I hate it when crap like this happens.
A few months ago, I built a new server (with hardware from ASA Computers, highly recommended--I own three of their boxes) to serve as a backup/spare for family.zawodny.com (the machine that hosts my blog and about 40 domains). The new machine was called friends.zawodny.com and had similar but slightly newer hardware (1GHz CPU, 512MB RAM, 2 80GB Maxtor disks in a RAID-1 setup).
Well, yesterday things went all bad. One of the disks appears to have died (or is dying). Several services died, including SSH, so I couldn't get in. Luckily it sits in a WCNet rack.
So I called to see what they could do. Couldn't get a console login. Damn. The only other option is to reboot. So we tried that. It didn't come up.
To make a long story short, the machine has two IDE cables. Each is the primary (master) and has one of the drives on it. However, after fiddling around, we determined that it won't boot from either one. This really sucks.
With one of the drives, it does the "01 01 01 01 01 ..." at bootup. With the other, it boots to "LI" and stops.
The reason this really bothers me is that I tested all the failure modes before I shipped it off to get racked. I could have sworn that is should still be working--unless both disks happened to magically die at the same time.
Now I either need to pul the spare 30GB disk off my shelf, build an OS and ship it out. Or I need to get the whole machine shipped back here. I have time for neither right now.
Anyway, the lesson is simple. Next time around I'll just spend the extra bucks and get a 3Ware controller and be done with it.
Damn you, Murphy! This couldn't have happned to the machine that is just a few miles away from me in California? Noooo. It had to be one of the machines I have in Ohio.