实验环境:
LVS和keepalived介绍:
1.keepalived的设计目标是构建高可用的LVS负载均衡群集,可以调用ipvsadm工具来创建虚拟服务器,管理服务器池,而不仅仅用来做双机热备。
使用keepalived构建LVS群集更加简便易用,主要优势体现在:
对LVS负载调度器实现热备切换,提高可用性;
对服务器池中的节点进行健康检查,自动移除失效节点,恢复后再重新加入。
2.在基于LVS+Keepalived实现的LVS群集结构中,至少包括两台热备的负载调度器,两台以上的节点服务器,本例将以DR模式的LVS群集为基础,增加一台从负载调度器,使用Keepalived来实现主、从调度器的热备,从而构建建有负载均衡、高可用两种能力的LVS网站群集平台。
3.通过LVS+Keepalived构建的LVS集群,LVS负载均衡用户请求到后端tomcat服务器,Keepalived的作用是检测web服务器的状态,如果有一台web服务器死机,或工作出现故障,Keepalived将检测到,并将有故障的web服务器从系统中剔除,当web服务器工作正常后Keepalived自动将web服务器加入到服务器群中,这些工作全部自动完成,不需要人工干涉,需要人工做的只是修复故障的web服务器。
Keepalived简介
什么是Keepalived:keepalived可以实现服务的高可用或热备,用来防止单点故障的问题;而Keepalived的核心VRRP协议,VRRP协议主要实现了在路由器或三层交换机处的冗余;Keepalived就是使用VRRP协议来实现高可用的;下面一起来看一下Keepalived的原理图:
Keepalived启动后会有三个进程:
父进程:内存管理,子进程管理
子进程:VRRP子进程
子进程:healthchecker子进程
由上图可知:
两个子进程都被系统WatchDog看管,两个子进程各自复杂自己的事,healthchecker子进程复杂检查各自服务器的健康程度,例如HTTP,LVS等,如果healthchecker子进程检查到MASTER上服务不可用了,就会通知本机上的VRRP子进程,让它删除通告,并且去掉虚拟IP,转换为BACKUP状态;
实验架构图:
实验过程:
HA高可用集群构建前提:
1.proxy和proxy2节点时间必须同步;
建议使用ntp协议进行;
参考博客:http://sohudrgon.blog.51cto.com/3088108/1598314
2、节点之间必须要通过主机名互相通信;
建议使用hosts文件;
通信中使用的名字必须与其节点为上“uname -n”命令展示出的名字保持一致;
[root@proxy ~]# cat /etc/hosts 127.0.0.1 localhost localhost.localdomain localhost4 localhost4.localdomain4 ::1 localhost localhost.localdomain localhost6 localhost6.localdomain6 172.16.0.1 server.magelinux.com server 172.16.31.52 proxy.stu31.com proxy 172.16.31.53 proxy2.stu31.com proxy2 172.16.31.50 tom1.stu31.com tom1 172.16.31.51 tom2.stu31.com tom2
3、节点之间彼此root用户能基于ssh密钥方式进行通信;
节点proxy: # ssh-keygen -t rsa -P "" # ssh-copy-id -i .ssh/id_rsa.pub proxy2 节点proxy2: # ssh-keygen -t rsa -P "" # ssh-copy-id -i .ssh/id_rsa.pub proxy
测试ssh无密钥通信:
[root@proxy ~]# date ; ssh proxy2 date Sat Jan 17 15:29:26 CST 2015 Sat Jan 17 15:29:26 CST 2015
一.配置安装LVS+keepalived
1.在Master与Backup服务器上分别安装Ipvsadm、Keepalived软件包、这里使用的是yum安装方式
# yum install -y keepalived ipvsadm
注释:这里安装Ipvsadm工具主要是为了查看lvs规则使用,不安装ipvsadm工具也能成功配置规则,但不方便;
2.修改keepalived配置文件
keepalived的文件路径/etc/keepalived/keepalived.conf
修改Master节点proxy的主配置文件
[root@proxy ~]# vim /etc/keepalived/keepalived.conf ! Configuration File for keepalived global_defs { notification_email { #设置报警通知邮件地址,可以设置多个 root@localhost } notification_email_from [email protected] #设置邮件的发送地址 smtp_server 127.0.0.1 #设置smtp server的地址,该地址必须是存在的 smtp_connect_timeout 30 #设置连接smtp server的超时时间 router_id LVS_DEVEL #运行Keepalived服务器的标识,发邮件时显示在邮件标题中的信息 } vrrp_instance VI_1 { #定义VRRP实例,实例名自定义 state MASTER #指定Keepalived的角色,MASTER为主服务器,BACKUP为备用服务器 interface eth0 #指定HA监测的接口 virtual_router_id 51 #虚拟路由标识,这个标识是一个数字(1-255),在一个VRRP实例中主备服务器ID必须一样 priority 100 #优先级,数字越大优先级越高,在一个实例中主服务器优先级要高于备服务器 advert_int 1 #设置主备之间同步检查的时间间隔单位秒 authentication { #设置验证类型和密码 auth_type PASS #验证类型有两种{PASS|HA} auth_pass oracle #设置验证密码,在一个实例中主备密码保持一样 } virtual_ipaddress { #定义虚拟IP地址,可以有多个,每行一个 172.16.31.188 } } virtual_server 172.16.31.188 { #设置虚拟服务器,需要指定虚拟IP与服务端口,用空格分隔 delay_loop 6 #设置健康状态检查时间,单位为秒 lb_algo rr #设置负载高度算法,rr为轮询 lb_kind DR #设置LVS实现负载均衡的机制,可以为{NAT|TUN|DR}三种 nat_mask 255.255.0.0 #设置掩码 persistence_timeout 50 #会话保持时间,单位为秒;这个选项对于动态网页是非常有用的,为集群系统中session共享提供了一个很好的解决方案 protocol TCP #指定转发协议类型可以设置{TCP|UDP}两种 real_server 172.16.31.50 80 { #后端服务器节点,需要指定Real_server的IP与端口,用空格分隔 weight 1 #配置服务节点的权重,数字越大,权重越高 HTTP_GET { #设置检测Realserver的方式为http协议 url { path / status_code 200 #设定返回状态码为200表示Realserver是存活的 } connect_timeout 3 #设置响应超时时间 nb_get_retry 3 #设置超时重试次数 delay_before_retry 3 #设置超时后重试间隔 } } real_server 172.16.31.51 80 { weight 1 HTTP_GET { url { path / status_code 200 } connect_timeout 3 nb_get_retry 3 delay_before_retry 3 } } }
3.将Master服务器proxy节点上的主配置文件拷贝到Backup服务器proxy2节点稍作修改
# scp /etc/keepalived/keepalived.conf proxy2:/etc/keepalived/ ######修改如下两项 [root@proxy2 ~]# vim /etc/keepalived/keepalived.conf state BACKUP priority 99
4.启动两台服务器上的Keepalived服务并设置为开机自启动
# chkconfig --add keepalived ; ssh proxy2 "chkconfig --add keepalived" # chkconfig keepalived on ; ssh proxy2 "chkconfig keepalived on" # chkconfig --list keepalived; ssh proxy2 "chkconfig --list keepalived" keepalived 0:off 1:off 2:on 3:on 4:on 5:on 6:off keepalived 0:off 1:off 2:on 3:on 4:on 5:on 6:off
启动服务:
# service keepalived start; ssh proxy2 "service keepalived start"
5.观察各节点的服务启动后的日志:
MASTER节点的日志;
[root@proxy keepalived]# tail -f /var/log/messages Jan 17 15:57:42 proxy Keepalived[15019]: Starting Keepalived v1.2.13 (10/15,2014) Jan 17 15:57:42 proxy Keepalived[15021]: Starting Healthcheck child process, pid=15024 Jan 17 15:57:42 proxy Keepalived[15021]: Starting VRRP child process, pid=15025 Jan 17 15:57:42 proxy Keepalived_vrrp[15025]: Netlink reflector reports IP 172.16.31.52 added Jan 17 15:57:42 proxy Keepalived_vrrp[15025]: Netlink reflector reports IP fe80::a00:27ff:fe3b:2360 added Jan 17 15:57:42 proxy Keepalived_vrrp[15025]: Registering Kernel netlink reflector Jan 17 15:57:42 proxy Keepalived_vrrp[15025]: Registering Kernel netlink command channel Jan 17 15:57:42 proxy Keepalived_vrrp[15025]: Registering gratuitous ARP shared channel Jan 17 15:57:42 proxy Keepalived_healthcheckers[15024]: Netlink reflector reports IP 172.16.31.52 added Jan 17 15:57:42 proxy Keepalived_healthcheckers[15024]: Netlink reflector reports IP fe80::a00:27ff:fe3b:2360 added Jan 17 15:57:42 proxy Keepalived_healthcheckers[15024]: Registering Kernel netlink reflector Jan 17 15:57:42 proxy Keepalived_healthcheckers[15024]: Registering Kernel netlink command channel Jan 17 15:57:42 proxy Keepalived_vrrp[15025]: Opening file '/etc/keepalived/keepalived.conf'. Jan 17 15:57:42 proxy Keepalived_vrrp[15025]: Configuration is using : 63202 Bytes Jan 17 15:57:42 proxy Keepalived_vrrp[15025]: Using LinkWatch kernel netlink reflector... Jan 17 15:57:42 proxy Keepalived_healthcheckers[15024]: Opening file '/etc/keepalived/keepalived.conf'. Jan 17 15:57:42 proxy Keepalived_healthcheckers[15024]: Configuration is using : 16989 Bytes Jan 17 15:57:42 proxy Keepalived_vrrp[15025]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(10,11)] Jan 17 15:57:42 proxy Keepalived_healthcheckers[15024]: Using LinkWatch kernel netlink reflector... Jan 17 15:57:42 proxy Keepalived_healthcheckers[15024]: Activating healthchecker for service [172.16.31.50]:8080 Jan 17 15:57:42 proxy Keepalived_healthcheckers[15024]: Activating healthchecker for service [172.16.31.51]:8080 Jan 17 15:57:43 proxy Keepalived_vrrp[15025]: VRRP_Instance(VI_1) Transition to MASTER STATE Jan 17 15:57:44 proxy Keepalived_vrrp[15025]: VRRP_Instance(VI_1) Entering MASTER STATE Jan 17 15:57:44 proxy Keepalived_vrrp[15025]: VRRP_Instance(VI_1) setting protocol VIPs. Jan 17 15:57:44 proxy Keepalived_healthcheckers[15024]: Netlink reflector reports IP 172.16.31.188 added Jan 17 15:57:44 proxy Keepalived_vrrp[15025]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 172.16.31.188
BACKUP节点的日志:
[root@proxy2 keepalived]# tail -f /var/log/messages Jan 17 15:57:42 proxy2 Keepalived[15005]: Starting Keepalived v1.2.13 (10/15,2014) Jan 17 15:57:42 proxy2 Keepalived[15007]: Starting Healthcheck child process, pid=15009 Jan 17 15:57:42 proxy2 Keepalived[15007]: Starting VRRP child process, pid=15010 Jan 17 15:57:42 proxy2 Keepalived_vrrp[15010]: Netlink reflector reports IP 172.16.31.53 added Jan 17 15:57:42 proxy2 Keepalived_healthcheckers[15009]: Netlink reflector reports IP 172.16.31.53 added Jan 17 15:57:42 proxy2 Keepalived_vrrp[15010]: Netlink reflector reports IP fe80::a00:27ff:fe6e:bd28 added Jan 17 15:57:42 proxy2 Keepalived_vrrp[15010]: Registering Kernel netlink reflector Jan 17 15:57:42 proxy2 Keepalived_vrrp[15010]: Registering Kernel netlink command channel Jan 17 15:57:42 proxy2 Keepalived_vrrp[15010]: Registering gratuitous ARP shared channel Jan 17 15:57:42 proxy2 Keepalived_healthcheckers[15009]: Netlink reflector reports IP fe80::a00:27ff:fe6e:bd28 added Jan 17 15:57:42 proxy2 Keepalived_healthcheckers[15009]: Registering Kernel netlink reflector Jan 17 15:57:42 proxy2 Keepalived_healthcheckers[15009]: Registering Kernel netlink command channel Jan 17 15:57:42 proxy2 Keepalived_vrrp[15010]: Opening file '/etc/keepalived/keepalived.conf'. Jan 17 15:57:42 proxy2 Keepalived_healthcheckers[15009]: Opening file '/etc/keepalived/keepalived.conf'. Jan 17 15:57:42 proxy2 Keepalived_healthcheckers[15009]: Configuration is using : 16991 Bytes Jan 17 15:57:42 proxy2 Keepalived_vrrp[15010]: Configuration is using : 63204 Bytes Jan 17 15:57:42 proxy2 Keepalived_vrrp[15010]: Using LinkWatch kernel netlink reflector... Jan 17 15:57:42 proxy2 Keepalived_vrrp[15010]: VRRP_Instance(VI_1) Entering BACKUP STATE Jan 17 15:57:42 proxy2 Keepalived_vrrp[15010]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(10,11)] Jan 17 15:57:42 proxy2 Keepalived_healthcheckers[15009]: Using LinkWatch kernel netlink reflector... Jan 17 15:57:42 proxy2 Keepalived_healthcheckers[15009]: Activating healthchecker for service [172.16.31.50]:8080 Jan 17 15:57:42 proxy2 Keepalived_healthcheckers[15009]: Activating healthchecker for service [172.16.31.51]:8080
6.开启Master节点与Backup节点服务器的路由转发功能
#sed -i 's/net.ipv4.ip_forward = 0/net.ipv4.ip_forward = 1/g' /etc/sysctl.conf
二.安装配置后端的tomcat服务器节点
1.配置后端tomcat服务器节点;
具体配置参考我的前篇博客,都是连续性的实验;
配置完成后,访问测试如图:
tom1节点:
tom2节点:
实现了负载均衡访问;
2.为两台Realserver提供Sysv格式的脚本来自动修改内核参数与虚拟IP并运行脚本
#vim rs.sh . /etc/rc.d/init.d/functions VIP=172.16.31.188 host=`/bin/hostname` case "$1" in start) # Start LVS-DR real server on this machine. /sbin/ifconfig lo down /sbin/ifconfig lo up echo 1 > /proc/sys/net/ipv4/conf/lo/arp_ignore echo 2 > /proc/sys/net/ipv4/conf/lo/arp_announce echo 1 > /proc/sys/net/ipv4/conf/all/arp_ignore echo 2 > /proc/sys/net/ipv4/conf/all/arp_announce /sbin/ifconfig lo:0 $VIP broadcast $VIP netmask 255.255.255.255 up /sbin/route add -host $VIP dev lo:0 ;; stop) # Stop LVS-DR real server loopback device(s). /sbin/ifconfig lo:0 down echo 0 > /proc/sys/net/ipv4/conf/lo/arp_ignore echo 0 > /proc/sys/net/ipv4/conf/lo/arp_announce echo 0 > /proc/sys/net/ipv4/conf/all/arp_ignore echo 0 > /proc/sys/net/ipv4/conf/all/arp_announce ;; status) # Status of LVS-DR real server. islothere=`/sbin/ifconfig lo:0 | grep $VIP` isrothere=`netstat -rn | grep "lo:0" | grep $VIP` if [ ! "$islothere" -o ! "isrothere" ];then # Either the route or the lo:0 device # not found. echo "LVS-DR real server Stopped." else echo "LVS-DR real server Running." fi ;; *) # Invalid entry. echo "$0: Usage: $0 {start|status|stop}" exit 1 ;; esac
注释:脚本中的VIP定义的是虚拟IP地址
3.脚本两个节点都存放一份;然后执行此脚本:
#sh rs.sh start
4.查看设置后的状态:
# sh rs.sh status LVS-DR real server Running.
查看节点的VIP地址设置完成与否:
tom1节点;
[root@tom1 ~]# ip addr show lo 1: lo: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 inet 172.16.31.188/32 brd 172.16.31.188 scope global lo:0 inet6 ::1/128 scope host valid_lft forever preferred_lft forever
tom2节点:
[root@tom2 ~]# ip addr show lo 1: lo: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 inet 172.16.31.188/32 brd 172.16.31.188 scope global lo:0 inet6 ::1/128 scope host valid_lft forever preferred_lft forever
5.查看MASTER节点proxy的ipvs规则:
[root@proxy keepalived]# ipvsadm -L -n IP Virtual Server version 1.2.1 (size=4096) Prot LocalAddress:Port Scheduler Flags -> RemoteAddress:Port Forward Weight ActiveConn InActConn TCP 172.16.31.188:0 rr persistent 50 -> 172.16.31.50:8080 Route 1 0 0 -> 172.16.31.51:8080 Route 1 0 0
从规则中可以看出虚拟IP与Port及调度算法为rr;其中有两个Realserver
三.访问测试及故障模拟ces
1.访问测试:
由上可知,使用的是rr调度算法,在访问测试时可能需要多访问几次或换个浏览器来测试访问
2.模拟Master服务器出现故障,将Master主机上的Keepalived服务停止,查看Backup服务器是否接管所有服务
[root@proxy keepalived]# service keepalived stop Stopping keepalived: [ OK ]
3.观察节点切换后的系统日志:
MASTER节点的日志:
[root@proxy keepalived]# tail -f /var/log/messages Jan 17 16:20:40 proxy Keepalived[15021]: Stopping Keepalived v1.2.13 (10/15,2014) Jan 17 16:20:40 proxy Keepalived_vrrp[15025]: VRRP_Instance(VI_1) sending 0 priority Jan 17 16:20:40 proxy Keepalived_vrrp[15025]: VRRP_Instance(VI_1) removing protocol VIPs. Jan 17 16:20:40 proxy Keepalived_healthcheckers[15024]: Netlink reflector reports IP 172.16.31.188 removed Jan 17 16:20:40 proxy Keepalived_healthcheckers[15024]: Removing service [172.16.31.50]:8080 from VS [172.16.31.188]:0 Jan 17 16:20:40 proxy Keepalived_healthcheckers[15024]: Removing service [172.16.31.51]:8080 from VS [172.16.31.188]:0 Jan 17 16:20:40 proxy kernel: IPVS: __ip_vs_del_service: enter
BACKUP节点的日志:
[root@proxy2 keepalived]# tail -f /var/log/messages Jan 17 16:20:41 proxy2 Keepalived_vrrp[15010]: VRRP_Instance(VI_1) Transition to MASTER STATE Jan 17 16:20:42 proxy2 Keepalived_vrrp[15010]: VRRP_Instance(VI_1) Entering MASTER STATE Jan 17 16:20:42 proxy2 Keepalived_vrrp[15010]: VRRP_Instance(VI_1) setting protocol VIPs. Jan 17 16:20:42 proxy2 Keepalived_healthcheckers[15009]: Netlink reflector reports IP 172.16.31.188 added Jan 17 16:20:42 proxy2 Keepalived_vrrp[15010]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 172.16.31.188 Jan 17 16:20:47 proxy2 Keepalived_vrrp[15010]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 172.16.31.188
4.查看各节点的VIP地址状态及IPVS规则:
原MASTER节点:
[root@proxy keepalived]# ip addr show 1: lo: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: mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 08:00:27:3b:23:60 brd ff:ff:ff:ff:ff:ff inet 172.16.31.52/16 brd 172.16.255.255 scope global eth0 inet6 fe80::a00:27ff:fe3b:2360/64 scope link valid_lft forever preferred_lft forever
[root@proxy keepalived]# ipvsadm -L -n IP Virtual Server version 1.2.1 (size=4096) Prot LocalAddress:Port Scheduler Flags -> RemoteAddress:Port Forward Weight ActiveConn InActConn
由上可见Master服务器上已删除虚拟IP与LVS规则
BACKUP节点:
[root@proxy2 keepalived]# ip addr show 1: lo: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: mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 08:00:27:6e:bd:28 brd ff:ff:ff:ff:ff:ff inet 172.16.31.53/16 brd 172.16.255.255 scope global eth0 inet 172.16.31.188/32 scope global eth0 inet6 fe80::a00:27ff:fe6e:bd28/64 scope link valid_lft forever preferred_lft forever
由上可见,虚拟IP地址已成功在Backup服务器启动
[root@proxy2 keepalived]# ipvsadm -L -n IP Virtual Server version 1.2.1 (size=4096) Prot LocalAddress:Port Scheduler Flags -> RemoteAddress:Port Forward Weight ActiveConn InActConn TCP 172.16.31.188:0 rr persistent 50 -> 172.16.31.50:8080 Route 1 0 0 -> 172.16.31.51:8080 Route 1 0 0
LVS的规则也已成功配置在Backup服务器上面
5.再次访问测试服务器是否正常提供服务
6.假如Master服务器修复好已重新上线,则虚拟IP地址与LVS规则会重新配置到Master服务器上而在Backup服务器上删除
[root@proxy keepalived]# service keepalived start Starting keepalived: [ OK ]
观察启动日志:
MASTER节点的日志:
[root@proxy keepalived]# tail -f /var/log/messages Jan 17 16:34:50 proxy Keepalived_vrrp[15366]: VRRP_Instance(VI_1) Transition to MASTER STATE Jan 17 16:34:50 proxy Keepalived_vrrp[15366]: VRRP_Instance(VI_1) Received lower prio advert, forcing new election Jan 17 16:34:51 proxy Keepalived_vrrp[15366]: VRRP_Instance(VI_1) Entering MASTER STATE Jan 17 16:34:51 proxy Keepalived_vrrp[15366]: VRRP_Instance(VI_1) setting protocol VIPs. Jan 17 16:34:51 proxy Keepalived_vrrp[15366]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 172.16.31.188 Jan 17 16:34:51 proxy Keepalived_healthcheckers[15365]: Netlink reflector reports IP 172.16.31.188 added Jan 17 16:34:56 proxy Keepalived_vrrp[15366]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth0 for 172.16.31.188
自动进入了MASTER状态;
BACKUP节点的日志:
[root@proxy2 keepalived]# tail -f /var/log/messages Jan 17 16:34:50 proxy2 Keepalived_vrrp[15258]: VRRP_Instance(VI_1) Received higher prio advert Jan 17 16:34:50 proxy2 Keepalived_vrrp[15258]: VRRP_Instance(VI_1) Entering BACKUP STATE Jan 17 16:34:50 proxy2 Keepalived_vrrp[15258]: VRRP_Instance(VI_1) removing protocol VIPs. Jan 17 16:34:50 proxy2 Keepalived_healthcheckers[15257]: Netlink reflector reports IP 172.16.31.188 removed
自动进入BACKUP状态,移除VIP;
查看各节点的VIP 和IPVS规则:
MASTER节点的的VIP生成成功,IPVS规则生成:
[root@proxy keepalived]# ip addr show 1: lo: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: mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 08:00:27:3b:23:60 brd ff:ff:ff:ff:ff:ff inet 172.16.31.52/16 brd 172.16.255.255 scope global eth0 inet 172.16.31.188/32 scope global eth0 inet6 fe80::a00:27ff:fe3b:2360/64 scope link valid_lft forever preferred_lft forever [root@proxy keepalived]# ipvsadm -L -n IP Virtual Server version 1.2.1 (size=4096) Prot LocalAddress:Port Scheduler Flags -> RemoteAddress:Port Forward Weight ActiveConn InActConn TCP 172.16.31.188:0 rr persistent 50 -> 172.16.31.50:8080 Route 1 0 0 -> 172.16.31.51:8080 Route 1 0 0
BACKUP节点自动降级,VIP自动移除,但是ipvs规则还是存在的,这对我们是没有影响的,没有了IP地址只有规则也是不生效的
[root@proxy2 keepalived]# ip addr show 1: lo: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: mtu 1500 qdisc pfifo_fast state UP qlen 1000 link/ether 08:00:27:6e:bd:28 brd ff:ff:ff:ff:ff:ff inet 172.16.31.53/16 brd 172.16.255.255 scope global eth0 inet6 fe80::a00:27ff:fe6e:bd28/64 scope link valid_lft forever preferred_lft forever [root@proxy2 keepalived]# ipvsadm -L -n IP Virtual Server version 1.2.1 (size=4096) Prot LocalAddress:Port Scheduler Flags -> RemoteAddress:Port Forward Weight ActiveConn InActConn TCP 172.16.31.188:0 rr persistent 50 -> 172.16.31.50:8080 Route 1 0 0 -> 172.16.31.51:8080 Route 1 0 0
7.如果后端Realserver出现故障,则LVS规则会清除相应Realserver的规则
我们将后端的tom1节点的tomcat服务停止:
[root@tom1 ~]# catalina.sh stop Using CATALINA_BASE: /usr/local/tomcat Using CATALINA_HOME: /usr/local/tomcat Using CATALINA_TMPDIR: /usr/local/tomcat/temp Using JRE_HOME: /usr/java/latest Using CLASSPATH: /usr/local/tomcat/bin/bootstrap.jar:/usr/local/tomcat/bin/tomcat-juli.jar [root@tom1 ~]#
观察LVS节点的启动日志:
MASTER节点:
出现连接错误,发送了邮件:
[root@proxy keepalived]# tail -f /var/log/messages Jan 17 16:40:39 proxy Keepalived_healthcheckers[15365]: Error connecting server [172.16.31.50]:8080. Jan 17 16:40:39 proxy Keepalived_healthcheckers[15365]: Removing service [172.16.31.50]:8080 from VS [172.16.31.188]:0 Jan 17 16:40:39 proxy Keepalived_healthcheckers[15365]: Remote SMTP server [127.0.0.1]:25 connected. Jan 17 16:40:39 proxy Keepalived_healthcheckers[15365]: SMTP alert successfully sent.
BACKUP节点:
出现连接错误,发送了邮件:
[root@proxy2 keepalived]# tail -f /var/log/messages Jan 17 16:40:42 proxy2 Keepalived_healthcheckers[15257]: Error connecting server [172.16.31.50]:8080. Jan 17 16:40:42 proxy2 Keepalived_healthcheckers[15257]: Removing service [172.16.31.50]:8080 from VS [172.16.31.188]:0 Jan 17 16:40:42 proxy2 Keepalived_healthcheckers[15257]: Remote SMTP server [127.0.0.1]:25 connected. Jan 17 16:40:42 proxy2 Keepalived_healthcheckers[15257]: SMTP alert successfully sent.
在到MASTER节点查看ipvs规则:
[root@proxy keepalived]# ipvsadm -L -n IP Virtual Server version 1.2.1 (size=4096) Prot LocalAddress:Port Scheduler Flags -> RemoteAddress:Port Forward Weight ActiveConn InActConn TCP 172.16.31.188:0 rr persistent 50 -> 172.16.31.51:8080 Route 1 0 0 You have new mail in /var/spool/mail/root
规则已经将故障的后端tomcat服务器移除,并且我们还收到了邮件;查看邮件:
[root@proxy keepalived]# mail Heirloom Mail version 12.4 7/29/08. Type ? for help. "/var/spool/mail/root": 16 messages 12 new 14 unread 略..... N 12 [email protected] Sat Jan 17 15:44 17/594 "[LVS_DEVEL] Realserver [172.16.31.51]:80 - DOWN" N 13 [email protected] Sat Jan 17 15:44 17/594 "[LVS_DEVEL] Realserver [172.16.31.50]:80 - DOWN" N 14 [email protected] Sat Jan 17 15:50 17/594 "[LVS_DEVEL] Realserver [172.16.31.51]:80 - DOWN" N 15 [email protected] Sat Jan 17 15:50 17/594 "[LVS_DEVEL] Realserver [172.16.31.50]:80 - DOWN" N 16 [email protected] Sat Jan 17 16:40 17/596 "[LVS_DEVEL] Realserver [172.16.31.50]:8080 - DOWN" & 16 Message 16: From [email protected] Sat Jan 17 16:40:39 2015 Return-Path:X-Original-To: root@localhost Delivered-To: [email protected] Date: Sat, 17 Jan 2015 08:40:39 +0000 From: [email protected] Subject: [LVS_DEVEL] Realserver [172.16.31.50]:8080 - DOWN X-Mailer: Keepalived To: [email protected] Status: R => CHECK failed on service : connection error <= & quit Held 16 messages in /var/spool/mail/root
注:我在这里还测试了优先级相同时,LVS的主备节点的选举情况,如果优先级都是100的话,我们停止MASTER节点的keepalived服务后,VIP和IPVS规则自动转移到了BACKUP节点,但是我们将MASTER节点的keepalived服务启动后,VIP还是在 BACKUP节点上,ipvs规则会在MASTER节点上生成;
所以我们在生产环境中一定要注意优先级的设置,主节点的优先级必须比备用节点的优先级高;
至此,LVS+Keepalived实现高可用负载均衡集群已全部完成