Well it finally happened – my hard drive decided that it could no longer take all the beatings that I was putting it through. Clicks, whirs and spinning up and down were the order of the day so I set about to recover as much data from it as possible. Bearing in mind of course that it was split into more than one operating system. Damn these rubbishy small drives. At least I have backups of most everything.
Being that it would take at least a week for my favourite on-line retailer to get me a hard drive, and next day delivery was not really an option I went to the local Aldi, and bought a Western Digital hard drive (250gb, 8Mb Cache, 7200 RPM) for the princely sum of €125, which is a little over 50¢ a gigabyte, which is damned good value.
Thankfully there is no corruption of the partition table, so as I performed for the laptop upgrade, I used the trusty dd command.
dd if=/dev/hda of=/dev/hdb bs=512 conv=noerror,sync
The difference here is of course that I’ve added the essential conv=noerror,sync to the command line. This means that it ignores input/output errors, and the sync makes sure that when I/O errors happen that the destination drive gets padded with zeros. It’s progressing nicely, and I’ve only seen 72 i/o errors, and they look to be sequential on the drive. whatever was on the drive at that location has just taken one for the team 🙂
I’ll fire up partition magic once I’ve all the content transferred. 72 sectors of 512 bytes is not a huge amount of data to lose, and I have a feeling it’s in the pagefile.
Update 2005-08-24 12:51 – it seems to be working well so far.