What an annoying problem. Power in South Africa is so flaky now that you can expect a server to have died sometime over the weekend. You don't expect the server to have woken up, but be halted at a "strike F1 to continue" screen, which is precisely what I'm facing with the DELL PowerEdge 2900 machine. Philosophical question: If a network tree fails in a cluster, and no-one is around to press F1, is there a way to revive it?
The user manual for the DELL is here
(nothing useful in here)
In our case, the problem is that the DELL POWEREDGE 2900 spits out the "press F1 to continue" when there is an error. Since it always came up straight after the PERC RAID controller stuff, one presumed it was related, but it's not!
After going into the BIOS and checking out the stuff, I noticed that SATA PORT A is an "unknown device" and when I went into it, the setting was "AUTO" ... when I switched it OFF, and there was no longer a SATA PORT A device (only SATA PORT B device which was the CD-ROM), it no longer spewed forth said error. Turns out it was confuzzled! Now I can allow my machine to wake itself up without my interventions. Job done. :)