That gives only the strings, I needed the entire text blocks to reconstruct the source files.

After about 4 weeks into my first programming job (~15 years ago), I lost all the source files in my cwd by mistakingly typing "rm *>o" instead of "rm *.o".
Of course at that time there were no tape no backups and my last commit was about a week or two earlier. I went to see the sysadmin and explained my situation.

In about 2 minutes he wrote a C program that opened and read from /dev/hda in blocks, looking for some variable/function names that I had provided him. This
yielded a big text file, and it took me about a day to untangle and reconstruct the various source files. I remember that had impressed the hell out of me.

One book that blew my mind was "The Data Warehouse Toolkit: The Complete Guide to Dimensional Modeling" by Ralph Kimball. It's pretty easy to read and offer some very interesting techniques for designing data warehouse schemas.

Here's how we do it:

- Run your services in a few vservers on the same physical server:
    * DNS + DHCP
    * mail
    * ftp
    * www
- Have a backup server where your stuff is rsynced daily. This allows for quick restores in case of disaster.

Vservers are great because they isolate you from the hardware. Server becomes too small? Buy another one, move your vservers to it and you're done. Need to upgrade a service? Copy the vserver, upgrade, test, swap it with the old one when you are set. It's a great advantage to be able to move stuff easily from one box to another.