I am really drilling down on Configuration Drift.
I am of the general opinion that most CD is caused by good intentions & that a little diversity in the environment is likely good for stability [a kind of Simian Army if you will].
In any event, one of the serious struggles about have a less then zero tolerance for CD Variance is that it leads to challenging provisioning issues–defining the Gold Standard from a pile of snow flakes can be a very non exact science & documenting all the tunings that went into a system is proving to be a very difficult archival challenge.
One of the resources I have started to use to uncover some of this is the great Stewart Brand‘s How Buildings Learn. Like so many other areas, Brand’s work has deep resonance for many of the challenges I am facing and provides very concrete guidance on the factors that should be used to dig my way out.
Leave a comment
No comments yet.