[root@example ~]# yum install ipvsadm -y 安装ip_vs管理工具 理论部分我们在上篇文章中详细介绍了，这次进入实操阶段 [root@example ~]# modprobe ip_vs 加载ip_vs模块 [root@example ~]# modprobe ip_vs_rr [root@example ~]# lsmod | grep ip_vs ip_vs_rr 1420 0 ip_vs 126534 2 ip_vs_rr libcrc32c 1246 1 ip_vs ipv6 335589 271 ip_vs [root@example ~]# ip addr add /24 dev eth0 label eth0:1 手动加vip地址 [root@example ~]# ip addr 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 /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:29:5a:e1:a3 brd ff:ff:ff:ff:ff:ff inet /24 brd scope global eth0 inet /24 scope global secondary eth0:1 inet6 fe80::20c:29ff:fe5a:e1a3/64 scope link valid_lft forever preferred_lft forever
本次环境规划如下 操作系统
[root@example ~]# curl web01_ [root@example ~]# curl web02_ 手工执行配置lvs服务器并添加两台rs服务 查看相关的帮助命令 [root@example ~]# ipvsadm --help ipvsadm 2008/5/15 (compiled with popt and IPVS ) Usage: ipvsadm -A|E -t|u|f service-address [-s scheduler] [-p [timeout]] [-M netmask] [--pe persistence_engine] ipvsadm -D -t|u|f service-address ipvsadm -C ipvsadm -R ipvsadm -S [-n] ipvsadm -a|e -t|u|f service-address -r server-address [options] ipvsadm -d -t|u|f service-address -r server-address ipvsadm -L|l [options] ipvsadm -Z [-t|u|f service-address] ipvsadm --set tcp tcpfin udp ipvsadm --start-daemon state [--mcast-interface interface] [--syncid sid] ipvsadm --stop-daemon state ipvsadm -h Commands: Either long or short options are allowed. --add-service -A add virtual service with options --edit-service -E edit virtual service with options --delete-service -D delete virtual service --clear -C clear the whole table --restore -R restore rules from stdin --save -S save rules to stdout --add-server -a add real server with options --edit-server -e edit real server with options --delete-server -d delete real server --list -L|-l list the table --zero -Z zero counters in a service or all services --set tcp tcpfin udp set connection timeout values --start-daemon start connection sync daemon --stop-daemon stop connection sync daemon --help -h display this help message [root@example ~]# ipvsadm -C 清空全部配置 [root@example ~]# ipvsadm -A -t :80 -s wrr 指定vip并设定调度算法 [root@example ~]# ipvsadm -a -t :80 -r :80 -g -w 1 添加rs [root@example ~]# ipvsadm -a -t :80 -r :80 -g -w 3 [root@example ~]# ipvsadm -Ln IP Virtual Server version (size=4096) Prot LocalAddress:Port Scheduler Flags -> RemoteAddress:Port Forward Weight ActiveConn InActConn TCP :80 wrr -> :80 Route 1 0 0 -> :80 Route 3 0 0
手工在rs服务器上面绑定操作 第一台 [root@example ~]# ifconfig lo:132 netmask up [root@example ~]# route add -host dev lo [root@example ~]# ip addr 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 /8 scope host lo inet /32 brd scope global lo:132 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:29:66:66:33 brd ff:ff:ff:ff:ff:ff inet /24 brd scope global eth0 inet6 fe80::20c:29ff:fe66:6633/64 scope link valid_lft forever preferred_lft forever 第二台 [root@example ~]# ip addr add /32 dev lo label lo:133 [root@example ~]# route add -host dev lo [root@example ~]# ip addr 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 /8 scope host lo inet /32 scope global lo:133 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:29:6c:a9:25 brd ff:ff:ff:ff:ff:ff inet /24 brd scope global eth0 inet6 fe80::20c:29ff:fe6c:a925/64 scope link valid_lft forever preferred_lft forever 手工在rs端抑制arp响应 第一台 [root@example ~]# echo "1" >/proc/sys/net/ipv4/conf/lo/arp_ignore [root@example ~]# echo "2" >/proc/sys/net/ipv4/conf/lo/arp_announce [root@example ~]# echo "1" >/proc/sys/net/ipv4/conf/all/arp_ignore [root@example ~]# echo "2" >/proc/sys/net/ipv4/conf/all/arp_announce 第二台 [root@example ~]# echo "1" >/proc/sys/net/ipv4/conf/lo/arp_ignore [root@example ~]# echo "2" >/proc/sys/net/ipv4/conf/lo/arp_announce [root@example ~]# echo "1" >/proc/sys/net/ipv4/conf/all/arp_ignore [root@example ~]# echo "2" >/proc/sys/net/ipv4/conf/all/arp_announce
测试访问，不停的刷新，同时打开web服务器的log观察
web01_132日志 [root@example ~]# tail -f /var/log/httpd/access_log - - [13/Aug/2016:14:34:21 +0800] "GET / HTTP/" 200 22 "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:24 +0800] "GET / HTTP/" 200 22 "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:25 +0800] "GET / HTTP/" 200 22 "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" web01_133日志 [root@example ~]# tail -f /var/log/httpd/access_log - - [13/Aug/2016:14:33:54 +0800] "GET / HTTP/" 200 22 "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:33:54 +0800] "GET / HTTP/" 404 290 "/" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:05 +0800] "GET / HTTP/" 304 - "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:17 +0800] "GET / HTTP/" 304 - "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:20 +0800] "GET / HTTP/" 304 - "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:22 +0800] "GET / HTTP/" 200 22 "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:23 +0800] "GET / HTTP/" 304 - "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:23 +0800] "GET / HTTP/" 304 - "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:24 +0800] "GET / HTTP/" 200 22 "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:25 +0800] "GET / HTTP/" 304 - "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:25 +0800] "GET / HTTP/" 304 - "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:26 +0800] "GET / HTTP/" 200 22 "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/" - - [13/Aug/2016:14:34:26 +0800] "GET / HTTP/" 304 - "-" "Mozilla/ (Windows NT ; WOW64) AppleWebKit/ (KHTML, like Gecko) Chrome/ Safari/"
[root@example ~]# ipvsadm -Ln --stats 可以看到基本符合我们设定的权重1:3 IP Virtual Server version (size=4096) Prot LocalAddress:Port Conns InPkts OutPkts InBytes OutBytes -> RemoteAddress:Port TCP :80 66 335 0 44474 0 -> :80 16 82 0 10978 0 -> :80 50 253 0 33496 0
上面日志可能不够直观，最好再开一个客户端去执行for循环测试，这里就不在演示
以上手工创建只是为了让我更好的理解lvs的过程，好了下面开始我们的keepalived部分
安装keepalived
[root@example ~]#yum install keepalived -y
配置keepalived
主节点上面
[root@example ~]# cat /etc/keepalived/
global_defs {
notification_email {
1335120568@
}
notification_email_from @
smtp_server
smtp_connect_timeout 30
router_id LVS_01
}
vrrp_instance web {
state MASTER
interface eth0
virtual_router_id 51
priority 150
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
/24
}
}
virtual_server 80 {
delay_loop 6
lb_algo wrr
lb_kind DR
nat_mask
persistence_timeout 50
protocol TCP
real_server 80 {
weight 1
TCP_CHECK {
connect_timeout 8
nb_get_retry 3
delay_before_retry 3
connect_port 80
}
}
real_server 80 {
weight 1
TCP_CHECK {
connect_timeout 8
nb_get_retry 3
delay_before_retry 3
connect_port 80
}
}
}
备用节点上面keepalived配置
[root@example ~]# cat /etc/keepalived/
global_defs {
notification_email {
1335120568@
}
notification_email_from @
smtp_server
smtp_connect_timeout 30
router_id LVS_02
}
vrrp_instance web {
state BACKUP
interface eth0
virtual_router_id 51
priority 100
advert_int 1
authentication {
auth_type PASS
auth_pass 1111
}
virtual_ipaddress {
/24
}
}
virtual_server 80 {
delay_loop 6
lb_algo wrr
lb_kind DR
nat_mask
persistence_timeout 50
protocol TCP
real_server 80 {
weight 1
TCP_CHECK {
connect_timeout 8
nb_get_retry 3
delay_before_retry 3
connect_port 80
}
}
real_server 80 {
weight 1
TCP_CHECK {
connect_timeout 8
nb_get_retry 3
delay_before_retry 3
connect_port 80
}
}
}
启动主备keepalived的观察log [root@example ~]# /etc//keepalived start Starting keepalived: [ OK ] 主节点上面 [root@example keepalived]# tail -f /var/log/messages Aug 13 15:11:10 example Keepalived[45366]: Starting Keepalived (03/19,2015) Aug 13 15:11:10 example Keepalived[45367]: Starting Healthcheck child process, pid=45369 Aug 13 15:11:10 example Keepalived[45367]: Starting VRRP child process, pid=45370 Aug 13 15:11:10 example Keepalived_healthcheckers[45369]: Netlink reflector reports IP added Aug 13 15:11:10 example Keepalived_vrrp[45370]: Netlink reflector reports IP added Aug 13 15:11:10 example Keepalived_vrrp[45370]: Netlink reflector reports IP added Aug 13 15:11:10 example Keepalived_vrrp[45370]: Netlink reflector reports IP fe80::20c:29ff:fe5a:e1a3 added Aug 13 15:11:10 example Keepalived_vrrp[45370]: Registering Kernel netlink reflector Aug 13 15:11:10 example Keepalived_vrrp[45370]: Registering Kernel netlink command channel Aug 13 15:11:10 example Keepalived_vrrp[45370]: Registering gratuitous ARP shared channel Aug 13 15:11:10 example Keepalived_healthcheckers[45369]: Netlink reflector reports IP added Aug 13 15:11:10 example Keepalived_healthcheckers[45369]: Netlink reflector reports IP fe80::20c:29ff:fe5a:e1a3 added Aug 13 15:11:10 example Keepalived_healthcheckers[45369]: Registering Kernel netlink reflector Aug 13 15:11:10 example Keepalived_healthcheckers[45369]: Registering Kernel netlink command channel Aug 13 15:11:10 example Keepalived_vrrp[45370]: Opening file '/etc/keepalived/'. Aug 13 15:11:10 example Keepalived_vrrp[45370]: Configuration is using : 63278 Bytes Aug 13 15:11:10 example Keepalived_vrrp[45370]: Using LinkWatch kernel netlink reflector... Aug 13 15:11:10 example Keepalived_healthcheckers[45369]: Opening file '/etc/keepalived/'. Aug 13 15:11:10 example Keepalived_healthcheckers[45369]: Configuration is using : 14742 Bytes Aug 13 15:11:10 example Keepalived_vrrp[45370]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(10,11)] Aug 13 15:11:10 example Keepalived_healthcheckers[45369]: Using LinkWatch kernel netlink reflector... Aug 13 15:11:10 example Keepalived_healthcheckers[45369]: Activating healthchecker for service []:80 Aug 13 15:11:10 example Keepalived_healthcheckers[45369]: Activating healthchecker for service []:80 Aug 13 15:11:11 example Keepalived_vrrp[45370]: VRRP_Instance(web) Transition to MASTER STATE Aug 13 15:11:11 example Keepalived_vrrp[45370]: VRRP_Instance(web) Received lower prio advert, forcing new election Aug 13 15:11:12 example Keepalived_vrrp[45370]: VRRP_Instance(web) Entering MASTER STATE Aug 13 15:11:12 example Keepalived_vrrp[45370]: VRRP_Instance(web) setting protocol VIPs. Aug 13 15:11:12 example Keepalived_vrrp[45370]: VRRP_Instance(web) Sending gratuitous ARPs on eth0 for Aug 13 15:11:17 example Keepalived_vrrp[45370]: VRRP_Instance(web) Sending gratuitous ARPs on eth0 for Aug 13 15:13:36 example dhclient[1239]: DHCPREQUEST on eth0 to port 67 (xid=0x3b1957c6) Aug 13 15:13:36 example dhclient[1239]: DHCPACK from (xid=0x3b1957c6) Aug 13 15:13:38 example dhclient[1239]: bound to -- renewal in 678 seconds. 备用节点上面 [root@example keepalived]# tail -f /var/log/messages Aug 13 15:11:03 example Keepalived[1937]: Starting Keepalived (03/19,2015) Aug 13 15:11:03 example Keepalived[1938]: Starting Healthcheck child process, pid=1940 Aug 13 15:11:03 example Keepalived[1938]: Starting VRRP child process, pid=1941 Aug 13 15:11:03 example Keepalived_vrrp[1941]: Netlink reflector reports IP added Aug 13 15:11:03 example Keepalived_healthcheckers[1940]: Netlink reflector reports IP added Aug 13 15:11:03 example Keepalived_healthcheckers[1940]: Netlink reflector reports IP fe80::20c:29ff:feea:4c15 added Aug 13 15:11:03 example Keepalived_healthcheckers[1940]: Registering Kernel netlink reflector Aug 13 15:11:03 example Keepalived_healthcheckers[1940]: Registering Kernel netlink command channel Aug 13 15:11:03 example Keepalived_vrrp[1941]: Netlink reflector reports IP fe80::20c:29ff:feea:4c15 added Aug 13 15:11:03 example Keepalived_vrrp[1941]: Registering Kernel netlink reflector Aug 13 15:11:03 example Keepalived_vrrp[1941]: Registering Kernel netlink command channel Aug 13 15:11:03 example Keepalived_vrrp[1941]: Registering gratuitous ARP shared channel Aug 13 15:11:03 example Keepalived_vrrp[1941]: Opening file '/etc/keepalived/'. Aug 13 15:11:03 example Keepalived_vrrp[1941]: Configuration is using : 63278 Bytes Aug 13 15:11:03 example Keepalived_vrrp[1941]: Using LinkWatch kernel netlink reflector... Aug 13 15:11:03 example Keepalived_vrrp[1941]: VRRP_Instance(web) Entering BACKUP STATE Aug 13 15:11:03 example Keepalived_vrrp[1941]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(10,11)] Aug 13 15:11:03 example Keepalived_healthcheckers[1940]: Opening file '/etc/keepalived/'. Aug 13 15:11:03 example Keepalived_healthcheckers[1940]: Configuration is using : 14742 Bytes Aug 13 15:11:04 example Keepalived_healthcheckers[1940]: Using LinkWatch kernel netlink reflector... Aug 13 15:11:04 example Keepalived_healthcheckers[1940]: Activating healthchecker for service []:80 Aug 13 15:11:04 example Keepalived_healthcheckers[1940]: Activating healthchecker for service []:80 Aug 13 15:11:04 example kernel: IPVS: [wrr] scheduler registered. Aug 13 15:11:07 example Keepalived_vrrp[1941]: VRRP_Instance(web) Transition to MASTER STATE Aug 13 15:11:08 example Keepalived_vrrp[1941]: VRRP_Instance(web) Entering MASTER STATE Aug 13 15:11:08 example Keepalived_vrrp[1941]: VRRP_Instance(web) setting protocol VIPs. Aug 13 15:11:08 example Keepalived_vrrp[1941]: VRRP_Instance(web) Sending gratuitous ARPs on eth0 for Aug 13 15:11:08 example Keepalived_healthcheckers[1940]: Netlink reflector reports IP added Aug 13 15:11:10 example ntpd[1525]: Listen normally on 6 eth0 UDP 123 Aug 13 15:11:11 example Keepalived_vrrp[1941]: VRRP_Instance(web) Received higher prio advert Aug 13 15:11:11 example Keepalived_vrrp[1941]: VRRP_Instance(web) Entering BACKUP STATE Aug 13 15:11:11 example Keepalived_vrrp[1941]: VRRP_Instance(web) removing protocol VIPs. Aug 13 15:11:11 example Keepalived_healthcheckers[1940]: Netlink reflector reports IP removed Aug 13 15:11:13 example ntpd[1525]: Deleting interface #6 eth0, #123, interface stats: received=0, sent=0, dropped=0, active_time=3 secs 主节点vip地址情况 [root@example ~]# ip addr 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 /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:29:5a:e1:a3 brd ff:ff:ff:ff:ff:ff inet /24 brd scope global eth0 inet /24 scope global secondary eth0:1 inet6 fe80::20c:29ff:fe5a:e1a3/64 scope link valid_lft forever preferred_lft forever 备用节点vip地址情况 [root@example ~]# ip addr 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 /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:ea:4c:15 brd ff:ff:ff:ff:ff:ff inet /24 brd scope global eth0 inet6 fe80::20c:29ff:feea:4c15/64 scope link valid_lft forever preferred_lft forever
rs服务器配置（两台都需要做）
[root@example ~]# echo "1" >/proc/sys/net/ipv4/conf/lo/arp_ignore
[root@example ~]# echo "2" >/proc/sys/net/ipv4/conf/lo/arp_announce
[root@example ~]# echo "1" >/proc/sys/net/ipv4/conf/all/arp_ignore
[root@example ~]# echo "2" >/proc/sys/net/ipv4/conf/all/arp_announce
[root@example ~]# ip addr
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 /8 scope host lo
inet /32 scope global lo:1
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:29:66:66:33 brd ff:ff:ff:ff:ff:ff
inet /24 brd scope global eth0
inet6 fe80::20c:29ff:fe66:6633/64 scope link
valid_lft forever preferred_lft forever
高可用性功能测试 高可用性是通过LVS的两个Director Server完成的，为了模拟故障，先将主Director Server 上面的Keepalived服务停止，然后观察备用Director Server上Keepalived的运行日志，信息如下： [root@example ~]# /etc//keepalived stop 主节点 Stopping keepalived: [ OK ] 主节点日志 Aug 13 15:18:53 example Keepalived[45367]: Stopping Keepalived (03/19,2015) Aug 13 15:18:53 example Keepalived_vrrp[45370]: VRRP_Instance(web) sending 0 priority Aug 13 15:18:53 example Keepalived_vrrp[45370]: VRRP_Instance(web) removing protocol VIPs. Aug 13 15:18:53 example Keepalived_healthcheckers[45369]: Netlink reflector reports IP removed Aug 13 15:18:53 example Keepalived_healthcheckers[45369]: Removing service []:80 from VS []:80 Aug 13 15:18:53 example Keepalived_healthcheckers[45369]: Removing service []:80 from VS []:80 Aug 13 15:18:53 example kernel: IPVS: __ip_vs_del_service: enter Aug 13 15:18:54 example ntpd[1525]: Deleting interface #6 eth0:1, #123, interface stats: received=0, sent=0, dropped=0, active_time=5131 secs 备用节点日志 Aug 13 15:18:53 example Keepalived_vrrp[1941]: VRRP_Instance(web) Transition to MASTER STATE Aug 13 15:18:54 example Keepalived_vrrp[1941]: VRRP_Instance(web) Entering MASTER STATE Aug 13 15:18:54 example Keepalived_vrrp[1941]: VRRP_Instance(web) setting protocol VIPs. Aug 13 15:18:54 example Keepalived_vrrp[1941]: VRRP_Instance(web) Sending gratuitous ARPs on eth0 for Aug 13 15:18:54 example Keepalived_healthcheckers[1940]: Netlink reflector reports IP added Aug 13 15:18:56 example ntpd[1525]: Listen normally on 7 eth0 UDP 123 Aug 13 15:18:59 example Keepalived_vrrp[1941]: VRRP_Instance(web) Sending gratuitous ARPs on eth0 for
主节点keepalived恢复观察日志情况 [root@example ~]# /etc//keepalived stop Stopping keepalived: [ OK ] 主节点日志 Aug 13 15:22:04 example Keepalived[45420]: Starting Keepalived (03/19,2015) Aug 13 15:22:04 example Keepalived[45421]: Starting Healthcheck child process, pid=45423 Aug 13 15:22:04 example Keepalived[45421]: Starting VRRP child process, pid=45424 Aug 13 15:22:04 example Keepalived_vrrp[45424]: Netlink reflector reports IP added Aug 13 15:22:04 example Keepalived_vrrp[45424]: Netlink reflector reports IP fe80::20c:29ff:fe5a:e1a3 added Aug 13 15:22:04 example Keepalived_vrrp[45424]: Registering Kernel netlink reflector Aug 13 15:22:04 example Keepalived_vrrp[45424]: Registering Kernel netlink command channel Aug 13 15:22:04 example Keepalived_vrrp[45424]: Registering gratuitous ARP shared channel Aug 13 15:22:04 example Keepalived_healthcheckers[45423]: Netlink reflector reports IP added Aug 13 15:22:04 example Keepalived_healthcheckers[45423]: Netlink reflector reports IP fe80::20c:29ff:fe5a:e1a3 added Aug 13 15:22:04 example Keepalived_healthcheckers[45423]: Registering Kernel netlink reflector Aug 13 15:22:04 example Keepalived_healthcheckers[45423]: Registering Kernel netlink command channel Aug 13 15:22:05 example Keepalived_vrrp[45424]: Opening file '/etc/keepalived/'. Aug 13 15:22:05 example Keepalived_vrrp[45424]: Configuration is using : 63278 Bytes Aug 13 15:22:05 example Keepalived_vrrp[45424]: Using LinkWatch kernel netlink reflector... Aug 13 15:22:05 example Keepalived_vrrp[45424]: VRRP sockpool: [ifindex(2), proto(112), unicast(0), fd(10,11)] Aug 13 15:22:05 example Keepalived_healthcheckers[45423]: Opening file '/etc/keepalived/'. Aug 13 15:22:05 example Keepalived_healthcheckers[45423]: Configuration is using : 14742 Bytes Aug 13 15:22:05 example Keepalived_healthcheckers[45423]: Using LinkWatch kernel netlink reflector... Aug 13 15:22:05 example Keepalived_healthcheckers[45423]: Activating healthchecker for service []:80 Aug 13 15:22:05 example Keepalived_healthcheckers[45423]: Activating healthchecker for service []:80 Aug 13 15:22:05 example Keepalived_vrrp[45424]: VRRP_Instance(web) Transition to MASTER STATE Aug 13 15:22:05 example Keepalived_vrrp[45424]: VRRP_Instance(web) Received lower prio advert, forcing new election Aug 13 15:22:06 example Keepalived_vrrp[45424]: VRRP_Instance(web) Entering MASTER STATE Aug 13 15:22:06 example Keepalived_vrrp[45424]: VRRP_Instance(web) setting protocol VIPs. Aug 13 15:22:06 example Keepalived_vrrp[45424]: VRRP_Instance(web) Sending gratuitous ARPs on eth0 for Aug 13 15:22:06 example Keepalived_healthcheckers[45423]: Netlink reflector reports IP added Aug 13 15:22:07 example ntpd[1525]: Listen normally on 7 eth0 UDP 123 Aug 13 15:22:11 example Keepalived_vrrp[45424]: VRRP_Instance(web) Sending gratuitous ARPs on eth0 for 备用节点 Aug 13 15:22:05 example Keepalived_vrrp[1941]: VRRP_Instance(web) Received higher prio advert Aug 13 15:22:05 example Keepalived_vrrp[1941]: VRRP_Instance(web) Entering BACKUP STATE Aug 13 15:22:05 example Keepalived_vrrp[1941]: VRRP_Instance(web) removing protocol VIPs. Aug 13 15:22:05 example Keepalived_healthcheckers[1940]: Netlink reflector reports IP removed Aug 13 15:22:07 example ntpd[1525]: Deleting interface #7 eth0, #123, interface stats: received=0, sent=0, dropped=0, active_time=191 secs Aug 13 15:22:54 example dhclient[1239]: DHCPREQUEST on eth0 to port 67 (xid=0x6d287b9) Aug 13 15:22:54 example dhclient[1239]: DHCPACK from (xid=0x6d287b9) Aug 13 15:22:56 example dhclient[1239]: bound to -- renewal in 680 seconds.
可以看到备用节点的vip地址被移除了，一切都符合预期
负载均衡测试，打开浏览器测试访问lvs的vip地址
故障切换测试
故障切换是测试在某个节点出现故障后，Keepalived监控模块是否能及时发现，然后屏蔽故障节点，同时将服务转移到正常节点上执行。
这里将real server 1节点服务停掉，假定这个节点出现故障，然后查看主、备机日志信息，相关日志如下：
[root@example ~]# /etc//httpd stop
Stopping httpd: [ OK ]
观察主节点keepalived 日志情况 Aug 13 15:30:05 example Keepalived_healthcheckers[45423]: TCP connection to []:80 failed !!! Aug 13 15:30:05 example Keepalived_healthcheckers[45423]: Removing service []:80 from VS []:80 Aug 13 15:30:05 example Keepalived_healthcheckers[45423]: Remote SMTP server []:25 connected. Aug 13 15:30:35 example Keepalived_healthcheckers[45423]: Timeout reading data to remote SMTP server []:25. 观察备用节点keepalived 日志情况 Aug 13 15:30:04 example Keepalived_healthcheckers[1940]: TCP connection to []:80 failed !!! Aug 13 15:30:04 example Keepalived_healthcheckers[1940]: Removing service []:80 from VS []:80 Aug 13 15:30:04 example Keepalived_healthcheckers[1940]: Remote SMTP server []:25 connected. Aug 13 15:30:34 example Keepalived_healthcheckers[1940]: Timeout reading data to remote SMTP server []:25. 通过日志可以看出，Keepalived监控模块检测到这台主机出现故障后，将此节点从集群系统中剔除掉了。 此时访问这个地址，应该只能看到“web02_”了， 这是因为节点1出现故障，而Keepalived监控模块将节点1从集群系统中剔除了 再次重启web01服务观察主备keepalived 日志情况 主节点keepalived 日志情况 Aug 13 15:33:47 example Keepalived_healthcheckers[45423]: TCP connection to []:80 success. Aug 13 15:33:47 example Keepalived_healthcheckers[45423]: Adding service []:80 to VS []:80 Aug 13 15:33:47 example Keepalived_healthcheckers[45423]: Remote SMTP server []:25 connected. Aug 13 15:34:17 example Keepalived_healthcheckers[45423]: Timeout reading data to remote SMTP server []:25. 备用节点keepalived 日志情况 Aug 13 15:33:46 example Keepalived_healthcheckers[1940]: TCP connection to []:80 success. Aug 13 15:33:46 example Keepalived_healthcheckers[1940]: Adding service []:80 to VS []:80 Aug 13 15:33:46 example Keepalived_healthcheckers[1940]: Remote SMTP server []:25 connected. Aug 13 15:34:16 example dhclient[1239]: DHCPREQUEST on eth0 to port 67 (xid=0x6d287b9) Aug 13 15:34:16 example dhclient[1239]: DHCPACK from (xid=0x6d287b9) Aug 13 15:34:16 example Keepalived_healthcheckers[1940]: Timeout reading data to remote SMTP server []:25. Aug 13 15:34:18 example dhclient[1239]: bound to -- renewal in 778 seconds.
从日志可知，Keepalived监控模块检测到这台主机恢复正常后，又将此节点加入到集群系统中。
此时再次访问这个地址，然后不断刷新此页面，应该又能分别看到“web01_”和
“web01_”页面了，这说明在web1节点恢复正常后，Keepalived监控模块将此节点加入到集群系统中。