With the release of Kublr 1.19, we are continuing the tradition of expanding customization capabilities available to end users and Kubernetes operators and administrators.
Kublr 1.19 includes numerous improvements to the customization of Kubernetes clusters deployed on AWS and other clouds. Key among these is support for mixed instance policies including spot and on-demand instances and multiple instance types. Also new are additional AWS resources included within the Kublr cluster specification and managed by Kublr, and more.
These updates and practical tips for using them are explained in this on-demand webinar: Setting up a CI/CD Pipeline with Kublr and Kubernetes. The webinar walks through how Kublr can be used to deploy a Kubernetes cluster on AWS in high availability mode on inexpensive spot instances — in just under 20 minutes. This includes EFS-based cloud native storage and ready to use in production CI/CD tools like Jenkins and Nexus. It also includes a link to a github repository with all the code available for reuse and experimentation.
With this release, the Kublr Control Plane also packs a number of operational improvements such as full support for deployment to Kubernetes 1.18 and running in a cluster with pod security policies restricted by default, and in-place upgrades. Before Kublr 1.19 you could create managed Kubernetes 1.18 clusters, but the Kublr Control Plane itself ran in a Kubernetes 1.17 cluster. With Kublr 1.19, the control plane also runs in a Kubernetes 1.18 cluster. In-place upgrades greatly simplify the job of IT operations teams in upgrading the Kublr Control Plane.
Kublr 1.19 introduces support for additional helm packages deployed and managed as a part of the Kublr cluster specification. You can now include helm packages into the cluster specification and deploy system software together with your clusters in one step. This greatly simplifies the job of DevOps engineers / SRE and makes GitOps and infrastructure-as-a-code implementation much easier. The same webinar demonstrates this new capability using Jenkins and Nexus helm packages as an example.
Other new features include Centos/RHEL 8.1+ support, high availability configuration for AWS NAT gateways for private subnets, master-only cluster creation via the user interface (UI). A refreshed look-and-feel introduces a more efficient and modular UI for cluster provisioning and editing, and much more.
Together, these changes continue to evolve Kublr as the most robust enterprise Kubernetes platform on the market and align with our ultimate goal to become a true meta-cloud. A new type of infrastructure with flexible, cloud native, enterprise-grade modules deployable across different public and private clouds, as well as on-premise. Stay in touch to find out when 1.20 is available.