CPU-based autoscaling also requires limit ranges to be set on CPU requests for the pods being scaled, so we can use the LimitRange definition from one of the previous sections.
# cat my-limits.yaml apiVersion: v1kind: LimitRangemetadata: name: my-limitsspec: limits: - type: Pod min: cpu: 50m memory: 64Mi max: cpu: 150m memory: 128Mi - type: Container min: cpu: 50m memory: 64Mi max: cpu: 150m memory: 128Mi# oc create -f my-limits.yaml limitrange "my-limits" created
Depending on your host machine's CPU, you might have to tweak the values in the file above in order for autoscaling to work, that is why in the listing above they are different than in the beginning of the chapter.
The autoscaling feature can be applied to ...
