Train - Transport Interface

Project State: Active
Issues Response SLA: 3 business days
Pull Request Response SLA: 3 business days

For more information on project states and SLAs, see this documentation.

Umbrella Project: Chef Foundation

Project State: Active

Issues Response Time Maximum: 14 days

Pull Request Response Time Maximum: 14 days

Train lets you talk to your local or remote operating systems and APIs with a unified interface.

It allows you to:

execute commands via run_command
interact with files via file
identify the target operating system via os
authenticate to API-based services and treat them like a platform

Train supports:

Local execution
SSH
WinRM
Docker and Podman
Mock (for testing and debugging)
AWS as an API
Azure as an API
VMware via PowerCLI
Habitat

Examples

Setup

Local

require 'train'
train = Train.create('local')

SSH

require 'train'
train = Train.create('ssh',
  host: '1.2.3.4', port: 22, user: 'root', key_files: '/vagrant')

If you don't specify the key_files and password options, SSH agent authentication will be attempted. For example:

require 'train'
train = Train.create('ssh', host: '1.2.3.4', port: 22, user: 'root')

SSH transport has an ssh_config_file option to set the SSH config file path. This is set by default to true to read the values from the default SSH config file path. For example, ~/.ssh/config, /etc/ssh_config, /etc/ssh/ssh_config. Precedence is given to the options set through the arguments.

WinRM

require 'train'
train = Train.create('winrm',
  host: '1.2.3.4', user: 'Administrator', password: '...', ssl: true, self_signed: true)

Docker

require 'train'
train = Train.create('docker', host: 'container_id...')

You can use user option to connect with privileged user on non root user images.

require 'train'
train = Train.create('docker', host: 'container_id...', user: 'root')

For Docker Desktop for Windows, you must enable the "Expose daemon on tcp://localhost:2375 without TLS" setting. To change the URL that train uses to connect to Docker, use the docker_url option:

require 'train'
train = Train.create('docker', host: 'container_id...', docker_url: 'tcp://localhost:1234')

Windows Docker containers require PowerShell, which is assumed to be installed as powershell.exe. If it is installed as pwsh, use the --shell-command option to specify pwsh as the shell.

Podman

require 'train'
train = Train.create('podman', host: 'container_id...')

require 'train'
train = Train.create('podman', host: 'container_id...', podman_url: 'tcp://localhost:1234')

To connect to a Podman container, the Podman API endpoint URL needs to set. It can be set through the podman_url option or else Train will check for the CONTAINER_HOST environment variable. If neither is defined then Train will try to connect to the default URL, that is unix:///run/user/UID/podman/podman.sock for non-root users and unix:///run/podman/podman.sock for the root user. Precedence is given to options set through the arguments.

AWS

To use AWS API authentication, setup an AWS client profile to store the Access Key ID and Secret Access Key.

require 'train'
train = Train.create('aws', region: 'us-east-2', profile: 'my-profile')

You may also use the standard AWS CLI environment variables, AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY and AWS_REGION.

require 'train'
train = Train.create('aws')

VMware

require 'train'
Train.create('vmware', viserver: '10.0.0.10', user: 'demouser', password: 'securepassword')

You may also use environment variables by setting VISERVER, VISERVER__USERNAME, and VISERVER_PASSWORD

require 'train'
Train.create('vmware')

Configuration

To get a list of available options for a plugin:

puts Train.options('ssh')

This will provide all configuration options:

{
  :host     => { :required => true},
  :port     => { :default  => 22, :required => true},
  :user     => { :default  => "root", :required => true},
  :keys     => { :default  => nil},
  :password => { :default  => nil},
  ...

Usage

# start or reuse a connection
conn = train.connection

# run a command on Linux/Unix/Mac
puts conn.run_command('whoami').stdout

# get OS info
puts conn.os[:family]
puts conn.os[:release]

# access files
puts conn.file('/proc/version').content

# access specific API client functionality
ec2_client = train.connection.aws_client(Aws::EC2::Client)
puts ec2_client.describe_instances

# close the connection
conn.close

Testing

We perform unit, integration and windows tests.

unit tests ensure the intended behaviour of the implementation
integration tests run against VMs and docker containers
windows tests that run on appveyor for windows integration tests

Mac/Linux

bundle exec ruby -W -Ilib:test/unit test/unit/extras/stat_test.rb

Windows

# run windows tests
bundle exec rake test:windows

# run single tests
bundle exec ruby -I .\test\windows\ .\test\windows\local_test.rb

Kudos and Contributors

Train is heavily based on the work of:

test-kitchen

by Fletcher Nichol and a great community of contributors
ohai

by Adam Jacob, Chef Software Inc. and a great community of contributors

We also want to thank halo who did a great contribution by handing over the train gem name.

Contributing

Fork it
Create your feature branch (git checkout -b my-new-feature)
Commit your changes (git commit -am 'Add some feature')
Push to the branch (git push origin my-new-feature)
Create new Pull Request

License

| Author: | Dominik Richter ([email protected])

| Author: | Christoph Hartmann ([email protected])

| License: | Apache License, Version 2.0

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

inspec/train

inspec

Reviews

Repository Details