Today we’re excited to introduce native Kubernetes support to the Gremlin Reliability as a Service platform. Gremlin users can now easily discover, visualize, and target their Kubernetes objects within our web app and using our API. Now, when kicking off an experiment, Gremlin will intelligently select the containers underlying the specified Kubernetes objects so that you can be confident that your application runs the way you expect it to on Kubernetes.
In the past if you’ve installed Gremlin on a container in a Kubernetes cluster, you dealt with containers being spun up and down by the Kubernetes orchestrator. Targeting the containers underlying the service on which you’d like to experiment has been a bit like a game of whack-a-mole.
At the core, Kubernetes is a system to automate and abstract the complexity around deploying, scaling, and managing containerized applications. As Kubernetes has become a production ready platform used across the industry, we’ve brought this simplicity to Gremlin, allowing safe and easy targeting of your desired services when running Chaos Engineering experiments, allowing you to think about your infrastructure in abstract terms, exactly as you do with Kubernetes.
"Gremlin enables us to run chaos experiments on specific Kubernetes services", said Josh Simmonds, Senior Site Reliability Engineer at Ping Identity. "This significantly reduces the amount of time it takes to do negative testing and increases our systems' resilience to failure."
What you see next should be very familiar to you, but with the newly added Kubernetes option. Your list of clusters and namespaces are available and searchable to let you filter down the set of Kubernetes objects that are available for experimentation. Your objects are then broken down by Deployment, DaemonSet, ReplicaSet, StatefulSet, and Pod to allow easy and safe targeting. Try clicking on one or many objects and you’ll notice the visual map of your cluster light up to represent what will be impacted when this attack is run.
Once you’ve selected and configured the Gremlin of your choice, unleashing the attack will interpret the Kubernetes objects you’ve selected and intelligently target the appropriate containers. You no longer have to look for specific containers from a long list and wonder if you’ve found them all. Instead, the containers targeted by this attack will show up in the attack details screen, grouped together by the Kubernetes object they belong to.
All of the container details and logs will be available for your review as you’d expect when running any type of Gremlin experiment.
The native integration with Kubernetes is available now for Free and Pro users. Get started by heading to the web app, create a new attack and click on Kubernetes. Targeting Kubernetes will soon be possible within a Scenario, however is not available just yet.