kubevela/docs/examples/workflow

1. Apply CRD and Definitions:

   kubectl apply -f definition.yaml
   

   Check Policy and Workflow definitions:

   kubectl get policy
   kubectl get workflowstep
   

   Output:

   NAME        AGE
   foopolicy   41s
   
   NAME    AGE
   foowf   49s
   

   Check DefinitionRevision:

   kubectl get definitionrevision
   

   Output:

   NAMESPACE     NAME             REVISION   HASH               TYPE
   default       foopolicy-v1     1          8c340e1beaf9a3fa   Policy
   default       foowf-v1         1          83cf4e8246a89afa   WorkflowStep
   

2. Apply Application:

   kubectl apply -f app.yaml
   

3. Check workflow status in Application:

   kubectl get application first-vela-app -o=jsonpath='{.status.workflow[?(@.name=="my-wf")]}.phase'
   

   Output:

   running
   

4. Check Workflow objects:

   kubectl get foo my-wf -o=jsonpath='{.spec.key}'
   

   Output:

   test
   

   This means the resource has been rendered correctly.

5. Check workflow context:

   kubectl get foo my-wf -o=jsonpath='{.metadata.annotations.app\.oam\.dev/workflow-context}' | jq
   

   Output:

   {
     "appName": "first-vela-app",
     "appRevision": "first-vela-app-v1",
     "workflowIndex": 0,
     "resourceConfigMap": {
       "name": "first-vela-app-v1"
     }
   }
   

6. Patch condition status on workflow object:

   kubectl patch foo my-wf --type merge --patch "$(cat wf-patch.yaml)"
   

   Check workflow object status:

   kubectl get foo my-wf -o=jsonpath='{.status.conditions[?(@.type=="workflow-finish")]}' | jq
   

   Output:

   {
     "message": "{\"observedGeneration\":2}",
     "reason": "Succeeded",
     "status": "True",
     "type": "workflow-finish"
   }
   

   Note: The observedGeneration is 2 because the json patch will trigger generation increment.

7. Check workflow status in Application:

   kubectl get application first-vela-app -o=jsonpath='{.status.workflow[?(@.name=="my-wf")]}.phase'
   

   Output:

   succeeded
   

   The workflow phase has changed from running to succeeded due to the underlying object changing status condition.

8. cleanup:

   kubectl delete -f app.yaml
   kubectl delete -f definition.yaml