Setting up test nodes to use with Bolt
Difficulty: Basic
Time: Approximately 5 minutes
In this exercise you will create nodes that you can use to experiment with Bolt. You can also use existing nodes in your system if you prefer.
Prerequisites
To use an attached configuration file to set up test nodes, you must have one of the following installed on your machine:
Existing nodes
If you already have, or can easily launch, a few Linux or Windows nodes then you’re all set. These nodes must be accessible via SSH or WinRM; if you can access them via an SSH or WinRM client then Bolt can, too.
Using Vagrant
Note: These instructions assume that you are familiar with Vagrant and have a suitable hypervisor configured.
The attached Vagrantfile configures three CentOS 7 nodes and a Windows (Nano Server) node.
Save the following code as Vagrantfile or download the Vagrantfile attached to this exercise. To configure a different number of nodes, change the NODES environment variable.
# -*- mode: ruby -*-
# vi: set ft=ruby :
$nodes_count = 3
if ENV['NODES'].to_i > 0 && ENV['NODES']
$nodes_count = ENV['NODES'].to_i
end
Vagrant.configure('2') do |config|
config.vm.box = 'centos/7'
config.ssh.forward_agent = true
config.vm.network "private_network", type: "dhcp"
(1..$nodes_count).each do |i|
config.vm.define "node#{i}"
end
config.vm.define :windows do |windows|
windows.vm.box = "mwrock/WindowsNano"
windows.vm.guest = :windows
windows.vm.communicator = "winrm"
end
end
From the command line, ensure you’re in the directory where you stored the Vagrantfile file and enter vagrant up.
Generate the SSH configuration so Bolt knows how to authenticate with the SSH daemon. The following command will output the required details.
vagrant ssh-config
You can save that so it will be automatically picked up by most SSH clients, including Bolt. This uses the ability to specify hosts along with their connection details in a configuration file.
mkdir ~/.ssh
vagrant ssh-config | sed /StrictHostKeyChecking/d | sed /UserKnownHostsFile/d >> ~/.ssh/config
By saving this SSH configuration file, you can use the node name, rather than the IP address. When passing nodes to Bolt in the following exercises with Linux you will use --nodes node1,node2.
Make sure you can SSH into all of your nodes. If you’ve used the vagrant nodes before you may have to remove entries from ~/.ssh/known_hosts.
ssh node1
ssh node2
ssh node3
Using Docker
Using Docker we can quickly launch a number of ephemeral SSH servers. To make that even easier we’ll use Docker Compose.
Save the following code as docker-compose.yml or download the docker-compose.yml file attached to this exercise.
version: '3'
services:
ssh:
build: .
ports:
- 22
Save the following code as Dockerfile or download the Dockerfile attached to this exercise.
FROM rastasheep/ubuntu-sshd:16.04
RUN ln -s /usr/bin/python3 /usr/bin/python
Launch a single SSH server in the background: docker-compose up -d. To launch more SSH servers, run: docker-compose up --scale ssh=3 -d.
View a list of running containers: docker-compose ps. The result should be similar to:
Name Command State Ports
-------------------------------------------------------------------------
2acquiringnodes_ssh_1 /usr/sbin/sshd -D Up 0.0.0.0:32768->22/tcp
2acquiringnodes_ssh_2 /usr/sbin/sshd -D Up 0.0.0.0:32769->22/tcp
Note the Ports column. We are forwarding a local port to the SSH server running in the container. Using the example above, you can SSH to 127.0.0.1:32768.
If you have a local SSH client, test the connection. Change the port to one you get from running the docker-compose ps command. The image sets the username and password to root.
ssh root@127.0.0.1 -p 32768
Make sure you can log into all the nodes before moving on. You may have to remove some entries from ~/.ssh/known_hosts
When passing nodes to Bolt in the next section you will use --nodes 127.0.0.1:32768,127.0.0.1:32769, replacing the ports with those you see when you run the docker-compose ps command.
Next steps
Now that you have set up test nodes to use with Bolt you can move on to: