Causes of Failover??

[lug]

Hi All,

I have just had a weird experience with a few of my clusters...

I have 6 SQL2000 clusters, 3 of them failed all their resource groups onto NodeB at 12:30 yesterday, whereas the other 3 remained fine. I was thinking that maybe the switch that these clusters go into might need investigating but then i realised that the heartbeats via crossover cables are what determines whether or not each node is up and if to failover. What could have caused this then? It just seems a bit wierd that 3 happened at exactly the same time, so surely it must be a network issue somehow?

please reply if you have any ideas!

Thanks!

 

[victorv]

hi,

why network and not Power ? Also it may be common on them.

May be done a micro-interruption of Power ? causing
failover of 3 servers ?


bye

 

[crazyadmin]

I am not positive about this, but I do believe that it monitors both the heartbeat(private) and the public interfaces to make sure the server is "up" and accessible from the network.