Most things rescued, but not exactly everything…

The past couple of days, a lot of problems arose:

  1. I lost all of my files on my MIT account
  2. This meant that this blog was temporarily disabled
  3. My web browser crashed and would not come back up
  4. I forgot to get my passport out of the safe deposit box at the bank

The first two were completely recoverable. MIT keeps a backup in a readonly mirror in a subdirectory on my account. That meant that a recursive copy cured the first problem. The second problem was only partly remedied in that the files were physically there, but they would not manifest themselves as web pages. A quick email to the very helpful tech consultants gave me two magic spells I could use on the command line to reconnect the script daemon with the directories involved. And to make things more interesting, the blog software expected one of the directories to be writeable. A slight modification to the aforementioned magic spell cured that also.

The third problem was a mystery regarding the file There seemed to be multiple copies littered on my system from old installs that didn’t work (discussed earlier). If you recall, the only thing that worked was to compile my own Firefox from source code. This time I deleted all other installations that had that I could find.  Still the same error. So, still having the source, I compiled me a new copy, and hours later (on dual core), I installed a new copy on top of the old one and it worked like new.

The fourth problem is also solveable, but because I had to make an early morning flight tomorrow, I have to pay a penalty for missing the flight and waiting on standby.