I came in today and the CAD server was down. Nothing had really changed, but everything was apparently broken and everyone had pretty well left by the time I rolled in at 2. I came upon my frazzled boss trying to troubleshoot the problem.
Me: Can I try a few things?
Him: Do you know what you’re doing?
Me: Does the host do automatic backups?
Him: Yes. Weekly.
Me: Then I know enough.
–30 minutes later–
Me: It’s up.
Him: How did you do that?
Me: I turned it off and turned it back on again.
Him: I tried that!
Me: How many times did you do it?
Me: Ah… There’s the problem. I did it three times and between the 2nd and 3rd tries I turned off and on all the services manually. Remember *whisper* Windows Server 2003 can smell fear.
Insanity sometimes receives the quant definition of doing the same thing over and over an expecting different results. When it comes to near million-dollar pieces of software deployed across multiple servers with a Java frontend, it appears insanity is a requirement of operation.