【转】Keepalived无法绑定VIP故障排查经历

一 故障描述
我在台湾合作方给定的两台虚拟机上部署HAProxy+Keepalived负载均衡高可用方案。在配置完Keepalived后，重新启动Keepalived，Keepalived没有绑定VIP。

Keepalived执行程序路径为/data/app_platform/keepalived/sbin/keepalived
配置文件路径为/data/app_platform/keepalived/conf/keepalived.conf  
Keepalived的启动脚本为/etc/init.d/keepalived

keepalived.conf的内容
LB1 Master
! Configuration File for keepalived 
  
global_defs { 
  notification_email { 
    admin@example.com 
  } 
  notification_email_from lb1@example.com 
  smtp_server 127.0.0.1 
  smtp_connect_timeout 30 
  router_id LB1_MASTER 
} 
  
 vrrp_script chk_haproxy { 
        script "killall -0 haproxy" 
        interval 2 
    weight 2 
    } 
  
  
  
  
vrrp_instance VI_1 { 
    state MASTER 
    interface eth1 
    virtual_router_id 51 
    priority 100 
    advert_int 1 
    authentication { 
        auth_type PASS 
        auth_pass 1111 
    } 
    virtual_ipaddress { 
  
    10.1.1.200/24      brd 10.1.1.255      dev eth1 label eth1:vip 
  
    } 
  
    track_script { 
        chk_haproxy 
                } 
}

重新启动Keepalived查看日志
Mar  3 18:09:00 cv00300005248-1 Keepalived[20138]: Stopping Keepalived v1.2.15 (02/28,2015) 
Mar  3 18:09:00 cv00300005248-1 Keepalived[20259]: Starting Keepalived v1.2.15 (02/28,2015) 
Mar  3 18:09:00 cv00300005248-1 Keepalived[20260]: Starting Healthcheck child process, pid=20261 
Mar  3 18:09:00 cv00300005248-1 Keepalived[20260]: Starting VRRP child process, pid=20262 
Mar  3 18:09:00 cv00300005248-1 Keepalived_vrrp[20262]: Registering Kernel netlink reflector 
Mar  3 18:09:00 cv00300005248-1 Keepalived_vrrp[20262]: Registering Kernel netlink command channel 
Mar  3 18:09:00 cv00300005248-1 Keepalived_vrrp[20262]: Registering gratuitous ARP shared channel 
Mar  3 18:09:00 cv00300005248-1 Keepalived_healthcheckers[20261]: Registering Kernel netlink reflector 
Mar  3 18:09:00 cv00300005248-1 Keepalived_healthcheckers[20261]: Registering Kernel netlink command channel 
Mar  3 18:09:00 cv00300005248-1 Keepalived_healthcheckers[20261]: Configuration is using : 3924 Bytes 
Mar  3 18:09:00 cv00300005248-1 Keepalived_healthcheckers[20261]: Using LinkWatch kernel netlink reflector... 
Mar  3 18:09:00 cv00300005248-1 Keepalived_vrrp[20262]: Configuration is using : 55712 Bytes 
Mar  3 18:09:00 cv00300005248-1 Keepalived_vrrp[20262]: Using LinkWatch kernel netlink reflector... 
Mar  3 18:09:18 cv00300005248-1 kernel: __ratelimit: 1964 callbacks suppressed 
Mar  3 18:09:18 cv00300005248-1 kernel: Neighbour table overflow. 
Mar  3 18:09:18 cv00300005248-1 kernel: Neighbour table overflow. 
Mar  3 18:09:18 cv00300005248-1 kernel: Neighbour table overflow. 
Mar  3 18:09:18 cv00300005248-1 kernel: Neighbour table overflow. 
Mar  3 18:09:18 cv00300005248-1 kernel: Neighbour table overflow. 
Mar  3 18:09:18 cv00300005248-1 kernel: Neighbour table overflow. 
Mar  3 18:09:18 cv00300005248-1 kernel: Neighbour table overflow. 
Mar  3 18:09:18 cv00300005248-1 kernel: Neighbour table overflow. 
Mar  3 18:09:18 cv00300005248-1 kernel: Neighbour table overflow. 
Mar  3 18:09:18 cv00300005248-1 kernel: Neighbour table overflow.

查看VIP绑定情况
$ ifconfig eth1:vip 
eth1:vip  Link encap:Ethernet  HWaddr 00:16:3E:F2:37:6B  
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1 
          Interrupt:13

没有VIP绑定

二 排查过程

1）检查VIP的配置情况
向合作方确认提供的VIP的详细情况

IPADDR  10.1.1.200
NETMASK  255.255.255.0
GATEWAY  10.1.1.1
Brodcast  10.1.1.255
  
这里设置的是

10.1.1.200/24      brd 10.1.1.255      dev eth1 label eth1:vip

2）检查iptables和selinux的设置情况
$ sudo service iptables stop 
$ sudo setenforce 0 
setenforce: SELinux is disabled

如果非要开启iptables的话，需要作些设定

iptables -I INPUT -i eth1 -d 224.0.0.0/8 -j ACCEPT 
service iptables save

keepalived使用224.0.0.18作为Master和Backup健康检查的通信IP

3）检查相关的内核参数
HAProxy+Keepalived架构需要注意的内核参数有：

# Controls IP packet forwarding
net.ipv4.ip_forward = 1
开启IP转发功能

net.ipv4.ip_nonlocal_bind = 1
开启允许绑定非本机的IP

如果使用LVS的DR或者TUN模式结合Keepalived需要在后端真实服务器上特别设置两个arp相关的参数。这里也设置好。
net.ipv4.conf.lo.arp_ignore = 1
net.ipv4.conf.lo.arp_announce = 2
net.ipv4.conf.all.arp_ignore = 1
net.ipv4.conf.all.arp_announce = 2

4）检查VRRP的设置情况
LB1 Master
state MASTER 
    interface eth1 
    virtual_router_id 51 
    priority 100

LB2 Backup
state BACKUP 
    interface eth1 
    virtual_router_id 51 
    priority 99

Master和Backup的virtual_router_id需要一样，priority需要不一样，数字越大，优先级越高

5）怀疑是编译安装Keepalived版本出现了问题
重新下载并编译2.1.13的版本，并重新启动keepalived，VIP仍然没有被绑定。

线上有个平台的keepalived是通过yum安装的，于是打算先用yum安装keepalived后将配置文件复制过去看看是否可以绑定VIP

rpm -ivh http://ftp.linux.ncsu.edu/pub/epel/6/x86_64/epel-release-6-8.noarch.rpm 
yum -y install keepalived

cp /data/app_platform/keepalived/conf/keepalived.conf  /etc/keepalived/keepalived.conf

重新启动keepalived

然后查看日志
Mar  4 16:42:46 xxxxx Keepalived_healthcheckers[17332]: Registering Kernel netlink reflector 
Mar  4 16:42:46 xxxxx Keepalived_healthcheckers[17332]: Registering Kernel netlink command channel 
Mar  4 16:42:46 xxxxx Keepalived_vrrp[17333]: Opening file '/etc/keepalived/keepalived.conf'. 
Mar  4 16:42:46 xxxxx Keepalived_vrrp[17333]: Configuration is using : 65250 Bytes 
Mar  4 16:42:46 xxxxx Keepalived_vrrp[17333]: Using LinkWatch kernel netlink reflector... 
Mar  4 16:42:46 xxxxx Keepalived_vrrp[17333]: VRRP sockpool: [ifindex(3), proto(112), unicast(0), fd(10,11)] 
Mar  4 16:42:46 xxxxx Keepalived_healthcheckers[17332]: Opening file '/etc/keepalived/keepalived.conf'. 
Mar  4 16:42:46 xxxxx Keepalived_healthcheckers[17332]: Configuration is using : 7557 Bytes 
Mar  4 16:42:46 xxxxx Keepalived_healthcheckers[17332]: Using LinkWatch kernel netlink reflector... 
Mar  4 16:42:46 xxxxx Keepalived_vrrp[17333]: VRRP_Script(chk_haproxy) succeeded 
Mar  4 16:42:47 xxxxx Keepalived_vrrp[17333]: VRRP_Instance(VI_1) Transition to MASTER STATE 
Mar  4 16:42:48 xxxxx Keepalived_vrrp[17333]: VRRP_Instance(VI_1) Entering MASTER STATE 
Mar  4 16:42:48 xxxxx Keepalived_vrrp[17333]: VRRP_Instance(VI_1) setting protocol VIPs. 
Mar  4 16:42:48 xxxxx Keepalived_vrrp[17333]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth1 for 10.1.1.200 
Mar  4 16:42:48 xxxxx Keepalived_healthcheckers[17332]: Netlink reflector reports IP 10.1.1.200 added 
Mar  4 16:42:53 xxxxx Keepalived_vrrp[17333]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth1 for 10.1.1.200

再查看IP绑定情况
$ ifconfig eth1:vip 
eth1:vip  Link encap:Ethernet  HWaddr 00:16:3E:F2:37:6B  
          inet addr:10.1.1.200  Bcast:10.1.1.255  Mask:255.255.255.0 
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1 
          Interrupt:13

再通过yum将keepalived卸载掉
yum remove keepalived
恢复到原来的启动脚本/etc/init.d/keepalived
重新启动keepalived后还是无法绑定VIP

怀疑是keepalived启动脚本/etc/init.d/keepalived的问题
检查/etc/init.d/keepalived
# Source function library. 
. /etc/rc.d/init.d/functions 
  
exec="/data/app_platform/keepalived/sbin/keepalived" 
prog="keepalived" 
config="/data/app_platform/keepalived/conf/keepalived.conf" 
  
[ -e /etc/sysconfig/$prog ] && . /etc/sysconfig/$prog 
  
lockfile=/var/lock/subsys/keepalived 
  
start() { 
    [ -x $exec ] || exit 5 
    [ -e $config ] || exit 6 
    echo -n $"Starting $prog: " 
    daemon $exec $KEEPALIVED_OPTIONS 
    retval=$? 
    echo 
    [ $retval -eq 0 ] && touch $lockfile 
    return $retval 
}

关键是这一行
daemon $exec $KEEPALIVED_OPTIONS

由于没有复制/etc/sysconfig/keepalived,所以将直接执行damon /data/app_platform/keepalived/sbin/keepalived
由于keepalived默认使用的是/etc/keepalived/keepalived.conf作为配置文件，而这里指定了不同的配置文件，所以要修改成为
daemon $exec -D -f $config

重新启动keepalived，查看日志和VIP绑定情况
$ ifconfig eth1:vip 
eth1:vip  Link encap:Ethernet  HWaddr 00:16:3E:F2:37:6B  
          inet addr:10.1.1.200  Bcast:10.1.1.255  Mask:255.255.255.0 
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1 
          Interrupt:13

6）将LB2 Backup的keepalived启动脚本也修改一下，观察VIP接管情况
查看LB1 Master

$ ifconfig eth1:vip 
eth1:vip  Link encap:Ethernet  HWaddr 00:16:3E:F2:37:6B  
          inet addr:10.1.1.200  Bcast:10.1.1.255  Mask:255.255.255.0 
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1 
          Interrupt:13

查看LB2 Backup
$ ifconfig eth1:vip 
eth1:vip  Link encap:Ethernet  HWaddr 00:16:3E:F2:37:6B  
          inet addr:10.1.1.200  Bcast:10.1.1.255  Mask:255.255.255.0 
          UP BROADCAST RUNNING MULTICAST  MTU:1500  Metric:1 
          Interrupt:13

问题出现了，LB1 Master和LB2 Backup都绑定了VIP 10.1.1.200，这是不正常的！！！！

在LB1和LB2上登录10.1.1.200看看

[lb1 ~]$ ssh 10.1.1.200  
Last login: Wed Mar  4 17:31:33 2015 from 10.1.1.200 
[lb1 ~]$

[lb2 ~]$ ssh 10.1.1.200 
Last login: Wed Mar  4 17:54:57 2015 from 101.95.153.246 
[b2  ~]$

在LB1上停掉keepalived，ping下10.1.1.200这个IP，发现无法ping通
在LB2上停掉keepalived，ping下10.1.1.200这个IP，发现也无法ping通

然后开启LB1上的keepalived，LB1上可以ping通10.1.1.200,LB2上不行

开启LB2上的keepalived，LB2上可以ping通10.1.1.200

由此得出，LB1和LB2各自都将VIP 10.1.1.200绑定到本机的eth1网卡上。两台主机并没有VRRP通信，没有VRRP的优先级比较。

7)排查影响VRRP通信的原因
重新启动LB1 Master的Keepalived查看日志
Mar  5 15:45:36 gintama-taiwan-lb1 Keepalived_vrrp[32303]: Configuration is using : 65410 Bytes 
Mar  5 15:45:36 gintama-taiwan-lb1 Keepalived_vrrp[32303]: Using LinkWatch kernel netlink reflector... 
Mar  5 15:45:36 gintama-taiwan-lb1 Keepalived_vrrp[32303]: VRRP sockpool: [ifindex(3), proto(112), unicast(0), fd(10,11)] 
Mar  5 15:45:36 gintama-taiwan-lb1 Keepalived_vrrp[32303]: VRRP_Script(chk_haproxy) succeeded 
Mar  5 15:45:37 gintama-taiwan-lb1 Keepalived_vrrp[32303]: VRRP_Instance(VI_1) Transition to MASTER STATE 
Mar  5 15:45:38 gintama-taiwan-lb1 Keepalived_vrrp[32303]: VRRP_Instance(VI_1) Entering MASTER STATE 
Mar  5 15:45:38 gintama-taiwan-lb1 Keepalived_vrrp[32303]: VRRP_Instance(VI_1) setting protocol VIPs. 
Mar  5 15:45:38 gintama-taiwan-lb1 Keepalived_vrrp[32303]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth1 for 10.1.1.200 
Mar  5 15:45:38 gintama-taiwan-lb1 Keepalived_healthcheckers[32302]: Netlink reflector reports IP 10.1.1.200 added 
Mar  5 15:45:43 gintama-taiwan-lb1 Keepalived_vrrp[32303]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth1 for 10.1.1.200

发现LB1 Master上的Keepalived直接进入Master状态，然后接管VIP

再重新启动LB2 Backup上的Keepalived，查看日志
Mar  5 15:47:42 gintama-taiwan-lb2 Keepalived_vrrp[30619]: Configuration is using : 65408 Bytes 
Mar  5 15:47:42 gintama-taiwan-lb2 Keepalived_vrrp[30619]: Using LinkWatch kernel netlink reflector... 
Mar  5 15:47:42 gintama-taiwan-lb2 Keepalived_vrrp[30619]: VRRP_Instance(VI_1) Entering BACKUP STATE 
Mar  5 15:47:42 gintama-taiwan-lb2 Keepalived_vrrp[30619]: VRRP sockpool: [ifindex(3), proto(112), unicast(0), fd(10,11)] 
Mar  5 15:47:46 gintama-taiwan-lb2 Keepalived_vrrp[30619]: VRRP_Instance(VI_1) Transition to MASTER STATE 
Mar  5 15:47:47 gintama-taiwan-lb2 Keepalived_vrrp[30619]: VRRP_Instance(VI_1) Entering MASTER STATE 
Mar  5 15:47:47 gintama-taiwan-lb2 Keepalived_vrrp[30619]: VRRP_Instance(VI_1) setting protocol VIPs. 
Mar  5 15:47:47 gintama-taiwan-lb2 Keepalived_vrrp[30619]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth1 for 10.1.1.200 
Mar  5 15:47:47 gintama-taiwan-lb2 Keepalived_healthcheckers[30618]: Netlink reflector reports IP 10.1.1.200 added 
Mar  5 15:47:52 gintama-taiwan-lb2 Keepalived_vrrp[30619]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth1 for 10.1.1.200

可以看到LB2上的Keepalived先进入BACKUP状态，然后又转为MASTER状态，然后接管VIP

这样就说明VRRP组播有问题。

既然VRRP组播有问题，就尝试使用单播发送VRRP报文。修改LB1和LB2的配置
LB1
添加以下配置
unicast_src_ip  10.1.1.12 
    unicast_peer { 
                  10.1.1.17 
                }

LB2

添加以下配置
unicast_src_ip  10.1.1.17 
 unicast_peer { 
                  10.1.1.12 
                }

unicast_src_ip 表示发送VRRP单播报文使用的源IP地址
unicast_peer 表示对端接收VRRP单播报文的IP地址

然后各自重新加载keepalived，观察日志

LB1
Mar  5 16:13:35 gintama-taiwan-lb1 Keepalived_vrrp[2551]: VRRP_Instance(VI_1) setting protocol VIPs. 
Mar  5 16:13:35 gintama-taiwan-lb1 Keepalived_vrrp[2551]: VRRP_Script(chk_haproxy) considered successful on reload 
Mar  5 16:13:35 gintama-taiwan-lb1 Keepalived_vrrp[2551]: Configuration is using : 65579 Bytes 
Mar  5 16:13:35 gintama-taiwan-lb1 Keepalived_vrrp[2551]: Using LinkWatch kernel netlink reflector... 
Mar  5 16:13:35 gintama-taiwan-lb1 Keepalived_vrrp[2551]: VRRP sockpool: [ifindex(3), proto(112), unicast(1), fd(10,11)] 
Mar  5 16:13:36 gintama-taiwan-lb1 Keepalived_vrrp[2551]: VRRP_Instance(VI_1) Transition to MASTER STATE 
Mar  5 16:13:48 gintama-taiwan-lb1 Keepalived_vrrp[2551]: VRRP_Instance(VI_1) Received lower prio advert, forcing new election 
Mar  5 16:13:48 gintama-taiwan-lb1 Keepalived_vrrp[2551]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth1 for 10.1.1.200 
Mar  5 16:13:48 gintama-taiwan-lb1 Keepalived_vrrp[2551]: VRRP_Instance(VI_1) Received lower prio advert, forcing new election 
Mar  5 16:13:48 gintama-taiwan-lb1 Keepalived_vrrp[2551]: VRRP_Instance(VI_1) Sending gratuitous ARPs on eth1 for 10.1.1.200

LB2
Mar  5 16:13:48 gintama-taiwan-lb2 Keepalived_vrrp[453]: VRRP_Instance(VI_1) Received higher prio advert 
Mar  5 16:13:48 gintama-taiwan-lb2 Keepalived_vrrp[453]: VRRP_Instance(VI_1) Entering BACKUP STATE 
Mar  5 16:13:48 gintama-taiwan-lb2 Keepalived_vrrp[453]: VRRP_Instance(VI_1) removing protocol VIPs. 
Mar  5 16:13:48 gintama-taiwan-lb2 Keepalived_healthcheckers[452]: Netlink reflector reports IP 10.1.1.200 removed

查看VIP绑定情况，发现LB2上的VIP已经移除

在LB1上LB2上执行ping 10.1.1.200这个VIP
[lb1 ~]$ ping -c 5 10.1.1.200 
PING 10.1.1.200 (10.1.1.200) 56(84) bytes of data. 
64 bytes from 10.1.1.200: icmp_seq=1 ttl=64 time=0.028 ms 
64 bytes from 10.1.1.200: icmp_seq=2 ttl=64 time=0.020 ms 
64 bytes from 10.1.1.200: icmp_seq=3 ttl=64 time=0.020 ms 
64 bytes from 10.1.1.200: icmp_seq=4 ttl=64 time=0.021 ms 
64 bytes from 10.1.1.200: icmp_seq=5 ttl=64 time=0.027 ms 
  
--- 10.1.1.200 ping statistics --- 
5 packets transmitted, 5 received, 0% packet loss, time 3999ms 
rtt min/avg/max/mdev = 0.020/0.023/0.028/0.004 ms

[lb2 ~]$ ping -c 5 10.1.1.200  
PING 10.1.1.200 (10.1.1.200) 56(84) bytes of data. 
  
--- 10.1.1.200 ping statistics --- 
5 packets transmitted, 0 received, 100% packet loss, time 14000ms

当LB1接管VIP的时候LB2居然无法ping通VIP，同样将LB1的Keepalived停掉，LB2可以接管VIP，但是在LB1上无法ping通这个VIP

在LB1和LB2上进行抓包

lb1 ~]$ sudo tcpdump -vvv  -i eth1 host 10.1.1.17 
tcpdump: listening on eth1, link-type EN10MB (Ethernet), capture size 65535 bytes 
16:46:04.827357 IP (tos 0xc0, ttl 255, id 328, offset 0, flags [none], proto VRRP (112), length 40) 
    10.1.1.12 > 10.1.1.17: VRRPv2, Advertisement, vrid 51, prio 102, authtype simple, intvl 1s, length 20, addrs: 10.1.1.200 auth "1111^@^@^@^@" 
16:46:05.827459 IP (tos 0xc0, ttl 255, id 329, offset 0, flags [none], proto VRRP (112), length 40) 
    10.1.1.12 > 10.1.1.17: VRRPv2, Advertisement, vrid 51, prio 102, authtype simple, intvl 1s, length 20, addrs: 10.1.1.200 auth "1111^@^@^@^@" 
16:46:06.828234 IP (tos 0xc0, ttl 255, id 330, offset 0, flags [none], proto VRRP (112), length 40) 
    10.1.1.12 > 10.1.1.17: VRRPv2, Advertisement, vrid 51, prio 102, authtype simple, intvl 1s, length 20, addrs: 10.1.1.200 auth "1111^@^@^@^@" 
16:46:07.828338 IP (tos 0xc0, ttl 255, id 331, offset 0, flags [none], proto VRRP (112), length 40) 
    10.1.1.12 > 10.1.1.17: VRRPv2, Advertisement, vrid 51, prio 102, authtype simple, intvl 1s, length 20, addrs: 10.1.1.200 auth "1111^@^@^@^@"

 

lb2 ~]$ sudo tcpdump -vvv -i eth1 host 10.1.1.12 
tcpdump: listening on eth1, link-type EN10MB (Ethernet), capture size 65535 bytes 
16:48:07.000029 IP (tos 0xc0, ttl 255, id 450, offset 0, flags [none], proto VRRP (112), length 40) 
    10.1.1.12 > 10.1.1.17: VRRPv2, Advertisement, vrid 51, prio 102, authtype simple, intvl 1s, length 20, addrs: 10.1.1.200 auth "1111^@^@^@^@" 
16:48:07.999539 IP (tos 0xc0, ttl 255, id 451, offset 0, flags [none], proto VRRP (112), length 40) 
    10.1.1.12 > 10.1.1.17: VRRPv2, Advertisement, vrid 51, prio 102, authtype simple, intvl 1s, length 20, addrs: 10.1.1.200 auth "1111^@^@^@^@" 
16:48:08.999252 IP (tos 0xc0, ttl 255, id 452, offset 0, flags [none], proto VRRP (112), length 40) 
    10.1.1.12 > 10.1.1.17: VRRPv2, Advertisement, vrid 51, prio 102, authtype simple, intvl 1s, length 20, addrs: 10.1.1.200 auth "1111^@^@^@^@" 
16:48:09.999560 IP (tos 0xc0, ttl 255, id 453, offset 0, flags [none], proto VRRP (112), length 40) 
    10.1.1.12 > 10.1.1.17: VRRPv2, Advertisement, vrid 51, prio 102, authtype simple, intvl 1s, length 20, addrs: 10.1.1.200 auth "1111^@^@^@^@"

在LB1和LB2所在物理机上的其他虚拟机进行VIP ping测试，同样只能是LB1上绑定的VIP只能是LB1所在的物理机上的虚拟机可以ping通，LB2所在的物理机上的虚拟机无法ping通，反之也是一样

有同行建议说VRRP和DHCP也有关系，经过查看对方提供的VM的IP地址居然是DHCP分配的，但是经过测试，VRRP和DHCP没有关系。线上环境最好不要使用DHCP来获取IP地址。

8）请对方技术人员配合检查VIP无法ping通的问题
最终查明对方的内网居然使用的虚拟网络，网关是没有实际作用的。所以部分虚拟机无法通过10.1.1.1这个网关去访问VIP。

让对方虚拟机提供方的技术人员到服务器调试HAProxy+Keepalived，他们通过网络设置使得10.1.1.200这个VIP可以通过内网访问。但是当我测试时，发现当HAProxy挂掉后，Keepalived无法作VIP的切换。

9）解决当HAProxy挂掉后，Keepalived无法对VIP切换的问题。
经过反复测试，发现当Keepalived挂掉后，VIP可以切换。但是当HAProxy挂掉后，VIP无法切换。
仔细检查配置文件和查阅相关资料，最终确定是Keepalived的weight和priority两个参数的大小设置问题。


原来的配置文件中我设置LB1的weight为2，priority为100。LB2的weight为2，priority为99

对方在调试的时候将LB1的priority更改为160.这样反复测试当LB1的HAProxy挂掉后，VIP都无法迁移到LB2上。将LB1上的priority更改为100就可以了。

这里需要注意的是：
主keepalived的priority值与vrrp_script的weight值相减的数字小于备用keepalived的priority 值即可！

vrrp_script 里的script返回值为0时认为检测成功，其它值都会当成检测失败

* weight 为正时，脚本检测成功时此weight会加到priority上，检测失败时不加。

主失败:
            主 priority < 从 priority + weight 时会切换。

主成功：
            主 priority + weight > 从 priority + weight 时，主依然为主


* weight 为负时，脚本检测成功时此weight不影响priority，检测失败时priority - abs(weight)

主失败:
          主 priority - abs(weight) < 从priority 时会切换主从

主成功:
          主 priority > 从priority 主依然为主。

最终的配置文件为：
! Configuration File for keepalived 
  
global_defs { 
  notification_email { 
    admin@example.com 
  } 
  notification_email_from lb1@example.com 
  smtp_server 127.0.0.1 
  smtp_connect_timeout 30 
  router_id LB1_MASTER 
} 
  
 vrrp_script chk_haproxy { 
    script "killall -0 haproxy" 
    interval 2 
    weight 2 
    } 
  
  
  
  
#设置外网的VIP 
vrrp_instance eth0_VIP { 
    state MASTER 
    interface eth0 
    virtual_router_id 51 
    priority 100 
    advert_int 1 
    authentication { 
        auth_type PASS 
        auth_pass 1111 
    } 
    unicast_src_ip  8.8.8.6        #使用VRRP的单播 
    unicast_peer { 
                  8.8.8.7 
                } 
    virtual_ipaddress { 
    8.8.8.8/25  brd 8.8.8.255  dev eth0 label eth0:vip 
  
    } 
  
    track_script { 
        chk_haproxy 
                } 
} 
  
  
#设置内网的VIP 
vrrp_instance eth1_VIP { 
    state MASTER 
    interface eth1 
    virtual_router_id 52 
    priority 100 
    advert_int 1 
    authentication { 
        auth_type PASS 
        auth_pass 1111 
    } 
    unicast_src_ip  10.1.1.12 
    unicast_peer { 
                  10.1.1.17 
                } 
    virtual_ipaddress { 
    10.1.1.200/24    brd 10.1.1.255      dev eth1 label eth1:vip 
  
    } 
  
    track_script { 
        chk_haproxy 
                } 
} 

三 排查总结
在配置Keepalived的时候，需要注意以下几点：

A.内核开启IP转发和允许非本地IP绑定功能，如果是使用LVS的DR模式还需设置两个arp相关的参数。
B.如果Keepalived所在网络不允许使用组播，可以使用VRRP单播
C.需要注意主备的weight和priority的值，这两个值如果设置不合理可能会影响VIP的切换。
D.如果使用的配置文件不是默认的配置文件，在启动Keepalived的时候需要使用 -f 参数指定配置文件。

CentOS 6.3下Haproxy+Keepalived+Apache配置笔记 http://www.linuxidc.com/Linux/2013-06/85598.htm

Haproxy + KeepAlived 实现WEB群集 on CentOS 6 http://www.linuxidc.com/Linux/2012-03/55672.htm

Keepalived+Haproxy配置高可用负载均衡 http://www.linuxidc.com/Linux/2012-03/56748.htm

Haproxy+Keepalived构建高可用负载均衡 http://www.linuxidc.com/Linux/2012-03/55880.htm

CentOS 7 ��配置LVS + Keepalived + ipvsadm http://www.linuxidc.com/Linux/2014-11/109237.htm

Keepalived高可用集群搭建 http://www.linuxidc.com/Linux/2014-09/106965.htm

Keepalived 的详细介绍：请点这里
Keepalived 的下载地址：请点这里

 

转自：https://blog.csdn.net/wh8_2011/article/details/53101885

posted @ 2018-09-12 10:35  caogen1991  阅读(3081)  评论(0)  编辑  收藏  举报