IPMP/HA: most common configuration

Hi,
based on your experience what is the most common IPMP configuration on a two nodes cluster for HA failover / scalable services?

...active-active, active-standby with probe-based failure detection etc.

Thanks

this depends on the service and the requirements... but i think the most used config is active-standy.

ok ...but to generalize and simplify a lot, is correct to say:

for HA failover services: active[+test address] - standby
for HA scalable services: active[+test address] - active[+test address]

no, it isn't.