homeblogupgrading your puppet managed windows 2008 servers

Upgrading your Puppet-managed Windows 2008 servers

As you probably already know, on 14 January 2020, Microsoft plans to end support for Windows Server 2008 as a way to accelerate innovation with improved cost savings and security. If you’ve modeled your existing infrastructure with Puppet, the good news is that it will be much easier to recreate this elsewhere, whether that be on-prem or in Microsoft Azure. Below we’ll share with you the best way to get started.

The side-by-side approach (migration)

While Puppet won’t move your actual data for you, if your infrastructure is already managed by Puppet, you will be able to rebuild your new server to look exactly like your existing one. You shouldn’t lose much time in getting the new system running and ready to receive the data. The best way to go about this is a side-by-side approach, also known as migration — where you rebuild the server with the new OS from scratch and migrate the application data. You can do this either on-prem or in Microsoft Azure. For instructions on how to migrate, see the Windows upgrade documentation.

Note that if you keep your Puppet code in Azure DevOps, specifically Azure Repos, Continuous Delivery for Puppet Enterprise 3.0 can validate changes to a module or control repository and perform impact analysis before triggering the relevant pipeline to test and deploy the change. This gives you the advantages that come with simplifying the continuous delivery of your Puppet code.

How much effort will the migration require? This depends on the additional software running on your server. The more software you have, and the worse the documentation, the bigger the job of getting a fresh install to match your original server. Either way, you’ll have a more successful migration if your configuration is modeled in Puppet’s infrastructure-as-code, as it provides up-to-date self-documented infrastructure which is automatically applied to the new server.

Give the migration approach a go, and hopefully you’ll be glad you invested in an infrastructure-as-code and automation solution.

Claire Cadman is a technical writer at Puppet and Kevin Reeuwijk is a principal sales engineer at Puppet.

Learn more