On Wed, Nov 4, 2009 at 9:07 PM, Kwan Lowe <kwan.lowe@xxxxxxxxx> wrote: [...] > The problem with relying on documentation, even great documentation, > is that it puts the responsibility on the admin to follow. Now we all > know of the bad admins that can't follow directions, but in my > experience, the culprit is often the good admin who knows another way > to do something or has done it so often that he/she skips a critical > but simple step or just is so busy that things get missed. So given a > choice between great documentation and a good kickstart file, I'd > generally prefer the kickstart. In fact, given 100 builds, I would > prefer if they are all wrong in the identical way than have a perfect > build document that builds a perfect server but varies from instance > to instance. > In my experience it's the admins who think they are so good that they don't follow (or write) documents at all. Those are the ones who get in more trouble. The ones who have and follow documentation get a much higher success rate. Most of the time it's a choice between docs or no docs. Once you have them, getting someone to follow them is the easier job. (However, there is great variation in quality of documentation, and that can lead to the problem you describe, such as when the process is defined in loose terms instead of exact keystrokes) Once you have the docs, then you can build the kickstart file. They get you 80% of the way there. _______________________________________________ CentOS mailing list CentOS@xxxxxxxxxx http://lists.centos.org/mailman/listinfo/centos