Fail to ping VIP after one node shutdown
Hi,
I have installed RAC 11gR2 on Windows 2008 R2 on 2 server nodes. I don't have DNS server and so SCAN IP is hardcoded in hosts file. Everything works fine including IPs, SCAN, database connection if both servers are up and running. Windows Firewall are turned off in both servers.
Problems occur when one node is shutdown. I have checked that all VIPs for this node have already failover to the survival node (by checking with ipconfig). I have also checked with crsctl and confirmed that the VIP resource is changed to INTERMEDIATE and FAILED OVER. However, they cannot be "ping" from client (Windows XP SP2) but can be "ping" in the survival node.
When the shutdown node is power up again, the VIPs fail back and then all of them can be "ping" again.
The problems get worst for SCAN IP as it does not fail back. The SCAN IP keeps fail in "ping" from client until it is "switch" back to the "original" node.
It seems that somehow the server node don't listen to VIPs failover from other nodes.
Anyone can help?