a charmed what, now?

We had a power glitch at the hosting facility for my machine the other day, and after that we had reports of single-bit errors coming from one drive in the RAID-1 pair during read. (Tip ‘o the hat to Nathaniel Smith for being brave enough to call me at 06:15 with the news.)

Yesterday, I bought a new drive to replace the erroring one — which was, itself, not all that old — and things seemed to be OK. Today, though, we’re seeing the same result, with the new drive, so I’m going to have to go back in and move it to the known-good IDE channel or something.

I hope it takes, because I’m leaving in less than 36 hours for Europe, and it’ll be really hard to do work on the box from there.

one comment to “a charmed what, now?”

  1. Amos Hayes
    entered 19 May 2005 @ 6:04 pm

    Were they Maxtors? I had a pair of 6 month old DiamondMax 9 drives fail simultaneously in a RAID-1 config. I tested the drives independantly on seperate hardware and confirmed it was the drives and not the controller. When I dug into it online, I found cases of people having bought 4 of these drives only to lose #2 and #3 while #1 was in transit to the manufacturer to be replaced.

    Reminds me of the old DeskStar drives. Anyway… Western Digital seems to have the best reputation for reliability by a significant margin.

    Just thought I would pass that info along in case you were inclined (as I was) to rule out a pair of drives failing at the same time.