ClusterAutoscaler not scaling new worker even when there are pending pods in OpenShift Container Platform 4
Issue
-
We noticed some issues on the OCP 4 cluster related to cluster autoscaler. Not sure why this is, but even when the load gets big and there are new pending pod requests it doesn't trigger autoscaling of OpenShift workers.
Warning FailedScheduling <unknown> default-scheduler 0/56 nodes are available: 1 Insufficient memory, 11 Insufficient cpu, 44 node(s) didn't match node selector.
Environment
- Red Hat OpenShift Container Platform (RHOCP) 4.x
Subscriber exclusive content
A Red Hat subscription provides unlimited access to our knowledgebase, tools, and much more.