Skip to content

Latest commit

 

History

History
114 lines (93 loc) · 4.95 KB

basic-chef-fluency-kata-cheatsheet.md

File metadata and controls

114 lines (93 loc) · 4.95 KB
Raw

Basic Chef Fluency Kata CHEAT SHEET

Please see the associated exercise kata for the Basic Chef Fluency Badge in the basic-chef-fluency-kata.md file.

Create a new repo for today's kata (or just branch the repo)

  • from code/chef/work create a new repo chef generate repo chefkata<#>
  • create /code/chef/work/chefkata<#>/.chef and copy the user pem inside the .chef directory
  • generate knife.rb "generate knife config" for that org to put in .chef folder, too

Create a new cookbook branch

  • from code/chef/work/chefkata<#>/cookbooks git clone https://github.com/anniehedgpeth/chefkata.git
  • from code/chef/work/chefkata<#>/cookbooks/chefkata create a branch, name it, and switch to it git checkout -b <branch-name>

Add the second cookbook for the runlist

  • add the ubuntu cookbook to that cookbooks directory and make sure it converges git clone https://github.com/anniehedgpeth/ubuntu-14-hardening.git berks install berks upload

Any time I change the cookbook, from that cookbook’s directory, I need to:

  • bump the version in metadata.rb or it won't upload the new one
  • (you would normally automate that in Jenkins) berks install berks upload

To bootstrap a new machine:

knife bootstrap chefkata<#>.southcentralus.cloudapp.azure.com -N chefkata<#> -r 'recipe[chefkata::default], recipe[ubuntu-14-hardening::default]' --ssh-user annie --sudo

knife bootstrap production<#>.southcentralus.cloudapp.azure.com -N chefkata<#>prod -r 'recipe[chefkata::default], recipe[ubuntu-14-hardening::default]' -E 'prod' --ssh-user annie --sudo

To converge on that node from here on out:

run sudo chef-client in an ssh session

To validate with InSpec Profile

First you have to add your private key to the local ssh (I don't know if it matters which directory you're in.) ssh-add ssh annie@chefkata<#>.southcentralus.cloudapp.azure.com inspec exec https://github.com/anniehedgpeth/chefkata_inspec -t ssh://annie@chefkata<#>.southcentralus.cloudapp.azure.com inspec exec https://github.com/anniehedgpeth/chefkata_inspec -t ssh://annie@production<#>.southcentralus.cloudapp.azure.com

To add run-list:

knife node run_list set chefkata<#> 'recipe[chefkata::default]'

To edit the run-list:

  • knife node show chefkata<#>
  • make sure the cookbook is uploaded
  • make sure it has a Berksfile
  • berks install
  • berks upload
  • knife node run_list add chefkata<#> 'ubuntu-14-hardening'

When adding an organization to manage.chef.io

  • add the org
  • reset user key generate knife config from UI copy the user key into .chef folder create a new org key and download knife.rb see if your user needs a new key knife node list

Search

knife search node "platform:ubuntu" knife search node "builder:Annie"

Data bags

  • shared data that your cookbooks can use
  • data_bags directory is in the chef_repo sibling to cookbooks directory
  • each data bag is a folder and each data_bag item is a .json file within that folder
  • the data_bag item is just a .json file of all of settings for that data_bag item
    • must include { "id":"<data_bag_item_name>" }
  • upload data_bag item to chef server so that you can use it in your cookbook
    • first create the bag on the server
    • knife data bag create website messages.json
    • run this from the top of the chef repo directory knife data bag from file BAG_NAME ITEM_NAME.json
    • knife data bag from file website messages.json
    • It's the same command to update the data bag if you edit it
  • Verify that it's there in UI
    • Policy > Data Bags > name
  • Verify that it's there from command line
  • knife data bag list

Using data bags in the recipe

  • First we access the data from that item messages = data_bag_item('website', 'messages')
  • Then we're calling the specific data element from that item message = messages['welcomeMessage']
  • Then call it in the recipe like content message

Using data bags in test kitchen

So kitchen can't look inside the "real" data bags directory, so you have to set up a dummy data bags directory just for test kitchen.

  • Create cookbooks/thiscookbook/test/integration/data_bags and copy your real data bag directory into that
  • Then edit the .kitchen.yml
suites:
 - name: default
   data_bags_path: "test/integration/data_bags"

Roles

Roles function just like data bags in the sense that they're sibling to the cookbooks directory, you have to upload them separately to the chef server, and they have their own attributes. They also have their own run-lists.

  • to upload to chef server knife role from file roles/security.json
  • to check knife role list
  • to add the role to the runlist knife node run_list add chefkata<#> 'role[security]'

Environments

knife environment create ENVIRONMENT_NAME -d DESCRIPTION knife environment from file FILE (options) knife environment edit ENVIRONMENT_NAME knife environment show ENVIRONMENT_NAME