JointInferenceService controller enhancement

[SherlockShemol]

What type of PR is this?

What this PR does / why we need it:

• create deployment resources to enable pod to rebuild after manual deletion.
• enable JointInferenceService Controller to update pod when modifying CRD.
• add test file to test files to ensure functionality stability and correctness.
  Which issue(s) this PR fixes:

Fixes #

[MooreZheng]

There are CI issues that remain to be resolved, see https://github.com/kubeedge/sedna/actions/runs/11249926500/job/31277758880?pr=445

[tangming1996]

/lgtm

[tangming1996]

Can't return directly; otherwise, the subsequent code won't be executed directly.

Modify the updateInferenceServices function as follows.

func (c *Controller) updateInferenceServices(old, cur interface{}) error {
	oldService := old.(*sednav1.JointInferenceService)
	newService := cur.(*sednav1.JointInferenceService)

	// Check if the cloud worker configuration has changed
	if !reflect.DeepEqual(oldService.Spec.CloudWorker, newService.Spec.CloudWorker) {
		// Update cloud worker and log any errors
		if err := c.updateCloudWorker(newService); err != nil {
			klog.Errorf("Failed to update cloud worker for service %s/%s: %v", newService.Namespace, newService.Name, err)
		}
	}

	// Retrieve the address of the cloud inference service
	var bigModelHost string
	svc, err := c.kubeClient.CoreV1().Services(oldService.Namespace).Get(context.Background(),
		strings.ToLower(oldService.Name+"-"+jointInferenceForCloud), metav1.GetOptions{})
	if err != nil {
		if errors.IsNotFound(err) {
			// If the service does not exist, create a new one and retrieve its address
			klog.Info("Cloud service not found, creating new service...")
			bigModelHost, err = runtime.CreateEdgeMeshService(c.kubeClient, oldService, jointInferenceForCloud, BigModelPort)
			if err != nil {
				klog.Errorf("Failed to create EdgeMesh service for service %s/%s: %v", oldService.Namespace, oldService.Name, err)
			}
		} else {
			klog.Errorf("Failed to get cloud service %s/%s: %v", oldService.Namespace, oldService.Name, err)
		}
	} else {
		bigModelHost = fmt.Sprintf("%s.%s", svc.Name, svc.Namespace)
	}

	// Check if the edge worker configuration has changed
	if !reflect.DeepEqual(oldService.Spec.EdgeWorker, newService.Spec.EdgeWorker) {
		// Update edge worker and log any errors
		if err := c.updateEdgeWorker(newService, bigModelHost); err != nil {
			klog.Errorf("Failed to update edge worker for service %s/%s: %v", newService.Namespace, newService.Name, err)
		}
	}

	return nil
}

[SherlockShemol]

modified.

[tangming1996]

Grouping of imports.

[SherlockShemol]

modified.

[tangming1996]

Leave the configuration of hostNetwork to the user. There is no need for mandatory configuration here.

[tangming1996]

ditto

[tangming1996]

After these pull requests are merged, the current joint-inference examples need to add the hostNetwork configuration in the YAML, similar to the following:

CLOUD_NODE="cloud-node-name"
EDGE_NODE="edge-node-name"


kubectl create -f - <<EOF
apiVersion: sedna.io/v1alpha1
kind: JointInferenceService
metadata:
  name: helmet-detection-inference-example
  namespace: default
spec:
  edgeWorker:
    model:
      name: "helmet-detection-inference-little-model"
    hardExampleMining:
      name: "IBT"
      parameters:
        - key: "threshold_img"
          value: "0.9"
        - key: "threshold_box"
          value: "0.9"
    template:
      spec:
        nodeName: $EDGE_NODE
        dnsPolicy: ClusterFirstWithHostNet
        hostNetwork: true
        containers:
        - image: kubeedge/sedna-example-joint-inference-helmet-detection-little:v0.3.0
          imagePullPolicy: IfNotPresent
          name:  little-model
          env:  # user defined environments
          - name: input_shape
            value: "416,736"
          - name: "video_url"
            value: "rtsp://localhost/video"
          - name: "all_examples_inference_output"
            value: "/data/output"
          - name: "hard_example_cloud_inference_output"
            value: "/data/hard_example_cloud_inference_output"
          - name: "hard_example_edge_inference_output"
            value: "/data/hard_example_edge_inference_output"
          resources:  # user defined resources
            requests:
              memory: 64M
              cpu: 100m
            limits:
              memory: 2Gi
          volumeMounts:
            - name: outputdir
              mountPath: /data/
        volumes:   # user defined volumes
          - name: outputdir
            hostPath:
              # user must create the directory in host
              path: /joint_inference/output
              type: Directory

  cloudWorker:
    model:
      name: "helmet-detection-inference-big-model"
    template:
      spec:
        nodeName: $CLOUD_NODE
        dnsPolicy: ClusterFirstWithHostNet
        containers:
          - image: kubeedge/sedna-example-joint-inference-helmet-detection-big:v0.3.0
            name:  big-model
            imagePullPolicy: IfNotPresent
            env:  # user defined environments
              - name: "input_shape"
                value: "544,544"
            resources:  # user defined resources
              requests:
                memory: 2Gi
EOF

[MooreZheng]

/lgtm

[tangming1996]

/lgtm

[jaypume]

/lgtm

[jaypume]

/approve

[kubeedge-bot]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: jaypume

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Needs approval from an approver in each of these files:

• OWNERS [jaypume]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment