- From: Dan Adams <
>
- To: <
>
- Subject: [chef] Re: Re: Re: Re: Re: Chef in "Pure Infrastructure-As-Code" Mode?
- Date: Sun, 24 Jun 2012 21:10:20 +0100
- Mail-reply-to: <
>
On 24.06.2012 19:07, Torben Knerr wrote:
Oh, is there a standard place for nodes in the chef repo structure?
Didn't know about that, so I added the $CHEF_REPO/nodes directory and
placed the node info as .json files (like what you get with `knife
node show -fjson`) there...
I don't know how standard it is, but we use /nodes to save node
definitions in .rb format. You are able to manage nodes as .rb files,
edit them and store them in git, and commit them and then you can then
knife import them or whatever, all the same as you do with the
cookbooks, environments etc. I'm sure some people use json, but its far
easier as a human to work directly with the rb file and import from
there.
I'm aware the documentation has something like 4-5 competing workflows
lkisted for working with nodes, which is a bit confusing, and none of
them are explained in full detail. But I find the above the easiest by
far.
Cheers Dan
Archive powered by MHonArc 2.6.16.