>> I don't actually see why Live Upgrade is necessarily easier than an >> in-place upgrade. At least with a normal in-place upgrade you're >> booted off the new version in a known state, rather than an old >> version of Solaris in an unknown state. >Actually, this may not be the case. If upgrade fails for some reason you >may not be able to boot your system. With live upgrade you can boot back >always to the last known good running version of Solaris. It provides a >fallback mechanism.
True; but what I referred to is that the software environment in which "install" runs is better controlled in the "in-place upgrade" process. You boot the current OS and toolset from the net and then proceed to do the in-place upgrade. Liveupgrade, OTOH, complicates the test matrix because it runs on an OS in an unknown state. E.g., liveupgrade using $LC_* and that may cause certain interesting issues in some cases. >Also, even if upgrade succeeds, with development ongoing in OpenSolaris >there may be a bug that causes a panic on boot that just keeps on going. >This makes the system unusable as well if you can't even get to the >failsafe mode. Live upgrade at least allows you to get the system back >up and running. That is a risk you run, granted, but the recovery is not automatic which I believe is a failing of the process. (Surely if the system panics immediately on boot it should reboot into the old boot environment; does Caiman provision for that?) Casper
