My system has been working fine, on potato, for a month or so. (Admittedly, I hadn't worked on getting sound working, and I had some tuning to do before I could say my printers worked flawlessly.... but I was able to get the work done, that I needed to get done.)
I tinkered with downloading some stuff from "testing"...stuff necessary to run the latest, balsa, I think...and then, on the debian-user mailing list, someone said something about a "bad lilo.conf".
"No problem," thought I, "I'll just check my lilo.conf and, for good measure, rerun lilo before I reboot." I guess I should have read the entire message the other person wrote.
In any case, my system is a bit unusual, in that I've got the main HD mounted off of a Promise UDMA/66 controller (so, when I installed, initally, I installed from HD, with the IDE patches already there.). Kernel 2.2.18pre21-ide, I think.
So, I rebooted and ended up with the dreaded "MBR L 01 01 01 01" situation.
I suspect there's something I can do, by moving my HD back to the mother-board, booting from a rescue CD (rescue root=/dev/hda2), and somehow manipulating the MBR, from there. True? How?
(I tried to mount each of the other file-systems, so I can run VI, and edit LILO.conf...but that's not working....)