Integrating vRealize Automation 7.3 with Puppet Enterprise

Starting with vRealize Automation 7.3 (vRA) there is native integration with Puppet. This means that there are now Puppet object within the vRA interface that you can drag and drop onto your blueprints as well as select Puppet constructs such as roles from dynamic drop down lists while building and deploying vRA blueprints. You can read the annoucement from Puppet here. Puppet has also provided some starter content to make getting up and running quick.

Versions used

vRealize 7.3
vCenter 6.5.0.5300
Puppet vRealize Orchestrator (vRO) plug-in 3.0.0.31
Puppet 2017.2.2
CentOS Minimal 7.3

Install Puppet Enterprise Trial

We need to get Puppet Enterprise so go to https://puppet.com/download-puppet-enterprise, grab it for your platform and copy it to your Puppet machine.

If you don’t want to worry about the firewall, you can perform the following on the Puppet server.

systemctl stop firewalld
systemctl mask firewalld
yum install -y iptables-services
systemctl enable iptables
iptables -F
service iptables save
reboot

Install Puppet by running the following commands:

Copy the Puppet files to the Puppet server
tar xzvf puppet-enterprise-2017.2.2-el-7-x86_64.tar.gz
cd puppet-enterprise-2017.2.2-el-7-x86_64
./puppet-enterprise-installer
<Enter>

You’ll then see the following:

In a minute you’ll be prompted to continue the installation through a browser:

Once you hit the Puppet server you’ll see:

If you can’t access the Puppet install page, you may need to drop your firewall or open port 3000. You can quickly drop the firewall by running iptables -F

On the following page I made these selections, but you’ll need to specifiy whatever makes sense for your environment:

Puppet master FQDN: puppet.vmware.local
Puppet master DNS alias: puppet
Install PostgreSQL on the Puppet master host for me
Console ‘admin’ password: ****

On the next pages, select Continue and then Deploy now. If everything goes well, you’ll see:

Were not going to do too much with the console in this guide but feel free to explore it.

Install the Puppet Content Starter Pack

The Puppet Content Starter Pack will allow us to get up and running with vRA and Puppet very quickly as it will configure a number of settings and create roles that we will consume in vRA. The Starter Pack is hosted on Github so we will need to install git if it’s not installed:

yum -y install git

Then run the following commands:

git clone https://github.com/puppetlabs/puppet-vro-starter_content.git
cd puppet-vro-starter_content
bash scripts/vra_nc_setup.sh
/opt/puppetlabs/bin/puppet agent -t

At the end you should get:

Install the Puppet vRO plug-in

You can grab the plug-in here. I’m using the embedded version of vRO in the vRA appliance. We need to access the control center, which is stopped by default, so access the console of the vRA appliance and run:

service vco-configurator start

If you haven’t done this before, the above is not a typo. vRO was renamed from vCO and the service still has the old name. Open a web browser and go to the address of your vRA appliance (not the vRA application where you login) and select vRealize Orchestrator Control Center:

If you haven’t changed the logon, the username should be administrator@vsphere.local. Scroll down until you see Plug-ins:

Select Browse and browse to where you downloaded the Puppet vRO plug-in:

Select Install:

Should now see the plug-in:

I didn’t have to restart to get the plug-in to work but if you need to, you can run the following from the vRA appliance:

service vco-server restart

Add a vCenter instance in vRO

You need to have a vCenter instance in vRO for the vCenters that you plan to deploy machines to and have Puppet manage them. You need to do this because when you deploy a machine with vRA the Puppet integration will kick off a workflow to find the machine in vCenter so it can install the Puppet agent.

Open the vRO client
Browse to Library > vCenter > Configuration
Run Add a vCenter Server instance

Here is what my workflow run looked like:

If the workflow executes successfully, select the Inventory tab and verify that you can browse your vCenter(s) inventory.

Create the vRA Puppet endpoint

We can create our Puppet endpoint in vRA by performing the following:

Log into the vRA web page
Navigate to Administration tab > vRO Configuration > Endpoints
Select New
Select the Puppet plug-in
Fill in the Puppet server details (password is puppetlabs)

Create vRA Blueprint

For this section, I’m going to assume that you know how to create blueprints and manage catalog items and entitlements so I’m not going to show steps for each of these.

Create a new blueprint
Drag a vSphere machine onto the canvas
Select Configuration Management and drag the Puppet icon onto the vSphere machine

The design canvas should look like this:

Select Puppet_1 and then the Server tab.

Select your Puppet Master
Select the Puppet environment
Select the three dots to bring up the role selection menu
Select role::linux_webserver
Enter S3cr3tP@ssw0rd! for the Shared secret. This is used to auto-sign certificate request from vRA provisioned machines to Puppet. The Puppet Starter Content set this value on the Puppet server.