DevOps in the Cluster: A deep dive into operators
summary
Explains that operators are Kubernetes controllers for custom resources that encode operational knowledge and actually manage apps not just install them, uses the Operator SDK and Operator Framework, demonstrates an application monitoring operator that deploys two other operators, and builds a monitoring stack with Prometheus and Grafana plus Alertmanager; shows the operator creating CRDs, rolebindings and ownerReferences, exposes a demo bug where a created flag prevents updates, and proves end-to-end deployment works after a small fix.
AI-written summary · about
chapters (23)
- 0:00Start
- 0:21Agenda
- 0:57Kubernetes Applications
- 1:08What Is a Kubernetes Application
- 2:39Helm
- 5:55Kubernetes Api
- 7:42Custom Resources
- 8:05Custom Resource
- 9:18Characteristics of Operators
- 10:14What's the Operator Sdk
- 10:41General Workflow When Using the Operator Sdk
- 11:17Bootstrap a New Operator
- 13:01Implementation
- 17:36Tips for Devon on Developing Operators
- 19:11Be Careful with Finalized Errs
- 20:07Allow for Deleting in any Operator State
- 22:41Never Rely on Local Testing Only
- 23:44Dependency Resolution
- 23:54Dependency Resolution with Operators
- 24:14The Operator Lifecycle Manager
- 24:51Demo
- 28:23Config Maps
- 29:14Requirements for Monitoring
related talks
14mDemystifying k8s operators
53mAutomating Stateful Applications with Kubernetes Operators
42mKubernetes Operators: What are They and Why would You Need Them?
21mOperators for Beginners
01h 29mTutorial: Deep Dive into the Operator Framework for Kubernetes
31mKubernetes Operators within Reach of Any Developer
39mKubernetes Operators: Managing Complex Software with Software
24mAutomating Stateful Apps with Kubernetes Operators