Migrating Applications

This section describes how to create a Deployment with the same specifications as that in CCE 1.0 on the CCE 2.0 console.

It is advised to delete the applications on CCE 1.0 only after you have successfully created these applications on CCE 2.0.

Before Migration

Common Kubernetes resources include Deployments, StatefulSets, jobs, DaemonSets, pods, ConfigMaps, secrets, and Services. To migrate applications, export the Kubernetes resources from CCE 1.0 and create them on the CCE 2.0 clusters.

Procedure

1. Export resource files from CCE 1.0.

   kubectl get {resource} {name} -n {namespace} -oyaml --export > {namespace}_{resource}_{name}.yaml

   Assume that the following resource files are exported:

   kubectl get svc app-svc-test -ndefault -oyaml --export > default_svc_app-svc-test.yaml

   kubectl get secret secret-test -ndefault -oyaml --export > default_secret_secret-test.yaml

   kubectl get cm cm-test -ndefault -oyaml --export > default_cm_cm-test.yaml

2. Switch to the CCE 2.0 clusters and run the following kubectl command to create the resources exported in 1.

   kubectl create -f{namespace}_{resource}_{name}.yaml

   Examples of creating resource files:

   kubectl create -f default_svc_app-svc-test.yaml

   kubectl create -f default_secret_secret-test.yaml

   kubectl create -f default_cm_cm-test.yaml

3. In CCE 1.0, the application type is ReplicationController, which needs to be changed into Deployment in CCE 2.0.

   To change the application type, replace ReplicationController in the ReplicationController YAML file exported from CCE 1.0 with Deployment.

   Assume that the ReplicationController YAML file exported is as follows:

   apiVersion: v1
   kind: ReplicationController
   metadata:
   annotations:
   cce/app-createTimestamp: 2019-09-09-08-59-13
   cce/app-description: ""
   cce/app-updateTimestamp: 2019-09-09-08-59-13
   creationTimestamp: null
   generation: 1
   labels:
   cce/appgroup: app-test
   name: ssss
   rollingupdate: "false"
   name: ssss
   selfLink: /api/v1/namespaces/default/replicationcontrollers/ssss
   spec:
   replicas: 1
   selector:
   cce/appgroup: app-test
   name: ssss
   rollingupdate: "false"
   template:
   metadata:
   annotations:
   scheduler.alpha.kubernetes.io/affinity: '{"nodeAffinity":{"requiredDuringSchedulingIgnoredDuringExecution":{"nodeSelectorTerms":[{"matchExpressions":[{"key":"failure-domain.beta.kubernetes.io/zone","operator":"In","values":["eu-de-02","eu-de-01"]}]}]}}}'
   creationTimestamp: null
   labels:
   cce/appgroup: app-test
   name: ssss
   rollingupdate: "false"
   spec:
   affinity:
   nodeAffinity:
   requiredDuringSchedulingIgnoredDuringExecution:
   nodeSelectorTerms:
   - matchExpressions:
   - key: failure-domain.beta.kubernetes.io/zone
   operator: In
   values:
   - eu-de-02
   - eu-de-01
   containers:
   - args:
   - "360000"
   command:
   - sleep
   image: 10.125.1.72:6443/otc00000000001000010000/busybox:latest
   imagePullPolicy: Always
   name: container01
   ports:
   - containerPort: 80
   protocol: TCP
   resources: {}
   securityContext:
   privileged: true
   terminationMessagePath: /dev/termination-log
   terminationMessagePolicy: File
   dnsPolicy: ClusterFirst
   imagePullSecrets:
   - name: myregistry
   restartPolicy: Always
   schedulerName: default-scheduler
   securityContext: {}
   terminationGracePeriodSeconds: 30
   status:
   replicas: 0
   

   Replace the content in bold in the preceding YAML file to generate a YAML file for Deployments in CCE 2.0.

   apiVersion: apps/v1
   kind: Deployment
   metadata:
     annotations:
       cce/app-createTimestamp: 2019-09-09-08-59-13
       cce/app-description: ""
       cce/app-updateTimestamp: 2019-09-09-08-59-13
     creationTimestamp: null
     generation: 1
     labels:
       cce/appgroup: app-test
       name: ssss
       rollingupdate: "false"
     name: ssss
     selfLink: /api/v1/namespaces/default/replicationcontrollers/ssss
   spec:
     replicas: 1
     selector:
     matchLabels:
       cce/appgroup: app-test
     template:
     metadata:
       annotations:
         scheduler.alpha.kubernetes.io/affinity: '{"nodeAffinity":{"requiredDuringSchedulingIgnoredDuringExecution":{"nodeSelectorTerms":[{"matchExpressions":[{"key
   ":"failure-domain.beta.kubernetes.io/zone","operator":"In","values":["eu-de-02","eu-de-01"]}]}]}}}'
       creationTimestamp: null
       labels:
         cce/appgroup: app-test
       name: ssss
       rollingupdate: "false"
     spec:
       affinity:
         nodeAffinity:
           requiredDuringSchedulingIgnoredDuringExecution:
             nodeSelectorTerms:
             - matchExpressions:
               - key: failure-domain.beta.kubernetes.io/zone
                 operator: In
                 values:
                   - eu-de-02
                   - eu-de-01
       containers:
           - args:
         - "360000"
       command:
         - sleep
       image: 10.125.7.25:20202/otc00000000001000010000/busybox:latest
       imagePullPolicy: Always
       name: container01
       ports:
           - containerPort: 80
       protocol: TCP
       resources: {}
       securityContext:
       privileged: true
       terminationMessagePath: /dev/termination-log
       terminationMessagePolicy: File
       dnsPolicy: ClusterFirst
       imagePullSecrets:
           - name: default-secret
       restartPolicy: Always
       schedulerName: default-scheduler
       securityContext: {}
       terminationGracePeriodSeconds: 30
       status:
         replicas: 0
   

   Note

   Rectify errors as prompted during cluster upgrading. If you have any questions, contact O&M personnel.

last updated: 2024-12-13 13:16 UTC - commit: 637711c1b4cbb67ff8f1429f87a3eb6e1d7d78bc