There are murmurs on the internet that tools like Puppet are being killed off by Kubernetes. The reality is a little more complex. True, Kubernetes poses some challenges to various players in the infrastructure automation market, but they nevertheless remain important tools for engineers charged with managing infrastructure.
Kubernetes is forcing this market to adapt – and with Puppet announcing new tools and features to its portfolio in Puppet Enterprise 2019.1 yesterday, this it’s clear that the team are making the necessary strides to remain a key part of the infrastructure automation landscape.
Update: This article was amended to highlight that Puppet Enterprise is a distinct product separate from Continuous Delivery for Puppet Enterprise.
What’s new for Puppet Enterprise 2019.1?
There are two key elements to the Puppet announcement: enhanced integration with Puppet Bolt – an open source, agentless task runner – and improved capabilities with Continuous Delivery for Puppet Enterprise.
Puppet Bolt
Puppet Bolt, the Puppet team argue, offers a really simple way to get started with infrastructure automation “without requiring an agent installed on a remote target.”
The Puppet team explain that Puppet Bolt essentially allows users to expand the scope of what they can automate without losing the consistency and control that you’d expect when using a tool like Puppet.
This has some significant benefits in the context of Kubernetes. Bryan Belanger, Principal Consultant at Autostructure, said “We love using Puppet Bolt because it leverages our existing Puppet roles and classifications allowing us to easily make changes to large groups of servers and upgrade Kubernetes clusters quicker, which is often a pain if done manually.”
Belanger continues, saying “with the help of Puppet Bolt, we were also able to fix more than 1,000 servers within five minutes and upgrade our Kubernetes clusters within four hours, which included coding and tasks.”
Continuous Delivery for Puppet Enterprise
Updates to the Continuous Delivery product aim to make DevOps practices easier – the Puppet team are clearly trying to make it easier for organizations to empower their colleagues and continue to build a culture where engineers are not simply encouraged to be responsible for code deployment, but also able to do it with minimal fuss.
Module Delivery Pipelines now mean modules can be independently deployed without blocking others, while Simplified Puppet Deployments aims to make it easier for engineers that aren’t familiar with Puppet to “push simple infrastructure changes immediately and easily perform complex rolling deployments to a group of nodes in batches in one step.”
But there is also another dimension that aims to help engineers take pro-active steps to tackle resiliency and security issues. With Impact Analysis teams will be able to look at the potential impact of a deployment before it’s done.
Read next: “This is John. He literally wrote the book on Puppet” – An Interview with John Arundel
What’s the big idea behind this announcement?
The over-arching narrative that’s coming from the top is about supporting teams to scale their DevOps processes. It’s about making organizations’ ‘automation footprint’ more manageable.
“IT teams need a simple way to get started with automation and a solution that grows with them as their automation footprint grows,” Matt Waxman, Head of Product at Puppet, explains. “You shouldn’t have to throw away your existing scripts or tools to scale automation across your organization. Organizations need a solution that is extensible — one that complements their current automation efforts and helps them scale beyond individuals to multiple teams.”
Puppet Enterprise 2019.1 will be out on general availability on May 7 2019. Learn more here.