I returned from my travels, sat down at the desk yesterday morning, fired up my email program, and — ffftt!!! — encountered one of those awful Windows system messages (something to the effect that the system had been unable to write back to the disk and it was very sorry but some data had been lost) that can mean only one thing: hardware failure. I took a breath, ran the disk utility, and got another message informing me that there was insufficient disk space to fix the bad clusters, which made no sense, since there was plenty of empty space on the drive — unless, like, massive gigabytes’-worth of clusters had gone south.
The bad news was, this was the big 500 GB drive I use to store basically everything. (Another drive is for the system and applications.) The good news is, I’ve become religious about backups, I have copies of all my important data, including online backups; but it’s scattered about.
I can recall many hours spent hostage to CHKDSK back in the late ’80s, fixing my own or friends’ wayward drives, watching patiently as the little grid filled out with marks for good and bad clusters. This was when the disks held 10 or 20 megabytes. Then in the mid 90s I wrestled with bad Mac disks on the cruddy machines Salon was trying to make do with. But since then, either the quality’s gotten a lot better or I’ve had a good run of luck. This is the first disk disaster I’ve had in a decade.
I took another breath and began poking at the drive with various tools and utilities. Sometimes the system could see it, sometimes it couldn’t. Then, at the lowest ebb of my fight, my system told me that there was zero Kb of data in zero files on the drive, and that the drive, in fact, could not be examined at all because “the type of the file system is RAW.”
Raw? As in, not cooked? By now I was doing all my research on the old laptop. Google told me that this error message indicates that a drive is close to unformatted — in, as it were, the state of nature. Only I knew there was tons of data on the disk, and it hadn’t been erased, and it didn’t seem like the disk had massively failed (a la “head crash”).
A couple of hours digging through the tech forums led me to a fine utility called GetDataBack. Several hours later, this program had painstakingly reassembled the file-system tables of my drive and enabled me to ferry my precious information back to safety in (roughly) one piece rather than having to reassemble it from patchwork backups.
This morning, I purchased and installed a new drive. Spent the rest of the day restoring files. Now I’m back. Phew. Only two days lost.
My motherboard fried during the middle of my labors on Dreaming in Code. So maybe I get one technical disaster per book project, and now the one for my new book is behind me. (I also notice that the previous disaster also occurred after a week away. I wonder whether it’s the temperature change of sitting unpowered after weeks of constant use that’s stressing the gear.)
If you’ve read this far, you don’t need me to remind you of the moral of this tale: you must have a backup plan.
Post Revisions:
There are no revisions for this post.