Hard drive failures suck. A couple weeks ago, I noticed this server box was starting to make some odd squeaking noises, and I was reasonably sure that there weren’t any mice trapped inside. I checked the SMART stats, and sure enough, bad sectors had started appearing on one of the drives. In the days after that, the number kept steadily rising, and just trying to access some directories started spitting out I/O errors.
The data was safe, as it’s pretty robustly backed up, but it presented me with a bit of a dilemma: do I just replace the drive, or do I consider replacing the whole system, as it’s now around five years old?
In the end, I wound up kind-of upgrading it. I kept the system, but upgraded it to 4GB of memory cannibalized from my recently replaced gaming box, and replaced both of the hard drives in it instead of just the failing one. Instead of two 1TB drives, it now has one 240GB SSD, and one 4TB hard drive. The SSD makes for a nice and snappy boot and home drive, but it obviously can’t hold much, so any large collections of data will go on the 4TB drive, with symlinks all over the place as needed. The CPU’s still adequate for its relatively light needs, so with these upgrades, this system should hopefully last for quite a while longer.
Now how long is it going to take to restore all those files…