"Bloody hell, it sounds like they were very badly managed if it was only three servers."
Actaully it was the fault of the previous Linux Admin. He configured each Linux Server his way and documented nothing. As we are a 24/7 environment and can afford less than zero downtime it wasn't as easy as just rebuilding the boxes, they had to be fixed in place.
I agree that it took a long time but with no coherent standards (something Windows does have) and the Linux zealots penchant for tinkering the boxes were a nightmare.
When I did impose standards (and got alot of flack from the new unix admin for it). We went from downtime once per week to almost no downtime for six months.
However the Unix admin still gives me grief on a daily basis and sometimes I feel like cutting off his hands so he doesn't "fix" the exisitng implementation.
Heh, I can sympathize with that. Our sysadmin has to get my authorization to touch any of the servers that are under my authority. I'm a firm believer in the "if it ain't broke, don't fix it" philosophy for my servers and I smack down requests to tinker all the time. If a server has perfect uptime (and many of ours have been up for a year or two), it is hard to justify messing with them.
I think part of the problem is that our primary Unix sysadmin is bored. Nothing ever breaks, and so he needs something to do.
There's your problem. Anyone who likes to tinker needs to get their own box and play around to their heart's content. I've generally found that people who like to tinker with computers have little interest in actually getting them to work. It's a game to them, like a jigsaw puzzle.