Why I'm not really interested in Solaris's Live Upgrade stuff
Solaris Live Upgrade is one of those things that I would like to love, but I can't seem to get interesting in, and here's why.
First, as I've written before, we don't patch our Solaris machines. They're only accessible internally and they're effectively black-box appliances; there's very little enthusiasm here for patching a working appliance.
When we do patch our Solaris machines, we never need to roll back patches because we test them first in a test environment. We have no choice about this; we cannot possibly deploy an untested patch into a production environment, fast reversion or no. (Sadly our experience with Solaris patches is that we definitely do have to test them.)
(Also, we don't patch machines in single-user mode regardless of what the Solaris patches claim to require; in fact we just apply all of our patches through pca.)
Solaris LU couldn't save us from a server downtime either. Even with LU you need to reboot the machine in order to activate your new patches, and rebooting a fileserver always requires a formal downtime (which is one reason that we don't have very much of an urge to patch fileservers unless it is really important).
(I've written before about how we do fast OS upgrades. LU might make those simpler, but it does so at the cost of making an upgraded machine subtly different from a machine that we (re)install from scratch, unless we make (re)installs take even longer than they already do.)
I admit that I would feel more interested in just exploring Solaris LU if it hadn't already hung on me once or twice. I have a very low tolerance for bad behavior from tools like Live Upgrade; any bugs are generally enough to make me give up on the tool entirely (for reasons that don't fit in this entry).