Hi all,
2 nodes clusters (2008r2 Sp1) with 2 NIC for nodes: LAN and HEARTBEAT. Heartbeat cables are connected through a single dedicated network switch.
Quorum type is "Node and Disk Majority"
This is what happens:
If I unplug both network cables from node 1, cluster disk resources go in "failed status" instead of moving to node 2. Only if I force a "bring online" cluster resources go to node 2.
If I try the same (unplugging both network cables from node 1) and at the same time unplugging Heartbeat cable from node 2, everything works fine (resources go to node 2)
If i manually move services from node 1 to node 2 , or if I shutdown node 1, everything goes to node 2 (so cluster seems to be OK).
It seems that ther' s something wrong about the HB network (wich is totally isolated, I tried many network switches and also using a FC to Ethernet media converter - I cannot use crossover cable, node are in 2 different server rooms)
Any idea?
Many thanks in advance