Feat: Support trait with custom resource be dispatched to hubcluster

[fourierr]

Signed-off-by: Xiangbo Ma maxiangboo@cmbchina.com

Description of your changes

Support trait with custom resource be dispatched to hubcluster
Fixes #3754

The following is an example of how to use controlPlaneOnly

• Firstly, hubcpuscaler is defined on the basis of cpusccaler, so that HPA can be deployed to hubcuster

hubcpuscaler: {
	type: "trait"
	annotations: {}
	labels: {
		"ui-hidden": "true"
	}
	description: "Automatically scale the component based on CPU usage."
	attributes: {
		appliesToWorkloads: ["deployments.apps"]
		controlPlaneOnly: true
	}
}

template: {
	outputs: hubcpuscaler: {
		apiVersion: "autoscaling/v1"
		kind:       "HorizontalPodAutoscaler"
		metadata: name: context.name
		spec: {
			scaleTargetRef: {
				apiVersion: parameter.targetAPIVersion
				kind:       parameter.targetKind
				name:       context.name
			}
			minReplicas:                    parameter.min
			maxReplicas:                    parameter.max
			targetCPUUtilizationPercentage: parameter.cpuUtil
		}
	}

	parameter: {
		// +usage=Specify the minimal number of replicas to which the autoscaler can scale down
		min: *1 | int
		// +usage=Specify the maximum number of of replicas to which the autoscaler can scale up
		max: *10 | int
		// +usage=Specify the average CPU utilization, for example, 50 means the CPU usage is 50%
		cpuUtil: *50 | int
		// +usage=Specify the apiVersion of scale target
		targetAPIVersion: *"apps/v1" | string
		// +usage=Specify the kind of scale target
		targetKind: *"Deployment" | string
	}
}

• Secondly, write application file and apply

apiVersion: core.oam.dev/v1beta1
kind: Application
metadata:
  name: fourierapp03
  namespace: fourier
spec:
  components:
    - name: fourierapp03-fouriercomponent-01
      type: webservice
      properties:
        image: busybox:latest
        imagePullPolicy: IfNotPresent
        cmd: [ "sleep", "10000" ]
        cpu: "0.1"
        mem: "100Mi"
      traits:
        - type: hubcpuscaler
          properties:
            min: 1
            max: 10
            cpuPercent: 60
        - type: expose
          properties:
            port: [ 80 ]
  policies:
    - name: fourierapp03-topology-01
      type: topology
      properties:
        clusters: [ "kind01" ]
        namespace: fourier01
    - name: fourierapp03-topology-02
      type: topology
      properties:
        clusters: [ "kind02" ]
        namespace: fourier02
  workflow:
    steps:
      - type: deploy
        name: fourierapp03-deploy-01
        properties:
          policies: [ "fourierapp03-topology-01","fourierapp03-topology-02" ]

• Finally, you can see that HPA was dispatched to cluster local, deployment and service were dispatched to the cluster kind01, kind02 .
• And if you remove the hubcpuscaler in application, the Custom Resource in local will be GC Simultaneously. as shown below

apiVersion: core.oam.dev/v1beta1
kind: Application
metadata:
  name: fourierapp03
  namespace: fourier
spec:
  components:
    - name: fourierapp03-fouriercomponent-01
      type: webservice
      properties:
        image: busybox:latest
        imagePullPolicy: IfNotPresent
        cmd: [ "sleep", "10000" ]
        cpu: "0.1"
        mem: "100Mi"
      traits:
        - type: expose
          properties:
            port: [ 80 ]
  policies:
    - name: fourierapp03-topology-01
      type: topology
      properties:
        clusters: [ "kind01" ]
        namespace: fourier01
    - name: fourierapp03-topology-02
      type: topology
      properties:
        clusters: [ "kind02" ]
        namespace: fourier02
  workflow:
    steps:
      - type: deploy
        name: fourierapp03-deploy-01
        properties:
          policies: [ "fourierapp03-topology-01","fourierapp03-topology-02" ]

I have:

• Read and followed KubeVela's contribution process.
• Related Docs updated properly. In a new feature or configuration option, an update to the documentation is necessary.
• Run make reviewable to ensure this PR is ready for review.
• Added backport release-x.y labels to auto-backport this PR if necessary.

[codecov]

Codecov Report

Merging #3774 (495aced) into master (6761b18) will increase coverage by 0.02%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #3774      +/-   ##
==========================================
+ Coverage   63.87%   63.90%   +0.02%     
==========================================
  Files         309      309              
  Lines       29439    29452      +13     
==========================================
+ Hits        18805    18822      +17     
+ Misses       8196     8191       -5     
- Partials     2438     2439       +1     

Flag	Coverage Δ
apiserver-unittests	35.32% <0.00%> (-0.03%)	⬇️
core-unittests	56.18% <68.75%> (-0.07%)	⬇️
e2e-multicluster-test	26.51% <68.75%> (+0.13%)	⬆️
e2e-rollout-tests	23.69% <37.50%> (-0.01%)	⬇️
e2etests	31.48% <37.50%> (+0.39%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
apis/core.oam.dev/v1beta1/core_types.go	25.00% <ø> (+25.00%)	⬆️
...troller/core.oam.dev/v1alpha2/application/apply.go	88.55% <100.00%> (-1.08%)	⬇️
...ler/core.oam.dev/v1alpha2/application/generator.go	86.06% <100.00%> (+0.89%)	⬆️
pkg/multicluster/utils.go	66.66% <100.00%> (+0.46%)	⬆️
...tepdefinition/workflowstepdefinition_controller.go	67.02% <0.00%> (-9.58%)	⬇️
...es/policydefinition/policydefinition_controller.go	64.21% <0.00%> (-5.27%)	⬇️
pkg/apiserver/sync/store.go	55.00% <0.00%> (-4.00%)	⬇️
pkg/controller/utils/capability.go	80.30% <0.00%> (+0.44%)	⬆️
.../core/scopes/healthscope/healthscope_controller.go	79.42% <0.00%> (+1.15%)	⬆️
... and 5 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 6761b18...495aced. Read the comment docs.

[fourierr]

The following is an example of how to use controlPlaneOnly.

• Firstly, hubcpuscaler is defined on the basis of cpusccaler, so that HPA can be deployed to hubcuster

hubcpuscaler: {
	type: "trait"
	annotations: {}
	labels: {
		"ui-hidden": "true"
	}
	description: "Automatically scale the component based on CPU usage."
	attributes: {
		appliesToWorkloads: ["deployments.apps"]
		controlPlaneOnly: true
	}
}

template: {
	outputs: hubcpuscaler: {
		apiVersion: "autoscaling/v1"
		kind:       "HorizontalPodAutoscaler"
		metadata: name: context.name
		spec: {
			scaleTargetRef: {
				apiVersion: parameter.targetAPIVersion
				kind:       parameter.targetKind
				name:       context.name
			}
			minReplicas:                    parameter.min
			maxReplicas:                    parameter.max
			targetCPUUtilizationPercentage: parameter.cpuUtil
		}
	}

	parameter: {
		// +usage=Specify the minimal number of replicas to which the autoscaler can scale down
		min: *1 | int
		// +usage=Specify the maximum number of of replicas to which the autoscaler can scale up
		max: *10 | int
		// +usage=Specify the average CPU utilization, for example, 50 means the CPU usage is 50%
		cpuUtil: *50 | int
		// +usage=Specify the apiVersion of scale target
		targetAPIVersion: *"apps/v1" | string
		// +usage=Specify the kind of scale target
		targetKind: *"Deployment" | string
	}
}

• Secondly, write application file and apply

apiVersion: core.oam.dev/v1beta1
kind: Application
metadata:
  name: fourierapp03
  namespace: fourier
spec:
  components:
    - name: fourierapp03-fouriercomponent-01
      type: webservice
      properties:
        image: busybox:latest
        imagePullPolicy: IfNotPresent
        cmd: [ "sleep", "10000" ]
        cpu: "0.1"
        mem: "100Mi"
      traits:
        - type: hubcpuscaler
          properties:
            min: 1
            max: 10
            cpuPercent: 60
        - type: expose
          properties:
            port: [ 80 ]
  policies:
    - name: fourierapp03-topology-01
      type: topology
      properties:
        clusters: [ "kind01" ]
        namespace: fourier01
    - name: fourierapp03-topology-02
      type: topology
      properties:
        clusters: [ "kind02" ]
        namespace: fourier02
  workflow:
    steps:
      - type: deploy
        name: fourierapp03-deploy-01
        properties:
          policies: [ "fourierapp03-topology-01","fourierapp03-topology-02" ]

• Finally, you can see that HPA was dispatched to cluster local, deployment and service were dispatched to the cluster kind01, kind02 .
• And if you remove the hubcpuscaler in application, the Custom Resource in local will be GC Simultaneously. as shown below

apiVersion: core.oam.dev/v1beta1
kind: Application
metadata:
  name: fourierapp03
  namespace: fourier
spec:
  components:
    - name: fourierapp03-fouriercomponent-01
      type: webservice
      properties:
        image: busybox:latest
        imagePullPolicy: IfNotPresent
        cmd: [ "sleep", "10000" ]
        cpu: "0.1"
        mem: "100Mi"
      traits:
        - type: expose
          properties:
            port: [ 80 ]
  policies:
    - name: fourierapp03-topology-01
      type: topology
      properties:
        clusters: [ "kind01" ]
        namespace: fourier01
    - name: fourierapp03-topology-02
      type: topology
      properties:
        clusters: [ "kind02" ]
        namespace: fourier02
  workflow:
    steps:
      - type: deploy
        name: fourierapp03-deploy-01
        properties:
          policies: [ "fourierapp03-topology-01","fourierapp03-topology-02" ]