Today we are happy to announce a new release of the Banzai Cloud logging operator. It’s been a long time from the first commits till today, and is always nice to look back, learn and reflect on the evolution of the project. The first major release, June 2018 This was the very first release, and among the first operators we made. The operator pattern was pretty new, and the goal of the first logging operator was fairly simple - automate the manual fluent ecosystem configurations we were doing for our customers with the Pipeline platform.

Kubernetes is a highly extensible framework that is built from a bunch of loosely coupled components. This gives a very high level of flexibility, but adds some new challenges to the operation compared to monolithic solutions of similar systems of the past. One of these challenges is observability, especially log collection. This post describes how to collect the logs of Kubernetes components in detail, but does not discuss the collection of application (workload) logs.

Frequent readers of our blog and users of our hybrid cloud container management platform, Pipeline, will be familiar with the integrated cluster services that come with it. These services are automated end-to-end solutions for centralized logging, federated monitoring, security scans, advanced credential management, autoscaling, registries and lots more (see, for example, automated DNS management for Kubernetes). Providing an automated logging solution, and making sure it works seamlessly across multiple clusters, has always been part of Pipeline.

On this blog we’ve already discussed our totally redesigned logging operator, which automates logging pipelines on Kubernetes. Thanks to the tremendous amount of feedback and the numerous contributions we received from our community, we’ve been able to rethink and redesign that operator from scratch, but the improvements aren’t going to stop coming any time soon. Our goal is to continue removing the burden from human operators, and to help them manage the complex architectures of Kubernetes.

About a year ago we published the first release of our popular logging-operator. The initial version of that operator was designed to fit Pipeline, the Banzai Cloud hybrid cloud container management platform. However, since then, all kinds of people have found it to be an extremely useful tool that helps them manage their logs on Kubernetes. Initially, Fluent ecosystem automation was enough to support the disparate needs of our userbase, but, as the popularity of the logging-operator grew, different setups were put in place by our community that revealed some of its limitations.