Optimize planned evacuation of a node in Red Hat OpenShift Container Platform
Issue
- We do maintenance on our nodes node by node, meaning we evacuate all pods, do our tasks. At the moment the
--evacuateoption immediately terminates all pods resulting in about 80-100 new pods which need to be scheduled, pulled and started. This also puts quite a lot of load on a lot of components (openshift, registry, nfs etc.). In our enviroment it took openshift between 5-15 minutes to fully restart and restore all pods. We know that applications should be scaled to more than one pod to minimize downtime. But a planned maintenance could be softer on the environment.
Environment
- Red Hat OpenShift Container Platform 3.2.1
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase of over 48,000 articles and solutions.
Welcome! Check out the Getting Started with Red Hat page for quick tours and guides for common tasks.
