chetanmeh opened a new pull request #493: Configure same affinity policies as array URL: https://github.com/apache/incubator-openwhisk-deploy-kube/pull/493 Currently affinity value are getting duplicated and override the values. For example rendering the controller-pod.yaml ``` helm template openwhisk --name bladerunner -x templates/controller-pod.yaml --output-dir ../build ``` Renders the yaml with following content ```yaml serviceAccountName: bladerunner-core restartPolicy: Always affinity: # prefer to not run on an invoker node (only prefer because of single node clusters) nodeAffinity: preferredDuringSchedulingIgnoredDuringExecution: - weight: 100 preference: matchExpressions: - key: openwhisk-role operator: NotIn values: - invoker # prefer to run on a core node nodeAffinity: preferredDuringSchedulingIgnoredDuringExecution: - weight: 80 preference: matchExpressions: - key: openwhisk-role operator: In values: - core # Fault tolerance: prevent multiple instances of bladerunner-controller from running on the same node podAntiAffinity: requiredDuringSchedulingIgnoredDuringExecution: - labelSelector: matchExpressions: - key: name operator: In values: - bladerunner-controller topologyKey: "kubernetes.io/hostname" initContainers: ``` As can be seen that under `affinity` there are 2 entries for `nodeAffinity` which would result in last one overwriting the first one. Ideally this should be flagged in `helm lint` however it looks like this check is missing per helm/helm#5524
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
