I have setup a two node cluster on 2012. I am trying to use the generic application to fail fail over and to pick up where it left off i.e. Move what is in memory to the fail over server. I can not get this to work. When it fails over it restarts the program and anything in memory is lost. I read that the memory was supposed to be written to disk for the fail over.
To make it easier to see what is going on I created a program that just counts to a text file on the C drive. It over writes itself so you will only have the last number in the text file. If I run the program and then do a fail over after a few minutes (when the text file should be at 100 or so)it actually starts back at one again.
Any ideas?