🔧 run_command

The run_command method executes shell commands on the target host and returns the result. It can be used in recipes, definitions, resource blocks, and local_ruby_block contexts.

📤 Return Value

run_command returns a Specinfra::CommandResult with:

📖 Usage

📄 In a recipe

result = run_command('echo -n Hello')
result.stdout       # => "Hello"
result.exit_status  # => 0

🧩 In a definition

define :my_setup do
  result = run_command('cat /etc/os-release')
  if result.stdout.include?('Ubuntu')
    package 'ubuntu-keyring'
  end
end

🧱 In a resource block

execute 'conditional command' do
  result = run_command('cat /etc/hostname')
  command "echo 'Running on #{result.stdout.strip}'"
end

💻 In a local_ruby_block

local_ruby_block 'check kernel' do
  block do
    result = run_command('uname -r')
    Itamae.logger.info "Kernel: #{result.stdout.strip}"
  end
end

❌ Error Handling

By default, run_command raises an error if the command exits with a non-zero status. The error includes stdout and stderr for debugging.

⚖️ Comparison with execute Resource

Use run_command for querying state and making decisions. Use the execute resource for commands that change state.