Support schedulerEstimator as a list in karmada helm chart

[wengyao04]

What would you like to be added:

• Support schedulerEstimator as a list in karmada helm chart
• add command line argument in karmada-scheulder.yaml to enable estimator like

  - --enable-scheduler-estimator=true
  

• Could we also make "helm.sh/hook-delete-policy": hook-succeeded configurable in post-install-job.yaml. We want to keep the job pod for debugging issues if post-install job fails

Why is this needed:
Hi, we want to install scheduler estimator using helm chart. But the helm char only supporting one cluster https://github.com/karmada-io/karmada/blob/master/charts/karmada/values.yaml#L748. I manually install one for another member cluster. Could we make them as a list ?

Thank you !

[chaosi-zju]

/assign

[chaosi-zju]

hi @wengyao04, as for first problem, how about helm chart change like this PR: #4358

karmada/charts/karmada/values.yaml

Lines 741 to 876 in d4ad908

	## karmada scheduler estimator
	schedulerEstimator:
	## @param schedulerEstimator.memberClusters each cluster requires an estimator component, fill in the information for each cluster
	memberClusters:
	## @param schedulerEstimator.memberClusterInfo[0].clusterName the name of the member cluster
	- clusterName: ""
	## @param schedulerEstimator.memberClusterInfo[0].replicaCount target replicas
	replicaCount: 1
	## kubeconfig of the member cluster
	kubeconfig:
	## @param schedulerEstimator.memberClusterInfo[0].kubeconfig.server apiserver of the member cluster
	server: ""
	## @param schedulerEstimator.memberClusterInfo[0].kubeconfig.caCrt ca of the certificate
	caCrt: |
	-----BEGIN CERTIFICATE-----
	XXXXXXXXXXXXXXXXXXXXXXXXXXX
	-----END CERTIFICATE-----
	## @param schedulerEstimator.memberClusterInfo[0].kubeconfig.crt crt of the certificate
	crt: |
	-----BEGIN CERTIFICATE-----
	XXXXXXXXXXXXXXXXXXXXXXXXXXX
	-----END CERTIFICATE-----
	## @param schedulerEstimator.memberClusterInfo[0].kubeconfig.key key of the certificate
	key: |
	-----BEGIN RSA PRIVATE KEY-----
	XXXXXXXXXXXXXXXXXXXXXXXXXXX
	-----END RSA PRIVATE KEY-----
	## @param schedulerEstimator.labels labels of the scheduler-estimator deployment
	labels:
	app: karmada-scheduler-estimator
	## @param schedulerEstimator.podAnnotations annotations of the scheduler-estimator pods
	podAnnotations: {}
	## @param schedulerEstimator.podLabels labels of the scheduler-estimator pods
	podLabels: {}
	## @param image.registry karmada schedulerEstimator image registry
	## @param image.repository karmada schedulerEstimator image repository
	## @param image.tag karmada schedulerEstimator image tag (immutable tags are recommended)
	## @param image.pullPolicy karmada schedulerEstimator image pull policy
	## @param image.pullSecrets Specify docker-registry secret names as an array
	##
	image:
	registry: docker.io
	repository: karmada/karmada-scheduler-estimator
	tag: *karmadaImageVersion
	pullPolicy: *karmadaImagePullPolicy
	## Optionally specify an array of imagePullSecrets.
	## Secrets must be manually created in the namespace.
	## Example:
	## pullSecrets:
	## - myRegistryKeySecretName
	##
	pullSecrets: []
	## @param schedulerEstimator.resources resource quota of the scheduler-estimator
	resources: {}
	# If you do want to specify resources, uncomment the following
	# lines, adjust them as necessary, and remove the curly braces after 'resources:'.
	# limits:
	# cpu: 100m
	# memory: 128Mi
	# requests:
	# cpu: 100m
	# memory: 128Mi
	## @param schedulerEstimator.nodeSelector node selector of the scheduler-estimator
	nodeSelector: {}
	## @param schedulerEstimator.affinity affinity of the scheduler-estimator
	affinity: {}
	## @param schedulerEstimator.tolerations tolerations of the scheduler-estimator
	tolerations: []
	# - key: node-role.kubernetes.io/master
	# operator: Exists
	## @param schedulerEstimator.strategy strategy of the scheduler-estimator
	strategy:
	type: RollingUpdate
	rollingUpdate:
	maxUnavailable: 0
	maxSurge: 50%
	## @param apiServer.podDisruptionBudget
	podDisruptionBudget: *podDisruptionBudget
	## descheduler config
	descheduler:
	## @param descheduler.labels labels of the descheduler deployment
	labels:
	app: karmada-descheduler
	## @param descheduler.replicaCount target replicas of the descheduler
	replicaCount: 2
	## @param descheduler.podAnnotations annotations of the descheduler pods
	podAnnotations: {}
	## @param descheduler.podLabels labels of the descheduler pods
	podLabels: {}
	## @param image.registry karmada descheduler image registry
	## @param image.repository karmada descheduler image repository
	## @param image.tag karmada descheduler image tag (immutable tags are recommended)
	## @param image.pullPolicy karmada descheduler image pull policy
	## @param image.pullSecrets Specify docker-registry secret names as an array
	##
	image:
	registry: docker.io
	repository: karmada/karmada-descheduler
	tag: *karmadaImageVersion
	pullPolicy: *karmadaImagePullPolicy
	## Optionally specify an array of imagePullSecrets.
	## Secrets must be manually created in the namespace.
	## Example:
	## pullSecrets:
	## - myRegistryKeySecretName
	##
	pullSecrets: []
	## @param descheduler.resources resource quota of the descheduler
	resources: {}
	# If you do want to specify resources, uncomment the following
	# lines, adjust them as necessary, and remove the curly braces after 'resources:'.
	# limits:
	# cpu: 100m
	# memory: 128Mi
	# requests:
	# cpu: 100m
	# memory: 128Mi
	## @param descheduler.nodeSelector node selector of the descheduler
	nodeSelector: {}
	## @param descheduler.affinity affinity of the descheduler
	affinity: {}
	## @param descheduler.tolerations tolerations of the descheduler
	tolerations: []
	# - key: node-role.kubernetes.io/master
	# operator: Exists
	## @param descheduler.strategy strategy of the descheduler
	strategy:
	type: RollingUpdate
	rollingUpdate:
	maxUnavailable: 0
	maxSurge: 50%
	## @param descheduler.kubeconfig kubeconfig of the descheduler
	kubeconfig: karmada-kubeconfig
	## @param apiServer.podDisruptionBudget
	podDisruptionBudget: *podDisruptionBudget

[wengyao04]

Hi @chaosi-zju Thank you very much ! It works perfect for us !
BTW

Could we also add command line argument in karmada-scheulder.yaml to enable estimator like
- --enable-scheduler-estimator=true

[RainbowMango]

Could we also add command line argument in karmada-scheulder.yaml to enable estimator like

• --enable-scheduler-estimator=true

That makes sense to me. I wonder, would you like to send a PR for it?

[wengyao04]

Thanks @RainbowMango ! Yes we can send a PR for it

[chaosi-zju]

hi @wengyao04, I am not quite sure if this PR #4393 is what your problem 3 want?

please help explaining what's your perfect configurable way~

[RainbowMango]

/reopen
Seems there are two more tasks addressed by this issue. The first one was completed by #4358

[karmada-bot]

@RainbowMango: Reopened this issue.

In response to this:

/reopen
Seems there are two more tasks addressed by this issue. The first one completed by #4358

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.