-
Bug
-
Resolution: Unresolved
-
Undefined
-
None
-
4.15
-
None
-
No
-
2
-
OSDOCS Sprint 252, OSDOCS Sprint 253
-
2
-
False
-
https://issues.redhat.com/browse/OCPBUGS-19317
By default, the default minimum resources, e.g. "25m" cpu gets split across the number of containers in a pod
If there is more than one container in a pod (or you rename a container, and cause this very bug here), that can be a very small number, and might not be enough resources for health checks to succeed
If you want to avoid this, you need to specify a .spec.resourcePolicy.containerPolicies like:
apiVersion: "autoscaling.k8s.io/v1"
kind: VerticalPodAutoscaler
metadata:
name: hamster-vpa
spec:
targetRef:
apiVersion: "apps/v1"
kind: Deployment
name: sleeper
resourcePolicy:
containerPolicies:
- containerName: '*'
minAllowed:
cpu: 25
memory: 50Mi
controlledResources: ["cpu", "memory"]
- documents
-
-
- ASSIGNED
-