繼續求救，有相關日誌
在我做了cluster配置之後，我的rgmanager的配置如下：
<failoverdomains>
<failoverdomain name="igridnodes1" ordered="1" restricted="1">
<failoverdomainnode name="node1" priority="1"/>
<failoverdomainnode name="node2" priority="2"/>
<failoverdomainnode name="node3" priority="2"/>
</failoverdomain>
<failoverdomain name="igridnodes2" ordered="1" restricted="1">
<failoverdomainnode name="node1" priority="2"/>
<failoverdomainnode name="node2" priority="1"/>
<failoverdomainnode name="node3" priority="2"/>
</failoverdomain>
<failoverdomain name="igridnodes3" ordered="1" restricted="1">
<failoverdomainnode name="node1" priority="2"/>
<failoverdomainnode name="node2" priority="2"/>
<failoverdomainnode name="node3" priority="1"/>
</failoverdomain>
</failoverdomains>
<resources>
<ip address="192.168.11.31" monitor_link="1"/>
<ip address="192.168.11.32" monitor_link="1"/>
<ip address="192.168.11.33" monitor_link="1"/>
</resources>
<service autostart="1" domain="igridnodes1" name="sernode1">
<ip ref="192.168.11.31"/>
</service>
<service autostart="1" domain="igridnodes2" name="sernode2">
<ip ref="192.168.11.32"/>
</service>
<service autostart="1" domain="igridnodes3" name="sernode3">
<ip ref="192.168.11.33"/>
</service>
當我service rgmanager stop的時候，日誌：
May 29 11:39:03 node1 clurgmgrd: <notice> Stopping service sernode1
May 29 11:39:03 node1 clurgmgrd: : <info> Removing IPv4 address 192.168.11.31 from eth1
May 29 11:39:13 node1 clurgmgrd: <notice> Service sernode1 is stopped
May 29 11:39:15 node1 clurgmgrd: <notice> Shutdown complete, exiting
May 29 11:39:15 node1 rgmanager: : <notice> Cluster Service Manager is stopped.
resource IP被remove了
當我service rgmanager start后，日誌：
May 29 11:40:38 node1 clurgmgrd: <notice> Resource Group Manager Starting
May 29 11:40:38 node1 clurgmgrd: <info> Loading Service Data
May 29 11:40:38 node1 rgmanager: clurgmgrd startup succeeded
May 29 11:40:38 node1 clurgmgrd: <info> Initializing Services
May 29 11:40:39 node1 clurgmgrd: <info> Services Initialized
May 29 11:40:39 node1 clurgmgrd: <info> Logged in SG "usrm::manager"
May 29 11:40:39 node1 clurgmgrd: <info> Magma Event: Membership Change
May 29 11:40:39 node1 clurgmgrd: <info> State change: Local UP
May 29 11:40:39 node1 clurgmgrd: <info> State change: node3 UP
May 29 11:40:39 node1 clurgmgrd: <info> State change: node2 UP
May 29 11:40:50 node1 clurgmgrd: <notice> Starting stopped service sernode1
May 29 11:40:50 node1 clurgmgrd: <notice> Service sernode1 started
May 29 11:40:59 node1 clurgmgrd: : <warning> 192.168.11.31 is not configured
May 29 11:40:59 node1 clurgmgrd: <notice> status on ip "192.168.11.31" returned 1 (generic error)
May 29 11:40:59 node1 clurgmgrd: <notice> Stopping service sernode1
May 29 11:40:59 node1 clurgmgrd: <notice> Service sernode1 is recovering
May 29 11:40:59 node1 clurgmgrd: <notice> Recovering failed service sernode1
May 29 11:40:59 node1 clurgmgrd: <notice> Service sernode1 started
May 29 11:41:09 node1 clurgmgrd: : <warning> 192.168.11.31 is not configured
May 29 11:41:09 node1 clurgmgrd: <notice> status on ip "192.168.11.31" returned 1 (generic error)
May 29 11:41:09 node1 clurgmgrd: <notice> Stopping service sernode1
May 29 11:41:09 node1 clurgmgrd: <notice> Service sernode1 is recovering
May 29 11:41:09 node1 clurgmgrd: <notice> Recovering failed service sernode1
May 29 11:41:09 node1 clurgmgrd: <notice> Service sernode1 started
May 29 11:41:19 node1 clurgmgrd: : <warning> 192.168.11.31 is not configured
May 29 11:41:19 node1 clurgmgrd: <notice> status on ip "192.168.11.31" returned 1 (generic error)
May 29 11:41:19 node1 clurgmgrd: <notice> Stopping service sernode1
May 29 11:41:19 node1 clurgmgrd: <notice> Service sernode1 is recovering
May 29 11:41:19 node1 clurgmgrd: <notice> Recovering failed service sernode1
May 29 11:41:19 node1 clurgmgrd: <notice> Service sernode1 started
會重複的顯示<warning> 192.168.11.31 is not configured 。。為什麼這個IP起不來呢？
《解決方案》
你的IP都搞錯了。
<failoverdomainnode name="node1" priority="1"/>
<failoverdomainnode name="node2" priority="2"/>
<failoverdomainnode name="node3" priority="2"/>
這裡面是主機的IP地址，不是浮動地址！
<service autostart="1" domain="igridnodes1" name="sernode1">
<ip ref="192.168.11.31"/>
</service>
<service autostart="1" domain="igridnodes2" name="sernode2">
<ip ref="192.168.11.32"/>
</service>
<service autostart="1" domain="igridnodes3" name="sernode3">
<ip ref="192.168.11.33"/>
那裡面，應該是浮動地址！！！
《解決方案》
是的，集群對外地址，是浮動地址！
