Iterate over json with inspec and only target specific entries for tests

See also questions close to this topic

• Whats missing on my Ruby 'Inverse Of' relationship

  I know this topic has been addressed, but I have been at this for 2 days and I'm just stuck. I know inverse of does not create a new query, so should I use another method?

  Question: How to set up an 'inverse of' with a has_one, belongs_to situation & same class..

  Explanation: A user 'has_one :spouse' and 'belongs_to :spouse_from'. They are inverse of each other. When a User signs up, they can invite their significant other. For Example

  1. user_a invites & creates user_b
  2. user_b.spouse_id is set to user_a.id
  3. In a separate method I want to be able to update like.. user_a.spouse_id = user_a.spouse.id

  The only association that works at this point is user_b.spouse.

   Class User
    has_one :spouse, class_name: 'User', foreign_key: :spouse_id, dependent: :nullify, inverse_of: :spouse_from
  
  
    belongs_to :spouse_from, class_name: 'User', foreign_key: :spouse_id, inverse_of: :spouse, optional: true
  
• Library functions in Inspec Test

  Team --

  I have a ruby library helper in which I have defined multiple functions/methods. How do I reference those in Chef Inspec tests?

  def interface_name
    # some code with logic
  end
  

  In my specific use case, I am checking to see if a custom networking device was specified as a JSON parameter, and if it was, I am validating it make sure its real (incase they misspelled it) and also gathering its IP and state and other data as reported by Ohai

  This is what I have so far, but I'm not sure if this is correct

    describe file('/path/to/custom_attributes.json') do
      it { should exist }
      unless json('/path/to/custom_attributes.json').empty? do
        its(['networking']['interface_name']) { should_not be_empty }
        interface_name = file(json('/path/to/custom_attributes.json').params['networking']['interface_name'])
      end
    end
  
    describe file('/etc/NetworkManager/system-connections/wired_connection') do
      unless interface_name.empty?
        its('content') { should_not match(/^interface-name=wl.*/mx) }
      end
      its('content') { should match(%r{ca-cert=/etc/ssl/certs/ca-certificates\.crt}mx) }
      its('content') { should match(/id=\.corp.*type=ethernet.*autoconnect-priority=100.*dns-search=corp\.domain.com.*/mx) }
    end
  end
  

  The problem / question is that if I gather the parameter directly from the JSON file, then I am bypassing all the validation logic that I'm doing in the library, which defeats the purpose. So, how do I get access to that library function/method in the Inspec test?

  For reference, here is the function:

  def interface_name
    file = '/path/to/custom_attributes.json'
    if File.exist?(file) && !File.stat(file).zero?
      attributes = JSON.parse(File.read(file))
      device_name = attributes['networking']['interface_name']
      if device_name && !device_name.empty? && networking_devices.include?(device_name)
        interface = device_name
        Chef::Log.info("Valid custom interface provided, using \"#{device_name}\".")
      else
        Chef::Log.debug("Invalid interface (\"#{device_name}\") provided. Valid options are: \"#{networking_devices.keys}\"")
        interface = nil
      end
    else
      Chef::Log.debug('No custom interface provided.')
    end
    interface
  rescue JSON::ParserError
    nil
  end
  
• Writing structured facts

  I've wrote some ruby code which will run on a linux server and return details about the server as a fact. It does this by connecting to amazon and retrieving some json (it runs two separate commands one to retrieve a list of disks - e.g /dev/sda1, /dev/xvdb and then it maps this to a volumeID via another query).

  I've made some small amendments to the output and added some values I'm interested in. The code runs multiple times and returns multiple hashes (one for each disk - maybe I should merge them?). Anyway here's an example of a server which has two disks below (this is just some debug output):

  Hash is {"/dev/sda1"=>{"disk_mapping"=>"", "is_lvm_partitioned"=>"false", "volumeid"=>"vol1234"}}.
  Hash is {"/dev/xvdb1"=>{"disk_mapping"=>"xvdb1", "is_lvm_partitioned"=>"true", "volumeid"=>"vol5678"}}.
  

  The next thing I want to to is turn this into a structured fact (with the devices: /dev/sda1, /dev/xvdb1 as the "keys"). Here's a rough idea of how I've done it (I've skipped a lot of the irrelevant code).

  json_string = {
    "#{path}" => {
    "disk_mapping"       => "#{disk_mapping}",
    "is_lvm_partitioned" => "#{is_lvm_partitioned}",
    "volumeid"           => "#{getVolumes}"
    }
  }.to_json
  hash = JSON.parse(json_string)
  if hash.is_a? Hash then
    debug_msg("Hash is #{hash}.")
    hash["#{path}"].each do |key, child|
      debug_msg("Setting key: #{key} child: #{child}.")
    end
  end
  

  I've never really wrote any ruby before so this is copied from multiple places, but besides aggregated facts I can't find a way to do this; I've tried to do something like this:

  Facter.add(:test["#{path}"]["#{key}"]) do
    setcode do
      #{child}
    end
  end
  

  So I guess in order I want to know:

  1. Should I merge the hash somehow? I originally assumed I did but found this incredibly hard due to not knowing how many hash's I'd have.
  2. Should I be using an aggregated fact or a "standard" one?
  3. How do I retain the structure of the hash and then call it with a single query (e.g facts test).
  4. Any examples which are similar to my code (the puppet ones I've found quite hard to follow).

  Here's what I'm looking for at the end:

  [root@blah ~]# facter test
  {
    /dev/sda1 => {
      disk_mapping => "",
      is_lvm_partitioned => "false",
      volumeid => "vol1234"
    },
    /dev/xvdb => {
      disk_mapping => "xvdb1",
      is_lvm_partitioned => "true",
      volumeid => "vol5678"
    }
  }
  

  Thanks.
• (Terraform) Error 400: Invalid request: instance name (pg_instance)., invalid

  On GCP, I'm trying to create a Cloud SQL instance with this Terraform code below:

  resource "google_sql_database_instance" "postgres" {
      name                = "pg_instance"
      database_version    = "POSTGRES_13"
      region              = "asia-northeast1"
      deletion_protection = false
  
      settings {
          tier      = "db-f1-micro"
          disk_size = 10
      }
  }
  
  resource "google_sql_user" "users" {
      name     = "postgres"
      instance = google_sql_database_instance.postgres.name
      password = "admin"
  }
  

  But I got this error:

  Error: Error, failed to create instance pg_instance: googleapi: Error 400: Invalid request: instance name (pg_instance)., invalid

  Are there any mistakes for my Terraform code?
• How does Kubernetes and Terraform work seamlessly together and what role do they each undertake?

  I am a bit confused about the individual roles of Kubernetes and Terraform when using them both on a project.

  Until very recently, I had a very clear understanding of both their purposes and everything made sense to me. But, then I heard in one of Nana's videos on Terraform, that Terraform was also very advanced in orchestration and I got confused.

  Here's my current understanding of both these tools:

  • Kubernetes: Orchestration software that controls many docker containers working together seamlessly. Kubernetes makes sure that new containers are deployed based on the desired infrastructure defined in configuration files (written with the help of a tool like Terraform, as IaC).

  • Terraform: Tool for provisioning, configuring, and managing infrastructure as IaC.

  So, when we say that Terraform is a good tool for orchestration, do we mean that it's a good tool for orchestrating infrastructure states or docker containers as well?

  I hope someone can clear that out for me!
• Automate Azure Devops (FTP Upload) and Git to upload on Remote Server

  The current setup is as below

  • Version Control - Git
  • Repos and Branch hosted on - Azure DevOps
  • Codebase - External server

  The dev team clones Azure Repo into local git project and any staged changes are committed via Git and pushed to specific branch of Azure DevOps. In this setup we would want to upload the changes to external FTP servers and avoid manual upload. Currently trying to use Azure Devops FTP Upload Task (https://docs.microsoft.com/en-us/azure/devops/pipelines/tasks/utility/ftp-upload?view=azure-devops), however facing issues; yaml script as below

  trigger:
  - main
  
  pool:
    vmImage: 'ubuntu-latest'
  
  variables:
    phpVersion: 7.4
    webAppName: 'Test Project'
    buildConfiguration: 'Release'
    vmImageName: 'ubuntu-latest'
  
  steps:
  
  - publish: $(System.DefaultWorkingDirectory)/AzureRepoName
    artifact: Test Project Deploy
  
  - task: FtpUpload@2
    displayName: 'FTP Upload'
    inputs:
      credentialsOption: inputs
      serverUrl: 'ftps://00.00.00.00:22'
      username: ftp-username
      password: ftp-password
      rootDirectory: '$(System.DefaultWorkingDirectory)/AzureRepoName'
      remoteDirectory: '/home/public_html'
      clean: false
      cleanContents: false
      preservePaths: true
      trustSSL: true
  

  PROBLEM

  Following errors occur when I commit (for test purposes) something.

  Starting: PublishPipelineArtifact
  ==============================================================================
  Task         : Publish Pipeline Artifacts
  Description  : Publish (upload) a file or directory as a named artifact for the current run
  Version      : 1.199.0
  Author       : Microsoft Corporation
  Help         : https://docs.microsoft.com/azure/devops/pipelines/tasks/utility/publish-pipeline-artifact
  ==============================================================================
  Artifact name input: Test Project Deploy
  ##[error]Path does not exist: /home/vsts/work/1/s/AzureRepoName
  Finishing: PublishPipelineArtifact
  

  I want to upload any staged change that is committed to main branch on Azure Devops to be automatically deploy on the remote FTP server

  Thanks
• how to pass parameters to template in chef?

  I just started to study chef. These days, I'm testing the sample of templates on -- https://docs.chef.io/resources/template/

  But I failed all the times ... Here's my code --

  1. I created a cookbook named sample, and created a recipe named default.rb

  file '/srv/www/htdocs/index.html' do
    content 'Hello World!'
  end
  
  include_recipe '::e'
  

  2. Then I created another recipe e.rb --

  default['authorization']['sudo']['groups'] = %w(sysadmin wheel admin)
  
  default['authorization']['sudo']['users'] = %w(jerry greg)
  
  template '/tmp/test.txt' do
    source 'test.txt.erb'
    mode '0440'
    owner 'root'
    group 'root'
    variables(
      sudoers_groups: node['authorization']['sudo']['groups'],
      sudoers_users: node['authorization']['sudo']['users']
    )
  end
  

  3. In this cookbook's templates folder, I created a erb file -- test.txt.erb

  Defaults        !lecture,tty_tickets,!fqdn
  root          ALL=(ALL) ALL
  <% @sudoers_users.each  do |user| -%>
  <%= user %>   ALL=(ALL) <%= "NOPASSWD:" if @passwordless %>ALL
  <% end -%>
  
  %sysadmin     ALL=(ALL) <%= "NOPASSWD:" if @passwordless %>ALL
  
  <% @sudoers_groups.each do |group| -%>
  
  <%= group %> ALL=(ALL) <%= "NOPASSWD:" if @passwordless %>ALL
  <% end -%>
  
  

  4. Then after kick off "chef-client" , error message shows --

  [2022-05-06T18:01:07+08:00] FATAL: NameError: undefined local variable or method `default' for cookbook: sample, recipe: e :Chef::Recipe
  

  5. Since error shows can't find the variable named 'default' and in that sample -- it's using node['authorization']['sudo']['groups'] to pass parameters to sudoers_groups, I think the e.rb maybe should be this --

  node['authorization']['sudo']['groups'] = %w(sysadmin wheel admin)
  node['authorization']['sudo']['users'] = %w(jerry greg)
  
  template '/tmp/test.txt' do
    source 'test.txt.erb'
    mode '0440'
    owner 'root'
    group 'root'
    variables(
      sudoers_groups: node['authorization']['sudo']['groups'],
      sudoers_users: node['authorization']['sudo']['users']
    )
  end
  

  6. But it still fails --

  [2022-05-06T17:39:38+08:00] FATAL: NoMethodError: undefined method `[]' for nil:NilClass
  

  I really messed up by this official sample. Please kind help me, Thanks in advance for any ideas.

  Regards Eisen
• Nginx chef docker

  I am new to docker concept. For an application I need to create nginx container. The nginx configuration is in chef cookbook, hence I have example.conf.erb file and default.rb (containing setup nginx file) in my chef/cookbook/ directory . I am not sure how to containerise. I copied .conf.erb to /etc/nginx/conf.d/example.conf.erb. I am not sure what else to do. I am confused and no resource online need help immediate default.rb :

     include_recipe 'nginx_ldap_auth'
     include_recipe 'nginx'
     template 'nginx config' do
     source 'example.conf.erb'
     path '/etc/nginx/conf.d/example.conf.erb'
  
     owner 'root'
     group 'root'
     mode ''
     variables({'environment variables'})
     notifies :restart, 'service[nginx])'
  

  My Dockerfile currently look like this:

  FROM nginx:alpine
  
  COPY default.conf /etc/nginx/example.conf.erb/
  

  I am not sure if I need docker-compose. Apart from Dockerfile there is nothing much I have created. Please guide
• Can inspec use the output of a command to trigger and only_if skip of a control?

  I am trying to set something up like this:

    only_if('physical device') do
      command('hostnamectl') do
        its('stdout') { should match /Chassis: desktop/ }
      end
    end
    describe command('ifquery --list') do
      its ('stdout') { should eq "lo\nbond0\neth0\neth1\neth2\n" }
    end
  

  because I'm hitting up a group of physical and virtual machines using the same control and I want to check and see that all the physical devices have 3 ethernet interfaces and all the virtual devices has 1.

  There's only a handful of places you can go in linux to tell if you're running a vm or baremetal, but they almost all require grepping something out of a file, so I need to check stdout somehow.

  So, what I'm wondering is, how do I use stdout with a command to skip something?