awk&-F&&&&'{print&$1}')ð0_ip=$(/sbin/ifconfigð0&&|&sed&-n&'/inet&addr:/&s/inet&addr://pg'&|&awk&-F&&&&'{print&$1}')&#eth1_url=&http://${eth1_ip}:80&ð0_url=&http://${eth0_ip}:80&&&6.2&将第35行&$eth1_url,替换成&$eth0_url&if&curl&-IL&-A&&haproxy_check&&&$eth1_url&&&/dev/null&2&1&;then&6.3 将第48行 $eth1_url，替换成 $eth0_urlif&!&curl&-IL&-A&&haproxy_check&&&$eth1_url&&&/dev/null&2&1&;then
5.2&模拟网络故障&haproxy&切换测试&与&5.3&模拟haproxy&80端口故障&keepalived&切换测试&&五&haproxy&与&keepalived&整体测试，反过来测试也是没有问题的，不过要保证&haproxy.sh&keepalived.sh&这两个服务脚本已经启动了，可以使用pstree&查看&如：[root@four&cluster]#&pstree&init─┬─crond&&&&&&├─events/0&&&&&&├─events/1&&&&&&├─haproxy&&&&&&├─keepalived───keepalived&&&&&&├─keepalived.sh───haproxy_check.s───sleep&
&本文出自 “” 博客，转载请与作者联系！
了这篇文章
类别：┆阅读(0)┆评论(0)
19:49:13 13:38:08 11:23:29}
两大类热门资源免费畅读
续费一年阅读会员，立省24元！
Linux开源集群架构之keepalived应用详解
上传于||文档简介
&&k​e​e​p​a​l​i​v​e​d​安​装​
​
​应​用​
​
​架​设​高​可​用​M​Y​S​Q​L​
​
​整​合​L​V​S
阅读已结束，如果下载本文需要使用
想免费下载本文？
下载文档到电脑，查找使用更方便
还剩83页未读，继续阅读
你可能喜欢使用keepalived 实现MYSQL 高可用-mysql-电脑编程网使用keepalived 实现MYSQL 高可用作者： 和相关&&使用keepalived 实现MYSQL 高可用
keepalived
下载地址：http://www.keepalived.org/download.html
master:192.168.215.80
slave :192.168.215.28
VIP
:192.168.215.34
安装 keepalived
#=======================================================================
1:#wget http://www.keepalived.org/software/keepalived-1.1.20.tar.gz
2:#tar zxvf keepalived-1.1.20.tar.gz
3:#cd keepalived-1.1.20
4:#./configure
在这一步，可能看到的是如下：
#========================================================================
Keepalived configuration
------------------------
Keepalived version
: 1.1.20
Compiler
: gcc
Compiler flags
: -g -O2
Extra Lib
: -lpopt -lssl -lcrypto
Use IPVS Framework
: No
IPVS sync daemon support : No
Use VRRP Framework
: Yes
Use Debug flags
: No
#==========================================================================
出现3个NO，一个YES。解决的办法是编译时带上内核文件夹重新编译
# ./configure --with-kernel-dir=/usr/src/kernels/2.6.18-194.26.1.el5-x86_64/
Keepalived configuration
------------------------
Keepalived version
: 1.1.20
Compiler
: gcc
Compiler flags
: -g -O2
Extra Lib
: -lpopt -lssl -lcrypto
Use IPVS Framework
: Yes
IPVS sync daemon support : Yes
Use VRRP Framework
: Yes
Use Debug flags
: No
#==========================================================================
5:#make && make install
#=======================================================================
到这里，安装完成。下面写配置文件。
配置 keepalived(master,192.168.215.80)
#=======================================================================
#mkdir /etc/keepalived
#默认的配置文件获取的路径
#vi /etc/keepalived/keepalived.conf
#配置文件的名称（把下面复制即可）
! Configuration File for keepalived
global_defs {
#全局定义
&&&&¬ification_email {
#告警邮件地址组（故障发邮件在这里没有测试，我把它放到健康检查脚本里面了）
&&&&&
#发生故障，发告警的邮件地址，一般不用，可以随便写
&&&&&}
&&&&¬ification_email_from
#邮件发送地址
&&&&&smtp_server 127.0.0.1
#SMTPIP
&&&&&smtp_connect_timeout 30
#邮件连接超时最长时间
&&&&&router_id MySQL-ha
#路由标识。在全局定义里只有这个是必须的，其他都可以不要。
&&&&&}
vrrp_script check__alived {
#检查MYSQL状态的函数（注意这里的花括号和名称之间有空格）
&&&&script "/root/-moniter.sh"
#这里定义脚本，注意本行需要缩进，我原来没注意，在这上面整了很长时间。可以在这个脚本里检查MYSQL是否还活着，服务器的负载是否过高等，以及什么条件做什么操作。
&&&&&&&&interval 1
#运行上面检查脚本的时间间隔，单位为妙。keepalive的守护进程执行这个脚本
&&&&&&&&weight 5
#本脚本的运行权重，权重越高的，越优先检查（只有一个检查脚本，就无所谓了）
}
vrrp_instance VI_1 {
&&&&&state BACKUP
#两台配置此处均是BACKUP，一台宕机，另外一台把VIP抢过去。
&&&&&interface eth1
#在这里要注意，虚拟IP用的是哪个网卡进行通讯。我这里机器上的eth1是内网的，eth0是外网的。
&&&&&virtual_router_id 51
#虚拟路由ID。这个标识是同一个vrrp实例使用唯一的标识。即同一个vrrp_stance,MASTER和BACKUP的virtual_router_id是一致的，同时在整个vrrp内是唯一的。
&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&#keepalived通过发送和接受组广播包中的同一个virtual_router_id中的成员的存活，来确定对方的不可用，一旦检测到对方不可用，立即切换他的备份角色为主。
&&&&&priority 100
#权重，数值越大，权重越大，Master必须大于Slave。另一台改为90。
&&&&&advert_int 1
#Master和Slave负载均衡器之间同步检查的时间间隔，单位：秒
&&&&&#nopreempt
#抢占，在MASTER恢复后，把VIP抢过来。在SLAVE端也一样，如果MASTER出问题了，SLAVE就把VIP抢过来。
&&&&&authentication {
#Master和Slave之间认证的方式
&&&&&auth_type PASS
&&&&&auth_pass 1111
#认证的密码
&&&&&}
&&&&&track_script{
# 使用的VRRP脚本（检查健康脚步），上面是定义，这里是使用。
&&&&&&&&&check_mysql_alived
&&&&&}
&&&&&virtual_ipaddress {
#VIP的地址
&&&&&&&&&192.168.215.34
&&&
virtual_server 192.168.215.34 3306 {
&&&&&lb_algo wrr
&&&&&lb_kind DR
&&&&&persistence_timeout 60
#会话保持时间，单位为秒，当用户从远程帐号过行登陆网站时，有了这个会话保持功能，就能把用户的请求转发到同一个应用服务器
&&&&&protocol TCP
&&&&&real_server 192.168.215.80 3306 {
#真实IP（每台机器必须指定各自的真实IP）
&&&&&TCP_CHECK {
connect_timeout 10
#连接超时时间
nb_get_retry 3
delay_before_retry 3
#重连间隔时间
connect_port 3306
#健康检查端口
&&&&&&&&&&&&&}
&&&&&&&&&}
&&&&&}
#==========================================================================
MATER端配置到此结束。
编写健康服务检测的脚本
#==========================================================================
[root@l10ntermwiki bruce]# cat mysql-moniter.sh
#monitor mysql service status and server load
#!/bin/bash
function Send_Email()
#发邮件的函数。
{
ADMIN=""
MESSAGE_FILE="/tmp/monitor.log"
#邮件里面带上本机IP地址和时间，信息更明确。
&&&&&&&&echo "VIP has been changed AT `date`,IP:`cat /etc/hosts |grep -v "127.0.0.1"|awk '{print $1}'`,please check it. Message:$1" & $MESSAGE_FILE
&&&&&&&&/bin/mail -s "VIP has been changed" $ADMIN & $MESSAGE_FILE
rm -r $MESSAGE_FILE
}
CPU_NUMBER=`cat /proc/cpuinfo |grep processor |wc -l`
#获取CPU个数。
MAX_LOAD=`echo $CPU_NUMBER*2|bc`
#最大负载认为是CPU*2。设置成CPU个数更好。
if [ `service mysql status |awk '{print $2}'` = 'running' ] #检查mysql是否活着，这里用mysql作为服务名称，有些用mysqld作为服务名称。
then
&&&LOAD=`uptime`
&&&LOAD=${LOAD##* }
#
if [ `expr $LOAD \& $CPU_LOAD` = 1 ]
# another method,compare integer and decimal
&&&if [ `echo "$LOAD & $MAX_LOAD" | bc` = 1 ]
#如果负载过高。
&&&then
&&&&&&pkill keepalived
# if load is greater than max load then kill keepalived,take VIP point to another server.
&&&&&&Send_Email "Mysql is alived,Average load is busy,so kill keepalived"
&&&fi
else
#mysql服务挂了，VIP漂移走。
&&&pkill keepalived # if mysql service is not running ,then kill keepalived,take VIP point to another server.
&&&Send_Email "Mysql service is not alived,so kill keepalived."
fi
echo `date`&&/tmp/pk.log
#这里只是为了检查脚本是否被正常执行，实际使用删除掉。
#健康检查脚本完毕。
#chmod +x /home/bruce/MySQL.sh
#分配执行权限
#==========================================================================
启动keepalvice
#==========================================================================
[root@ldc2095 keepalived-1.1.20]# keepalived -D
#-D参数，会把日志记录到/var/log/message里面
[root@ldc2095 keepalived-1.1.20]# ps aux |grep keep
root
0:00 keepalived -D
root
0:00 keepalived -D
root
0:00 keepalived -D
root
0:00 grep keep
#===========================================================================
另外一个机器的配置和上面的一样。除了修改修改的几个地方。
健康检查脚本也一致。相关资料：｜｜｜｜｜｜｜使用keepalived 实现MYSQL 高可用来源网络，如有侵权请告知，即处理！编程Tags： & dngood 的BLOG
用户名：dngood
文章数：177
评论数：396
访问量：1754889
注册日期：
阅读量：24883
阅读量：235739
阅读量：980595
阅读量：140504
51CTO推荐博文
本文为& &haproxy + keepalived 一键安装包&& 的 第四篇
目的:帮助快速设置haproxy 与 keepalived 配置文档
(keepalived 为haproxy 做 HA 高可用）
第四篇 haproxy + keepalived &一键安装包&& 配置与测试HOWTO
一 haproxy.sh 与 keepalived.sh 服务脚本&
[root@four&cluster]#&ls&/usr/local/sbin/*.ah&haproxy.sh&&keepalived.sh&
二 设置自动启动 haproxy.sh 与 keepalived.sh
[root@four&cluster]#&cat&/etc/rc.local&&#!/bin/sh&#&#&This&script&will&be&executed&*after*&all&the&other&init&scripts.&#&You&can&put&your&own&initialization&stuff&in&here&if&you&don't&#&want&to&do&the&full&Sys&V&style&init&stuff.&&touch&/var/lock/subsys/local&ulimit&-SHn&65535&&/usr/local/sbin/haproxy.sh&start&/usr/local/sbin/keepalived.sh&start&
三 haproxy 相关配置
3.1&haroxy&程序目录&#ls&/usr/local/haproxy&doc&&haproxy.cfg&&html&&logs&&sbin&&share&&3.2&haproxy&配置文档&[root@four&cluster]#&file&/usr/local/haproxy/haproxy.cfg&&/usr/local/haproxy/haproxy.cfg:&UTF-8&Unicode&text&&3.3&编辑haproxy&配置文档&[root@four&cluster]#&vim&/usr/local/haproxy/haproxy.cfg&&第37&行&定义健康检查html文档，所以你需要在你的web服务上建立check.html文件用于haproxy&健康检查，也可以跟具需要自定义检查文档。&option&&httpchk&HEAD&/check.html&#用来做健康检查html文档&第38&39&行修10.0.1.252/253&替换成你的web&服务器ip地址&server&server1&10.0.1.252:80&cookie&server1&check&inter&2000&rise&3&&&&&fall&3&weight&3&server&server2&10.0.1.253:80&cookie&server2&check&inter&2000&rise&3&&&&&fall&3&maxconn&120&weight&3&&3.4&启动haproxy&输入如下命令&&[root@four&cluster]#&haproxy.sh&start&&3.5&检查是否有haproxy&进程&[root@four&cluster]#&ps&aux&|&grep&haproxy&|&grep&-v&'grep'&nobody&&&&&&0.2&&&&?&&&&&&&&Ss&&&20:35&&&0:00&/usr/local/haproxy/sbin/haproxy&-f&/usr/local/haproxy/haproxy.cfg&&3.6&检查haproxy&是否监听80&端口&[root@four&cluster]#&lsof&-i&:80&COMMAND&&PID&&&USER&&&FD&&&TYPE&DEVICE&SIZE&NODE&NAME&haproxy&4216&nobody&&&&4u&&IPv4&&&8229&&&&&&&TCP&*:http&(LISTEN)&&3.7&简单测试&haproxy&看看否返回&&200&OK&&[root@four&cluster]#&curl&-IL&http://10.0.1.254&(换成你的ip&)&HTTP/1.1&200&OK&Server:&nginx/1.0.5&Date:&Wed,&28&Dec&:30&GMT&Content-Type:&text/html&Content-Length:&159&Last-Modified:&Sat,&03&Dec&:10&GMT&Connection:&close&Accept-Ranges:&bytes&
四 keepalived 相关配置
4.1&keepalived&程序目录&[root@four&cluster]#&ls&/usr/local/keepalived&bin&&etc&&haproxy_check.sh&&keepalived.conf&&sbin&&share&&4.2&haproxy_check.sh&shell脚本监控程序,用于检查&80端口，来判断haproxy&是否可用&[root@four&cluster]#&file&/usr/local/keepalived/haproxy_check.sh&&/usr/local/keepalived/haproxy_check.sh:&Bourne-Again&shell&script&text&executable&&4.3&keepalived&配置文档&[root@four&cluster]#&file&/usr/local/keepalived/keepalived.conf&&/usr/local/keepalived/keepalived.conf:&ASCII&text&&4.4&编辑keepalived&配置文档&[root@four&cluster]#&vim&/usr/local/keepalived/keepalived.conf&&第16&行&interface&默认值ð1&,改成你的haproxy&外网卡(公网ip网卡)&interfaceð0&&第&34&行&默认值&10.0.1.250/24&devð1&，10.0.1.250&替换成你virtual&ip&地址,devð1&改成你的haproxy&外网卡(与第16行一致)&10.0.1.250/24&devð0&scope&global&&4.5&启动keepalived&输入如下命令&&[root@four&cluster]#&keepalived.sh&start&[root@four&cluster]#&4242&&&&&&&&&&&&&&&&#此处返回的是keepalived&进程id，敲下回车键即可&&4.6&检查是否有keepalived&进程&#&ps&aux&|&grep&keepalived&|&grep&-v&'grep'&&root&&&&&&&&0.0&&63844&&&464&pts/0&&&&S&&&&20:52&&&0:00&/bin/bash&/usr/local/sbin/keepalived.sh&start&root&&&&&&&&0.1&&35776&&&608&?&&&&&&&&Ss&&&20:52&&&0:00&/usr/local/keepalived/sbin/keepalived&-f&/usr/local/keepalived/keepalived.conf&root&&&&&&&&0.1&&3&?&&&&&&&&S&&&&20:52&&&0:00&/usr/local/keepalived/sbin/keepalived&-f&/usr/local/keepalived/keepalived.conf&root&&&&&&&&0.2&&6&pts/0&&&&S&&&&20:52&&&0:00&/bin/bash&/usr/local/keepalived/haproxy_check.sh&start&&4.7&查看keepalived&输出日志&[root@four&cluster]#&tail&/var/log/messages&&Dec&28&20:52:28&four&Keepalived_vrrp:&Registering&Kernel&netlink&reflector&Dec&28&20:52:28&four&Keepalived_vrrp:&Registering&Kernel&netlink&command&channel&Dec&28&20:52:28&four&Keepalived_vrrp:&Registering&gratutious&ARP&shared&channel&Dec&28&20:52:28&four&Keepalived:&Starting&VRRP&child&process,&pid=4241&Dec&28&20:53:08&four&Keepalived_vrrp:&Opening&file&'/usr/local/keepalived/keepalived.conf'.&&Dec&28&20:53:08&four&Keepalived_vrrp:&Configuration&is&using&:&63577&Bytes&Dec&28&20:53:08&four&Keepalived_vrrp:&Using&LinkWatch&kernel&netlink&reflector...&Dec&28&20:53:08&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&Dec&28&20:53:12&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Transition&to&MASTER&STATE&Dec&28&20:53:13&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&&&&&&&&&&&&&&#当前keepalived&状态为master&
五 haproxy 与 keepalived 整体测试
5.1&查看&haproxy&与&keepalived&进程树&[root@four&cluster]#&pstree&&init─┬─crond&&&&&&├─events/0&&&&&&├─events/1&&&&&&├─haproxy&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&#haproxy&进程&&&&&&├─keepalived───keepalived&&&&&&&&&&&&&&&&&&&&&&#keepalived&进程(2个)&&&&&&├─keepalived.sh───haproxy_check.s───sleep&&&&&&#keepalived.sh&调用&haproxy_check.sh&(监控脚本)&调用&sleep&（用于控检查间隔时间,2s）&&&5.2&模拟网络故障&haproxy&切换测试&&haproxy&1&日志&[root@one&~]#&tail&/var/log/messages&-f&Dec&28&21:09:43&one&kernel:&NFSD:&starting&90-second&grace&period&Dec&28&21:09:44&one&Keepalived:&Starting&Keepalived&v1.2.2&(12/25,2011)&&Dec&28&21:09:44&one&Keepalived_vrrp:&Registering&Kernel&netlink&reflector&Dec&28&21:09:44&one&Keepalived_vrrp:&Registering&Kernel&netlink&command&channel&Dec&28&21:09:44&one&Keepalived_vrrp:&Registering&gratutious&ARP&shared&channel&Dec&28&21:09:44&one&Keepalived:&Starting&VRRP&child&process,&pid=1993&Dec&28&21:10:24&one&Keepalived_vrrp:&Opening&file&'/usr/local/keepalived/keepalived.conf'.&&Dec&28&21:10:24&one&Keepalived_vrrp:&Configuration&is&using&:&63597&Bytes&Dec&28&21:10:24&one&Keepalived_vrrp:&Using&LinkWatch&kernel&netlink&reflector...&Dec&28&21:10:24&one&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&&haproxy&2&日志&[root@four&cluster]#&ifdownð0&[root@four&cluster]#&tail&/var/log/messages&&Dec&28&20:52:28&four&Keepalived:&Starting&VRRP&child&process,&pid=4241&Dec&28&20:53:08&four&Keepalived_vrrp:&Opening&file&'/usr/local/keepalived/keepalived.conf'.&&Dec&28&20:53:08&four&Keepalived_vrrp:&Configuration&is&using&:&63577&Bytes&Dec&28&20:53:08&four&Keepalived_vrrp:&Using&LinkWatch&kernel&netlink&reflector...&Dec&28&20:53:08&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&Dec&28&20:53:12&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Transition&to&MASTER&STATE&Dec&28&20:53:13&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&Dec&28&21:12:07&four&Keepalived_vrrp:&Kernel&is&reporting:&interfaceð0&DOWN&Dec&28&21:12:07&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&FAULT&STATE&Dec&28&21:12:07&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Now&in&FAULT&state&&&&&&&&&&&&&&&&&#keepalived&fault&状态&&haproxy&1&日志&[root@one&~]#&tail&/var/log/messages&-f&Dec&28&21:09:43&one&kernel:&NFSD:&starting&90-second&grace&period&Dec&28&21:09:44&one&Keepalived:&Starting&Keepalived&v1.2.2&(12/25,2011)&&Dec&28&21:09:44&one&Keepalived_vrrp:&Registering&Kernel&netlink&reflector&Dec&28&21:09:44&one&Keepalived_vrrp:&Registering&Kernel&netlink&command&channel&Dec&28&21:09:44&one&Keepalived_vrrp:&Registering&gratutious&ARP&shared&channel&Dec&28&21:09:44&one&Keepalived:&Starting&VRRP&child&process,&pid=1993&Dec&28&21:10:24&one&Keepalived_vrrp:&Opening&file&'/usr/local/keepalived/keepalived.conf'.&&Dec&28&21:10:24&one&Keepalived_vrrp:&Configuration&is&using&:&63597&Bytes&Dec&28&21:10:24&one&Keepalived_vrrp:&Using&LinkWatch&kernel&netlink&reflector...&Dec&28&21:10:24&one&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&Dec&28&21:12:08&one&Keepalived_vrrp:&VRRP_Instance(VI_1)&Transition&to&MASTER&STATE&Dec&28&21:12:09&one&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&&&&&&&&&&&&&#成功切换为&master&&&haproxy&1&检查vip&&[root@one&~]#&ip&add&1:&lo:&,UP,LOWER_UP&mtu&16436&qdisc&noqueue&&&&&&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:ð0:&,MULTICAST,UP,LOWER_UP&mtu&1500&qdisc&pfifo_fast&qlen&1000&&&&&link/ether&08:00:27:74:3c:05&brd&ff:ff:ff:ff:ff:ff&&&&&inet&10.0.2.251/24&brd&10.0.2.255&scope&globalð0&&&&&inet6&fe80::a00:27ff:fe74:3c05/64&scope&link&&&&&&&&&valid_lft&forever&preferred_lft&forever&3:ð1:&,MULTICAST,UP,LOWER_UP&mtu&1500&qdisc&pfifo_fast&qlen&1000&&&&&link/ether&08:00:27:3e:9a:2b&brd&ff:ff:ff:ff:ff:ff&&&&&inet&10.0.1.251/24&brd&10.0.1.255&scope&globalð1&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&& &&&&inet&10.0.1.250/24&scope&global&secondaryð1&
#virtual ip&&&&inet6&fe80::a00:27ff:fe3e:9a2b/64&scope&link&&&&&&&&&valid_lft&forever&preferred_lft&forever&4:&sit0:&&mtu&1480&qdisc&noop&&&&&&link/sit&0.0.0.0&brd&0.0.0.0&&5.3&模拟haproxy&80端口故障&keepalived&切换测试&&haproxy&2&恢复&[root@four&cluster]#&ifupð0&haproxy&2&日志&[root@four&cluster]#&tail&/var/log/messages&-f&Dec&28&20:53:08&four&Keepalived_vrrp:&Configuration&is&using&:&63577&Bytes&Dec&28&20:53:08&four&Keepalived_vrrp:&Using&LinkWatch&kernel&netlink&reflector...&Dec&28&20:53:08&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&Dec&28&20:53:12&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Transition&to&MASTER&STATE&Dec&28&20:53:13&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&Dec&28&21:12:07&four&Keepalived_vrrp:&Kernel&is&reporting:&interfaceð0&DOWN&Dec&28&21:12:07&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&FAULT&STATE&Dec&28&21:12:07&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Now&in&FAULT&state&Dec&28&21:15:34&four&kernel:&e1000:ð0&NIC&Link&is&Up&1000&Mbps&Full&Duplex,&Flow&Control:&RX&Dec&28&21:15:35&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&&&haproxy&1&进程&[root@one&~]#&pstree&init─┬─crond&&&&&&├─dbus-daemon&&&&&&├─events/0&&&&&&├─haproxy&&&&&&├─keepalived───keepalived&&&&&&├─keepalived.sh───haproxy_check.s───sleep&&停止&haproxy&进程&[root@one&~]#&killall&haproxy&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&#停止haproxy&进程&[root@one&~]#&tail&/usr/local/keepalived/fault.log&&&&&&&&&&&&&&&&&&&#fault&log&&_21:17:50&http://10.0.1.251:80&1&&&&&&fault&&_21:17:51&http://10.0.1.251:80&2&&&&&&fault&&_21:17:52&http://10.0.1.251:80&3&&&&&&fault&&_21:17:53&http://10.0.1.251:80&4&&&&&&fault&&_21:17:54&http://10.0.1.251:80&5&&&&&&fault&&_21:17:55&http://10.0.1.251:80&6&&&&&&fault&&_21:17:56&http://10.0.1.251:80&7&&&&&&fault&&_21:17:57&http://10.0.1.251:80&8&&&&&&fault&&haproxy&1&日志&[root@one&~]#&tail&/var/log/messages&&Dec&28&21:12:09&one&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&Dec&28&21:17:58&one&Keepalived:&Terminating&on&signal&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&Dec&28&21:17:58&one&Keepalived:&Stopping&Keepalived&v1.2.2&(12/25,2011)&&&&&&&&&&&&&&&&&&&#keepalived&被haproxy_check.sh&监控脚本关闭&Dec&28&21:17:58&one&Keepalived_vrrp:&Terminating&VRRP&child&process&on&signal&Dec&28&21:17:58&one&kernel:&ADDRCONF(NETDEV_UP):ð0:&link&is¬&ready&Dec&28&21:17:58&one&kernel:&e1000:ð0&NIC&Link&is&Up&1000&Mbps&Full&Duplex,&Flow&Control:&RX&Dec&28&21:17:58&one&kernel:&ADDRCONF(NETDEV_CHANGE):ð0:&link&becomes&ready&Dec&28&21:18:00&one&kernel:&ADDRCONF(NETDEV_UP):ð1:&link&is¬&ready&Dec&28&21:18:00&one&kernel:&e1000:ð1&NIC&Link&is&Up&1000&Mbps&Full&Duplex,&Flow&Control:&RX&Dec&28&21:18:00&one&kernel:&ADDRCONF(NETDEV_CHANGE):ð1:&link&becomes&ready&&&haproxy&2&日志&[root@four&cluster]#&tail&/var/log/messages&-f&Dec&28&20:53:08&four&Keepalived_vrrp:&Configuration&is&using&:&63577&Bytes&Dec&28&20:53:08&four&Keepalived_vrrp:&Using&LinkWatch&kernel&netlink&reflector...&Dec&28&20:53:08&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&Dec&28&20:53:12&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Transition&to&MASTER&STATE&Dec&28&20:53:13&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&Dec&28&21:12:07&four&Keepalived_vrrp:&Kernel&is&reporting:&interfaceð0&DOWN&Dec&28&21:12:07&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&FAULT&STATE&Dec&28&21:12:07&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Now&in&FAULT&state&Dec&28&21:15:34&four&kernel:&e1000:ð0&NIC&Link&is&Up&1000&Mbps&Full&Duplex,&Flow&Control:&RX&Dec&28&21:15:35&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&BACKUP&STATE&Dec&28&21:18:02&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Transition&to&MASTER&STATE&Dec&28&21:18:03&four&Keepalived_vrrp:&VRRP_Instance(VI_1)&Entering&MASTER&STATE&&&&&&&&&&#keepalived&状态为&master&&&&&&&&&&&&&[root@four&cluster]#&ip&add&&&&&&&&&&&&&&&&&1:&lo:&,UP,LOWER_UP&mtu&16436&qdisc&noqueue&&&&&&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:ð0:&,MULTICAST,UP,LOWER_UP&mtu&1500&qdisc&pfifo_fast&qlen&1000&&&&&link/ether&08:00:27:27:ef:a6&brd&ff:ff:ff:ff:ff:ff&&&&&inet&10.0.2.254/24&brd&10.0.2.255&scope&globalð0&&&&&inet6&fe80::a00:27ff:fe27:efa6/64&scope&link&&&&&&&&&valid_lft&forever&preferred_lft&forever&3:ð1:&,MULTICAST,UP,LOWER_UP&mtu&1500&qdisc&pfifo_fast&qlen&1000&&&&&link/ether&08:00:27:cd:f3:db&brd&ff:ff:ff:ff:ff:ff&&&&&inet&10.0.1.254/24&brd&10.0.1.255&scope&globalð1&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&&& &&&&inet&10.0.1.250/24&scope&global&secondaryð1&
#virtual ip &&&&inet6&fe80::a00:27ff:fecd:f3db/64&scope&link&&&&&&&&&valid_lft&forever&preferred_lft&forever&4:&sit0:&&mtu&1480&qdisc&noop&&&&&&link/sit&0.0.0.0&brd&0.0.0.0&
脚本默认监控ð1&网卡，所以你需要根据自己的需要调整，如果你haproxy&virtual&ip&在eth0&网卡，那么你需要做如下更改:&6.1&使用&#&号注释掉&第14行&第16行，取消掉&第15行&第17行&#eth1_ip=$(/sbin/ifconfigð1&&|&sed&-n&'/inet&addr:/&s/inet&addr://pg'&|&awk&-F&&&&'{print&$1}')ð0_ip=$(/sbin/ifconfigð0&&|&sed&-n&'/inet&addr:/&s/inet&addr://pg'&|&awk&-F&&&&'{print&$1}')&#eth1_url=&http://${eth1_ip}:80&ð0_url=&http://${eth0_ip}:80&&&6.2&将第35行&$eth1_url,替换成&$eth0_url&if&curl&-IL&-A&&haproxy_check&&&$eth1_url&&&/dev/null&2&1&;then&6.3 将第48行 $eth1_url，替换成 $eth0_urlif&!&curl&-IL&-A&&haproxy_check&&&$eth1_url&&&/dev/null&2&1&;then
5.2&模拟网络故障&haproxy&切换测试&与&5.3&模拟haproxy&80端口故障&keepalived&切换测试&&五&haproxy&与&keepalived&整体测试，反过来测试也是没有问题的，不过要保证&haproxy.sh&keepalived.sh&这两个服务脚本已经启动了，可以使用pstree&查看&如：[root@four&cluster]#&pstree&init─┬─crond&&&&&&├─events/0&&&&&&├─events/1&&&&&&├─haproxy&&&&&&├─keepalived───keepalived&&&&&&├─keepalived.sh───haproxy_check.s───sleep&
&本文出自 “” 博客，转载请与作者联系！
了这篇文章
类别：┆阅读(0)┆评论(0)
19:49:13 13:38:08 11:23:29}
我要回帖
更多关于 keepalived 使用 的文章
更多推荐
- ·惠来哪家仿琉璃瓦屋顶效果图好？？
- ·自建房盖房子屋顶封顶用什么材料除了水泥板还有什么办法比较好？？
- ·怎样控制展馆布展费用组成的施工成本？
- ·在穆斯桥市留学的氛围怎么样？
- ·MD如何可以令md布料移动边精确距离与模特一起旋转？
- ·九阴真经华山派入门君子堂二内高阶华山派可以学吗
- ·白沟全城乐购帮帮主是谁
- ·天天天天抢红包是真的吗吗
- ·意迟迟的掌珠txt下载还更新吗
- ·汗拳拳皇13怎么解锁人物解
- ·震旦adc218驱动安装打印机驱动怎么安装
- ·想怎么能看业主在找保姆近清远聚龙湾温泉别墅别墅找保姆的工作怎么才能找到
- ·三星2951d驱动ml2951dn打印机网络ip设置
- ·现在买ipad mini2值得买吗合适吗
- ·千元手千元机什么牌子好好些什么型号好
- ·uvc clarifying lotion什么意思
- ·一个手机号可以注册几个小米账号更换手机号
- ·下载金立e8评测视频手机广告视频
- ·小米4移动高配版和红米note3 标配 高配标配哪个好用？建议买哪个？
- ·keepalived root权限必须用root用吗
- ·这样的户型图尺寸包括墙厚吗好吗？是手机户型图尺寸包括墙厚吗吗？
- ·热热鲁热热鲁360免费wifi手机版版
- ·手机寻回功能，微信密码忘了如何找回，怎样找回密码。
- ·属虎人手机尾号185aa8879aa6089好吗
- ·苹果5sa1530at/a
- ·怎样审请不用手机号码申请qqqq号码
- ·易奕的易小川手机铃声叫什么声
- ·如何怎样防止qq被盗点击微友链接时银行卡被盗刷
- ·有谁用过江南星星商用冰柜柜价绍下。
- ·win7 32 4g内存怎么显示4g内存
- ·用自己的微信绑定别人手机号被别人绑定在qq属于敲诈吗
- ·汽车音响三分频，50赫兹和60赫兹的区别HSK163和DLS的R6.3该怎么选择？那个效果好点，因为预算有限，比较纠
- ·如何修改密保问题保
- ·我想买一个日本买单反相机便宜吗点的
- ·新买苹果6s重新设置id手机有膜吗，自己还用去重新贴吗
