<-- Back
Time to complete: 45 minutes
In this lab we will install Puppet Enterprise 2016.5.1
- PE is free to install and evaluate
- When running PE without a license, you're limited to 10 agents
We are about to install Puppet Enterprise, and make sure you're logged into our puppet master host.
When connecting to a Vagrant-provisioned VM with vagrant ssh you will be logged in as the vagrant user (a normal/unprivileged user). However, we need root privileges when installing Puppet Enterprise.
In order to simplify the commands we run in the following sections, I will assume you know how to use sudo, and when to use it. All puppet commands need to be run as root, so the easiet thing to do would just to become root. We're in a training environment after all, and if you mess something up, you can just delete it and start over.
So, if you're already root, Great!
If you're not the root user yet, then become root, and be happy!
sudo su -
Now let's un-compress/de-archive the PE installation tarball, and install...
Change into the directory with the PE software:
cd /share/software/puppet
tar xzvf puppet-enterprise-2016.5.1-el-7-x86_64.tar.gz
cd puppet-enterprise-2016.5.1-el-7-x86_64Then run the installer:
./puppet-enterprise-installer
The installer will prompt you:
=============================================================
Puppet Enterprise Installer
=============================================================
Puppet Enterprise offers two different methods of installation.
[1] Guided install
Recommended for beginners. This method will install and configure a temporary
webserver to walk you through the various configuration options.
NOTE: This method requires you to be able to access port 3000 on this machine
from your desktop web browser.
[2] Text-mode
Recommended for advanced users. This method will open your $EDITOR (vi)
with a PE config file (pe.conf) for you to edit before you proceed with installation.
The pe.conf file is a HOCON formatted file that declares parameters and values needed to
install and configure PE.
We recommend that you review it carefully before proceeding.
=============================================================
How to proceed? [1]:
Accept the default [1] (just press Return/Enter).
Some packages will be installed, and then the installer will prompt you do open a page in your web browser...
## We're preparing the Web Installer...
2016-11-14 17:39:20,482 Running command: mkdir -p /opt/puppetlabs/puppet/share/installer/installer
2016-11-14 17:39:20,487 Running command: cp -pR /share/software/puppet/puppet-enterprise-2016.5.1-el-7-x86_64/* /opt/puppetlabs/puppet/share/installer/installer/
## Go to https://puppet.example.com:3000 in your browser to continue installation.
## Be sure to use 'https://' and that port 3000 is reachable through the firewall.
Remember that we have forwarded port 3000 to 22000 on our workstation, so...
- Use web browser to connect to: https://127.0.0.1:22000/
- Click Let's Get Started
- Click Monolithic Install
- For FQDN, enter puppet.example.com
- For DNS alias, enter puppet
- Select: Install PostgreSQL on the PuppetDB host for me.
- Enter an initial password for the admin user (we will change it later)
- Click Submit
- Click Continue
Note: Don't worry about the memory and disk space warnings. The amount of memory and disk space we've provisioned will be just fine for this training exercise. Obviously, when deploying for a real production environment, you'd want to pay attention to these!
You will see this:
We're checking to make sure the installation will work correctly
Verify local environment.
Verify that 127.0.0.1 can resolve puppet.example.com.
Verify root access on puppet.example.com.
Verify that DNS is properly configured for puppet.example.com.
Verify that your hardware meets requirements on puppet.example.com.
[puppet.example.com] We found 2,847 MB RAM. We recommend at least 4096 MB.
Verify that 127.0.0.1 has a PE installer that matches puppet.example.com's OS.
Verify that '/opt', '/var', and '/tmp' contain enough free space on puppet.example.com.
[puppet.example.com] Insufficient space in '/opt' (16 GB); we recommend at least 100 GB for a production environment.
Click Deploy Now
Intalling your deployment
Install Puppet Enterprise on puppet.
Verify that Puppet Enterprise is functioning on puppet.
[puppet] The puppet agent ran successfully.
Verify that MCollective is functioning on puppet.
Backup installer log files to puppet.
During the installation process you may click on 'Log view' to see what is happening behind the scenes, and then click 'Summary View' to return back to the overview.
Note: Once the installation completes, clicking the 'Start Using Puppet Enterprise' button will not work, as we are port-forwarding from a VM to our localhost. Use the link below instead.
We've forwarded port 443 from our puppet VM to port 22443 on our hosting workstation, so you should be able to connect to the PE Console via the URL:
- PE Console URL: https://127.0.0.1:22443/
Login as admin and enter the admin password you chose during the install.
If you forgot (or not sure what you typed) you can find the password in the answers file:
grep console_admin_password /opt/puppetlabs/puppet/share/installer/conf.d/puppet.example.com.conf
Probabbly a good idea to change it...! (You can change the admin password via the PE Console Web-Interface)
All log files for Puppet Enterprise and its components can be found in: /var/log/puppetlabs/
If you would like to look at the install logs, they are available within /var/log/puppetlabs/installer/
Believe it or not, that's all there is to installing a Monolithic puppet server.
Although we wont cover the Split install, nor the Large Environment Install in this tutorial, it may be helpful to understand what they are:
A Split install would consist of 3 hosts (or VMs) with the various components split across them as follows:
- Host 1
- Puppet Master (Compile Master)
- Certificate Authority (CA)
- ActiveMQ Broker (used by MCollective)
- Host 2
- Puppet Enterprise Console
- Host 3
- Puppet DB
- PostgreSQL Database (backing for PuppetDB)
A Large Environment Install takes it a step further and scales out the ActiveMQ infrastructure used by MCollective, and more importantly, the Puppet Compile Masters. In this example we've added multiple ActiveMQ Spokes and multiple Compile Masters (which the Puppet agents would point to). Only the PE infrastructure hosts would point to the MoM for their configuration. Additional Compile Masters and ActiveMQ Spoke Brokers could be added to scale out.
- Host 1
- Puppet Master of Masters (MoM)
- Certificate Authority (CA)
- ActiveMQ Spoke Broker (used by MCollective)
- Host 2
- Puppet Enterprise Console
- Host 3
- Puppet DB
- PostgreSQL Database (backing for PuppetDB)
- Host 4
- ActiveMQ Hub Broker
- Host 5
- ActiveMQ Spoke Broker
- Host 6
- ActiveMQ Spoke Broker
- Host 7
- Puppet Compile Master (load balanced)
- Host 8
- Puppet Compile Master (load balanced)
- Host 9
- Puppet Compile Master (load balanced)
A Split and/or Large Environment Install is a bit more work, as we'd be splitting up the various parts of PE on to separate VM's, as well as deploying multiple compile masters behind a loadbalancer. If you're interested in learning more about a split install, you may read about it in the PE Installation Guide. Here are the relevant pages:
Okay, let's get back to the lab...
Let's play around in the PE Console a bit. To change the 'admin' account password, click on 'admin' in the top right corner, and select 'My Account'. You should find a 'Reset password' link near the top/right of that page.
Look around the PE console. Try clicking on the Events, Nodes, and Classification tabs to see what you can see.
You should see 1 agent is registered called puppet.example.com. This is your Puppet Master!
Test your PE installation by running the agent manually from the shell prompt like this:
puppet agent -t
Note: Even on the Puppet Master, the Agent runs regularly in the background as a service. The Master configures itself through Puppet. Be aware, if you make a puppet change that affects all nodes, you will be affecting the master config as well (e.g. a global change to /etc/hosts).
Do not be fearful! Just be careful, and think about how your changes will affect every node in your ecosystem.
Do not disable the puppet agent on the master thinking that you're guarding against accidental changes that could break your puppet infrastructure. The agent runs on the master are required for mcollective key distribution (they get stored in the PuppetDB and then installed on the other nodes of puppet infra via exported resources). Also, there are certain configuration params in the puppet.conf on the master that are managed by puppet itself.
[root@puppet ~]# puppet agent -t
Info: Using configured environment 'production'
Info: Retrieving pluginfacts
Info: Retrieving plugin
Info: Loading facts
Info: Caching catalog for puppet.example.com
Info: Applying configuration version '1479158750'
Notice: Applied catalog in 15.20 secondsNot too exciting. We have confirmed that the agent run succeeds, so we know the puppetmaster is up and running and able to build a catalog for itself. Good.
If you're on a terminal that supports ANSI color, you'll notice that the text is in GREEN. If there are any puppet errors during a puppet run, those errors would show up in RED text.
Okay, let's move on...
Continue to Lab #4 --> Install Puppet Agent on agent node, and do test puppet run
These links are not needed for this Lab, but for reference here's the PE Install Guide at the PuppetLabs web site:
Quick Start Guide: https://docs.puppetlabs.com/pe/2016.5/quick_start_install_mono.html
Detailed Install Guide: https://docs.puppetlabs.com/pe/2016.5/install_basic.html
Split Install: https://docs.puppetlabs.com/pe/2016.5/install_pe_split.html
LEI Install: https://docs.puppetlabs.com/pe/2016.5/install_multimaster.html
<-- Back to Contents
Copyright © 2016 by Mark Bentley