Closed lottspot closed 7 years ago
I think the wise choice is to use an External Node Classifier to model our inventory. In addition to providing us with a powerful attribute-driven overview of the ecosystem we'll be managing, taking this approach will allow us to plug our inventory data into a configuration management tool of our choice once the time comes.
@ub1quit33 without doing more than skimming through their docs and some discussions, reclass smells better to me
This issue is pretty much obsolete
Before we can know what to do, we need to know what we have. We will need to gather a detailed inventory of what machines are currently active, what application stacks are installed on them, what provider they are hosted with, what web applications are hosted on them, and really every other detail we can get our hands on.
This task can be moved into the "ready" state once: