UPDATE!

This repository is no longer in use by the Puppet SE Team, and will be official Archived on Friday, July 1st. It will then be deleted on Friday, July 15th. If you are using any of the included code in any way, please take efforts to preserver your access to the code at your earlier convenience.

Thanks! - The Puppetlabs SE Team

Bolt Workshop automated deployment of PE with Bolt

To create a Bolt Workshop environment with this module, do the following:

Add the following to your Puppetfile for Bolt:

# Modules from the Puppet Forge
mod 'puppetlabs-yumrepo_core', '1.0.3'

# Modules from Git
mod 'workshop_deploy',
  :git => 'https://github.com/puppetlabs-seteam/workshop_deploy.git',
  :ref => 'master'

and run bolt puppetfile install to sync the modules.

If you are using Hydra to stand up a Puppet Enterprise environment, do the following:

Make sure you have deployed a Hydra environment with cd4pe: false, agents: false and student_machine_count set to the amount of students you have for the workshop. Add 1 more to have your own student machine to do demos with.

Retrieve the Bolt Inventory file for your Hydra environment from the #team-svcsport-chatter channel in Slack, and open it in an editor.

• Scroll down to the alllinux section and locate the name of your PE master (should be <branchname>-master0.classroom.puppet.com)
• Copy the name of the PE master you found
• Scroll up to the master section and replace the internal IP address (a 10.x.x.x IP) with the PE master name that you copied
• Save the file

Now make sure you have placed the training.pem file in ~/.ssh/training.pem before you run the Bolt plan.

Run the workshop_deploy::hydra plan as follows:

bolt plan run workshop_deploy::hydra [github_user=[github user] github_pwd=[github password]] [github_token=[github token]] demo_name=[Your Hydra demo branch] --inventoryfile [Bolt inventoryfile you got from #team-svcsport-chatter]

If your GitHub user has 2FA enabled you can not use your GitHub user together with the GitHub password. You need a GitHub token instead. The token needs to be created on the GitHub website. Move to Settings -> Developer settings -> Personal acess token and create a token with the following scopes: "repo", "admin:repo_hook" "read:user" and "delete_repo".

Example with Github username & password:

bolt plan run workshop_deploy::hydra github_user=user1 github_pwd='password' demo_name=user1-1 --inventoryfile ~/Downloads/user1-1-bolt-inventory.yaml

Example with GitHub token:

bolt plan run workshop_deploy::hydra github_token='abcdefghijklmn1234567' demo_name=user1-1 --inventoryfile ~/Downloads/user1-1-bolt-inventory.yaml

Optionally, you can move the parameters into a Bolt params file, which makes it easier to preconfigure support for multiple Hydra environments. A Bolt params file is in JSON format and looks like this for the workshop_deploy::hydra Plan:

{
"github_user": "user1",
"github_pwd": 'password',
"demo_name": "user1-1"
}

or

{
"github_token": "abcdefghijklmn1234567",
"demo_name": "user1-1"
}

To call the Bolt Plan with the params file (say the filename is user1-1-params.json), do this:

bolt plan run workshop_deploy::hydra --params @user1-1-params.json --inventoryfile ~/Downloads/user1-1-bolt-inventory.yaml

If you want to use AWSKit to stand up a Puppet Enterprise environment, do the following:

It is highly recommended to use the puppetseteam/puppet-aws-kit container image for AWSKit, see the Readme on the AWSKit repo.

Make sure you have previously configured AWSKit to deploy a CentOS image for the Bolt workshop that uses a fixed Elastic IP. You'll need the following data in Hiera for your desired AWS region:

awskit::create_bolt_workshop_targets::master_ip: '<available elastic ip>'
awskit::host_config:
  <your AWSKit username>-awskit-boltws-master:
    public_ip: '<available elastic ip>'

for example:

awskit::create_bolt_workshop_targets::master_ip: '35.180.221.85'
awskit::host_config:
  kevin-awskit-boltws-master:
    public_ip: '35.180.221.85'

Then run the Bolt Plan like this:

bolt plan run workshop_deploy bastion=[true|false] awsregion=[region] awsuser=[AWS user] [github_user=[github user] github_pwd=[github password]] [github_token=[github token]] --nodes [AWS public IP for PE master] --user centos --private-key [private key for SSH] --no-host-key-check

If your GitHub user has 2FA enabled you can not use your GitHub user together with the GitHub password. You need a GitHub token instead. The token needs to be created on the GitHub website. Move to Settings -> Developer settings -> Personal acess token and create a token with the following scopes: "repo", "admin:repo_hook" "read:user" and "delete_repo".

Example with Github username & password:

bolt plan run workshop_deploy bastion=true awsregion=eu-west-3 awsuser=user1 github_user=user1 github_pwd='password' --nodes 35.180.221.85 --user centos --private-key ./user1.key-eu-west-3.pem --no-host-key-check

Example with Github token:

bolt plan run workshop_deploy bastion=true awsregion=eu-west-3 awsuser=user1 github_token='abcdefghijklmn1234567' --nodes 35.180.221.85 --user centos --private-key ./user1.key-eu-west-3.pem --no-host-key-check

The parameters have the following meaning:

• bastion: Set to true for Bastion accounts in AWS, so to false for traditional accounts in AWS
• aws_region: The AWS region (in Hiera) that awskit must use for deployment
• aws_user: The AWS user (in Hiera) that awskit must use for deployment
• github_user: Your GitHub user account that has access to github.com/puppetlabs-seteam
• github_pwd: Your GitHub user account password
• github_token: Your Github personal access token, which can be used instead of github_user and github_pwd

You need the --user centos option for Bolt since CentOS instances in AWS must be accessed via the centos user. The plan is configured to elevate to root for the steps on the AWS instance, and therefor must not be called with the --run-as parameter.

Optionally, you can move the parameters into a Bolt params file, which makes it easier to preconfigure support for multiple regions. A Bolt params file is in JSON format and looks like this for the workshop_deploy Plan:

{
"bastion": true,
"awsregion": "eu-west-3",
"awsuser": "kevin",
"github_user": "kreeuwijk",
"github_pwd": 'password',
"nodes": "35.180.221.85"
}

or

{
"bastion": true,
"awsregion": "eu-west-3",
"awsuser": "kevin",
"github_token": "abcdefghijklmn1234567",
"nodes": "35.180.221.85"
}

To call the Bolt Plan with the params file (say the filename is eu-west3-params.json), do this:

bolt plan run workshop_deploy --params @eu-west3-params.json --user centos --private-key ./user1.key-eu-west-3.pem --no-host-key-check

Remember that you can eliminate the need to specify the connection info everytime, by adding the node to your inventory.yaml for Bolt!

Deploying Bolt targets when using AWSKit

To deploy targets, use the workshop_deploy::targets plan. This plan will:

• Deploy 1 Linux and 1 Windows target for the Bolt instructor
• Deploy 1 Linux and 1 Windows target for the amount of students specified

bolt plan run workshop_deploy::targets bastion=[true|false] awsregion=[region] awsuser=[AWS user] amount=[number of students]

for example:

bolt plan run workshop_deploy::targets bastion=true awsregion=eu-west-3 awsuser=kevin amount=5

The command above results in a total of 12 targets:

• 5 Linux targets for the students
• 5 Windows targets for the students
• 1 Linux and 1 Windows target for the Bolt instructor

Make sure the AWS region you select has a high enough limit for simultaneous running instances! Check here for your current limits: https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/ec2-resource-limits.html

The plan uses AWSkit to create the instances, and AWSkit has been configured to expect a 'bolt_ws_key' as the key for any region that you specify. This key may not have been setup yet for your region. To verify, look for a 'bolt_ws_key' entry in the 'Key Pairs' section in your AWS console.

If the key entry is missing, add it as follows:

• Download the private key from http://bit.ly/B0ltk3y
• Generate the public key from the private key by running ssh-keygen -y -f bolt_ws_key.pem
• Under 'Key Pairs' in the AWS console, click 'Import Key Pair', set the name to bolt_ws_key and paste the public key value