The reason you don't seen the printers online is because when the resource fails over to the other node, it doesn't know where the queue folder is and is most likely pulling the local queue from the C: drive, which is the default location for the spooler, and thus won't see the printers tied to the cluster resource that's migrating over. Look at the properties of the print server and make sure you relocate the queue from the local C: drive to the shared storage and set the location accordingly.
Anytime you cluster anything, the data that you're trying to make highly available, such as DHCP database, print spooler queue, SQL database, etc, must reside on shared storage.
This is a common mistake made by many, even me, so don't feel bad. We get so caught up in the complex stuff we forget to check for the obvious.
DOH! We all have our Homer Simpson moments :^0