DevConf.US 2021

It’s easy to get lost in logs and dashboards while navigating through problematic tests, builds, jobs resulting from the CI processes in applications with huge codebases like Kubernetes. By leveraging the data made available by Kubernetes and OpenShift testing, reporting, and visualization platforms like Prow, TestGrid, Bugzilla, etc. we can build intelligent open source AIOps tooling to better monitor builds, and help developers get to the root cause of failures. In this session, the speakers demonstrate how they’ve collected data from various CI/CD tools to calculate key performance indicator metrics. These metrics can help monitor the state of a CI workflow and can be shared via automated dashboards running on Kubeflow pipelines which can help investigate problematic tests, builds, or jobs. By collecting relevant metrics that can quantify the state of the CI workflow, we are presented with an opportunity to build more intelligent reporting tools that can aid an engineer's development lifecycle. By the end of the talk, the audience will understand how to build intelligent AIOps monitoring tools and dashboards that will enable them to get more visibility into their failures and ensure faster recovery to better support their CI/CD processes.