Helm Chart for Redis-HA is invalid, resource constraints also do not comply with GKE AutoPilot policies

Question

Helm Chart for Redis-HA is invalid, resource constraints also do not comply with GKE AutoPilot policies

kavinaravind opened this issue a year ago · 0 comments

kavinaravind commented a year ago

Category:

Kubernetes apps

Type:

Bug
Feature Request
Process

Redis has outdated specs for GKE Autopilot Clusters:

Error from server (GKE Warden constraints violations): error when creating "redis-ha-1_manifest.yaml": admission webhook "warden-validating.common-webhooks.networking.gke.io" denied the request: GKE Warden rejected the request because it violates one or more constraints.
Violations details: {"[denied by autogke-pod-limit-constraints]":["workload 'redis-ha-1-haproxy' cpu requests '250m' is lower than the Autopilot minimum required of '500m' for using pod anti affinity."]}
Requested by user: '<id>', groups: 'system:authenticated'.

Changing cpu: 100m to cpu: 500m was needed.

The code that was used to generate the helm template seems to have copied the name of the pointer to the variable rather than the label itself:
- https://github.com/GoogleCloudPlatform/click-to-deploy/blob/master/k8s/redis-ha/chart/redis-ha/templates/haproxy-deployment.yaml#L14
- https://github.com/GoogleCloudPlatform/click-to-deploy/blob/master/k8s/redis-ha/chart/redis-ha/templates/redis-statefulset.yaml#L5

The helm chart / Application CRD seems to be fairly outdated yet GCP seems to be using this as the click-to-deploy solution in https://console.cloud.google.com/kubernetes/application, is there a private fork that is being used?