不同子网上的 2 个节点的 IP 故障转移:无法从第二个节点 ping 虚拟 IP?

不同子网上的 2 个节点的 IP 故障转移:无法从第二个节点 ping 虚拟 IP?

我要设置冗余故障转移Redmine

  • 另一个实例安装在第二台服务器上没有问题
  • MySQL(与 Redmine 在同一台机器上运行)配置为主-主复制

因为它们位于不同的子网(192.168.3.x 和 192.168.6.x),所以VIPArip是唯一的选择。

/etc/ha.d/ha.cf在节点1上

logfacility none
debug 1
debugfile /var/log/ha-debug
logfile /var/log/ha-log
autojoin none
warntime 3
deadtime 6
initdead 60
udpport 694
ucast eth1 node2.ip
keepalive 1
node node1
node node2
crm respawn

/etc/ha.d/ha.cf在节点2上:

logfacility none
debug 1
debugfile /var/log/ha-debug
logfile /var/log/ha-log
autojoin none
warntime 3
deadtime 6
initdead 60
udpport 694
ucast eth0 node1.ip
keepalive 1
node node1
node node2
crm respawn

crm configure show

node $id="6c27077e-d718-4c82-b307-7dccaa027a72" node1
node $id="740d0726-e91d-40ed-9dc0-2368214a1f56" node2
primitive VIPArip ocf:heartbeat:VIPArip \
        params ip="192.168.6.8" nic="lo:0" \
        op start interval="0" timeout="20s" \
        op monitor interval="5s" timeout="20s" depth="0" \
        op stop interval="0" timeout="20s" \
        meta is-managed="true"
property $id="cib-bootstrap-options" \
        stonith-enabled="false" \
        dc-version="1.0.12-unknown" \
        cluster-infrastructure="Heartbeat" \
        last-lrm-refresh="1338870303"

crm_mon -1

============
Last updated: Tue Jun  5 18:36:42 2012
Stack: Heartbeat
Current DC: node2 (740d0726-e91d-40ed-9dc0-2368214a1f56) - partition with quorum
Version: 1.0.12-unknown
2 Nodes configured, unknown expected votes
1 Resources configured.
============

Online: [ node1 node2 ]

 VIPArip    (ocf::heartbeat:VIPArip):   Started node1

ip addr show lo

1: lo: <LOOPBACK,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
    inet 192.168.6.8/32 scope global lo
    inet6 ::1/128 scope host 
       valid_lft forever preferred_lft forever

我可以从节点 1(192.168.3.x)ping 192.168.6.8:

# ping -c 4 192.168.6.8
PING 192.168.6.8 (192.168.6.8) 56(84) bytes of data.
64 bytes from 192.168.6.8: icmp_seq=1 ttl=64 time=0.062 ms
64 bytes from 192.168.6.8: icmp_seq=2 ttl=64 time=0.046 ms
64 bytes from 192.168.6.8: icmp_seq=3 ttl=64 time=0.059 ms
64 bytes from 192.168.6.8: icmp_seq=4 ttl=64 time=0.071 ms

--- 192.168.6.8 ping statistics ---
4 packets transmitted, 4 received, 0% packet loss, time 3000ms
rtt min/avg/max/mdev = 0.046/0.059/0.071/0.011 ms

但无法从节点 2 (192.168.6.x) 和外部 ping 虚拟 IP。我是不是漏掉了什么?

PS:如果您得到如下信息,您可能需要IP2UTIL=/sbin/ip在资源代理脚本中进行设置:/usr/lib/ocf/resource.d/heartbeat/VIPArip

6月5日 11:08:10 node1 lrmd:[19832]:信息:RA输出:(VIPArip:stop:stderr)2012/06/05_11:08:10 错误:无效的OCF_RESK EY_ip [192.168.6.8]

http://www.clusterlabs.org/wiki/Debugging_Resource_Failures


回复@DukeLion:

哪个路由器接收 RIP 更新?

当我启动VIPArip资源时,ripd 使用以下配置文件运行(在节点 1 上):

/var/run/resource-agents/VIPArip-ripd.conf

hostname ripd
password zebra
debug rip events
debug rip packet
debug rip zebra
log file /var/log/quagga/quagga.log
router rip
!nic_tag
 no passive-interface lo:0
 network lo:0
 distribute-list private out lo:0
 distribute-list private in lo:0
!metric_tag
 redistribute connected metric 3
!ip_tag
access-list private permit 192.168.6.8/32
access-list private deny any

show ip route

Codes: K - kernel route, C - connected, S - static, R - RIP,
       O - OSPF, I - IS-IS, B - BGP, A - Babel,
       > - selected route, * - FIB route

K>* 0.0.0.0/0 via 192.168.3.1, eth1
C>* 127.0.0.0/8 is directly connected, lo
K>* 169.254.0.0/16 is directly connected, eth1
C>* 192.168.3.0/24 is directly connected, eth1
C>* 192.168.6.8/32 is directly connected, lo

sh ip rip status

Routing Protocol is "rip"
  Sending updates every 30 seconds with +/-50%, next due in 7 seconds
  Timeout after 180 seconds, garbage collect after 120 seconds
  Outgoing update filter list for all interface is not set
    lo:0 filtered by private
  Incoming update filter list for all interface is not set
    lo:0 filtered by private
  Default redistribution metric is 1
  Redistributing: connected
  Default version control: send version 2, receive any version 
    Interface        Send  Recv   Key-chain
  Routing for Networks:
    lo:0
  Routing Information Sources:
    Gateway          BadPackets BadRoutes  Distance Last Update
  Distance: (default is 120)

答案1

我认为问题不在于集群配置,而在于您的路由架构。

VIPArip 资源代理管理本地 quagga 以发送路由更新。但您还需要使用此路由更新来更改路由以指向活动服务器。我将尝试解释它的工作原理。

安息吧

看图。HA1 和 HA2 是 linux-ha 集群成员,其中 quagga 正在运行。蓝色路由器从两个网络链路监听 RIP。

当 vip 在 HA1 上启动时,quagga 向蓝色路由器发送 RIP 更新。它将 vip 前缀添加到其路由表中,下一跳为 192.168.1.2。

当发生故障转移时,HA1 上的 vip 会关闭,quagga 会完全停止,因此不会发送更新。蓝色路由器将在超时后删除路由表记录,即使 VIP 不会在 HA2 上启动。当 VIP 在 HA2 上启动时,它会启动 quagga 并发送 RIP 更新。蓝色路由器将使用 192.168.2.2 nexthop 将记录添加到路由表。

可以在更复杂的网络拓扑中使用 viparip,只需确保边界路由器在整个网络中获得路由更新。

相关内容