Daniel Schaaff

Adventures in Ruby

I’m learning ruby. Finding time to work towards this goal is proving difficult but I’m forcing myself to use ruby wherever possible to aid in my learning. I’ll be putting some of my lame code on here to chronicle my learning and hopefully get some feedback on how I can improve things. I recently came across a good opportunity when I needed to generate a list of nodes to use with the puppet catalog preview tool

I wanted to get a full picture of my infrastructure and represent all our nodes in the report output without having to manually type a large node list. Puppet already has all my node names so I just need to extract them . My first step was to query the nodes endpoint in puppetdb for all nodes and pipe it into a file.

curl http://puppetdb.example.com:8080/v3/nodes/ > nodesout.txt

The output of this is json with an array of hashes.

[{
"name" : "server2.example.com",
"deactivated" : null,
"catalog_timestamp" : "2016-11-28T19:28:14.828Z",
"facts_timestamp" : "2016-11-28T19:28:12.112Z",
"report_timestamp" : "2016-11-28T19:28:13.443Z"
},{
"name" : "server.example.com",
"deactivated" : null,
"catalog_timestamp" : "2016-11-28T19:28:14.828Z",
"facts_timestamp" : "2016-11-28T19:28:12.112Z",
"report_timestamp" : "2016-11-28T19:28:13.443Z"
}]

I only want the name of each node so I need to parse that out. It was a great opportunity to open pry and get some practice!

[1] pry(main)> require 'json'
=> true
[2] pry(main)> file = File.read('nodesout.txt')
pry(main)> data_hash = JSON.parse(file)
=> [{"name"=>"server.example.com",
"deactivated"=>nil,
"catalog_timestamp"=>"2016-11-29T00:37:03.202Z",
"facts_timestamp"=>"2016-11-29T00:37:00.972Z",
"report_timestamp"=>"2016-11-29T00:36:38.679Z"},
{"name"=>"server2.example.com",
"deactivated"=>nil,
"catalog_timestamp"=>"2016-11-29T00:37:03.202Z",
"facts_timestamp"=>"2016-11-29T00:37:00.972Z",
"report_timestamp"=>"2016-11-29T00:36:38.679Z"}]
[4] pry(main)> data_hash.class
=> Array
[5] pry(main)> def list_nodes(input)
[5] pry(main)*   File.open('nodes_out.txt', 'a') do |f|
[5] pry(main)*     input.each do |i|
[5] pry(main)*       f.puts i["name"]
[5] pry(main)*     end
[5] pry(main)*   end
[5] pry(main)* end
=> :list_nodes
[6] pry(main)> list_nodes(data_hash)
[7] pry(main)> exit

I now have the list of nodes I was looking for!

$ cat nodes_out.txt
server.example.com
server2.example.com

This accomplished what I needed and also saved me a lot of time (like putting the puppetdb query directly in the ruby stuff). I’m certain there may be a cleaner way to do this, but that’s what learning is for!

comments powered by Disqus