Does Windows Clustering provide true High Availability to Data Services?
I've used Server Groups before to provide HA with the understanding that it doesn't provide a fully automatic failover. By this I mean that if a node fails when a DataFlow is mid execution then we need to engineer some sort of recovery mechanism as there is the possibility that we have a partial data load.
I'm currently looking at Windows Clustering and I am not certain that it provides true HA either and so I'm not sure if there is any advantage over server groups. Am I right in thinking that with a windows cluster if a node fails when a dataflow is mid execution then we still need a recovery mechanism for any potential partial data loads or other data corruption?
Our target landscape is DS 4.2 sp4 on Windows