Merge branch 'master' into version-changes

example/integrations/tensorflow/tf-example.yaml

@@ -37,6 +37,7 @@ spec:
   minAvailable: 2
   schedulerName: kube-batch
   plugins:
+    env: []
     svc: []
   policies:
     - event: PodEvicted

example/job.yaml

@@ -6,30 +6,36 @@ spec:
   minAvailable: 3
   schedulerName: kube-batch
   policies:
-  - event: PodEvicted
-    action: RestartJob
-  input:
-    mountPath: "/myinput"
-  output:
-    mountPath: "/myoutput"
-    volumeClaim:
-      accessModes: [ "ReadWriteOnce" ]
-      storageClassName: "my-storage-class"
-      resources:
-        requests:
-          storage: 1Gi
+    - event: PodEvicted
+      action: RestartJob
+  plugins:
+    ssh: []
+    env: []
+    svc: []
+  maxRetry: 5
+  queue: default
+  volumes:
+    - mountPath: "/myinput"
+    - mountPath: "/myoutput"
+      volumeClaimName: "testvolumeclaimname"
+      volumeClaim:
+        accessModes: [ "ReadWriteOnce" ]
+        storageClassName: "my-storage-class"
+        resources:
+          requests:
+            storage: 1Gi
   tasks:
-  - replicas: 6
-    name: "default-nginx"
-    template:
-      metadata:
-        name: web
-      spec:
-        containers:
-        - image: nginx
-          imagePullPolicy: IfNotPresent
-          name: nginx
-          resources:
-            requests:
-              cpu: "1"
-        restartPolicy: OnFailure
+    - replicas: 6
+      name: "default-nginx"
+      template:
+        metadata:
+          name: web
+        spec:
+          containers:
+            - image: nginx
+              imagePullPolicy: IfNotPresent
+              name: nginx
+              resources:
+                requests:
+                  cpu: "1"
+          restartPolicy: OnFailure

hack/run-e2e-kind.sh

@@ -64,14 +64,20 @@ function install-volcano {
 
   echo "Install volcano plugin into cluster...."
   helm plugin install --kubeconfig ${KUBECONFIG} installer/chart/volcano/plugins/gen-admission-secret
-  helm gen-admission-secret --service integration-admission-service --namespace kube-system
+
+  #If failed to generate secret for admission service, return immediately
+  helm gen-admission-secret --service ${CLUSTER_NAME}-admission-service --namespace kube-system
+  if [[ $? != 0 ]]; then
+    echo "Failed to install secret for admission service, usually we need a retry."
+    exit 1
+  fi
 
   echo "Install volcano chart"
-  helm install installer/chart/volcano --namespace kube-system --name integration --kubeconfig ${KUBECONFIG} --set basic.image_tag_version=${TAG}
+  helm install installer/chart/volcano --namespace kube-system --name ${CLUSTER_NAME} --kubeconfig ${KUBECONFIG} --set basic.image_tag_version=${TAG} --wait
 }
 
 function uninstall-volcano {
-  helm delete integration --purge --kubeconfig ${KUBECONFIG}
+  helm delete ${CLUSTER_NAME} --purge --kubeconfig ${KUBECONFIG}
 }
 
 function generate-log {

installer/chart/volcano/plugins/gen-admission-secret/gen-admission-secret.sh

@@ -108,15 +108,15 @@ done
 # approve and fetch the signed certificate
 kubectl certificate approve ${csrName}
 # verify certificate has been signed
-for x in $(seq 15); do
+for x in $(seq 20); do
     serverCert=$(kubectl get csr ${csrName} -o jsonpath='{.status.certificate}')
     if [[ ${serverCert} != '' ]]; then
         break
     fi
     sleep 1
 done
 if [[ ${serverCert} == '' ]]; then
-    echo "ERROR: After approving csr ${csrName}, the signed certificate did not appear on the resource. Giving up after 15 attempts." >&2
+    echo "ERROR: After approving csr ${csrName}, the signed certificate did not appear on the resource. Giving up after 20 attempts." >&2
     exit 1
 fi
 echo ${serverCert} | openssl base64 -d -A -out ${tmpdir}/server-cert.pem

installer/chart/volcano/templates/batch_v1alpha1_job.yaml

@@ -32,36 +32,27 @@ spec:
           description: Specification of the desired behavior of a cron job, including
             the minAvailable
           properties:
-            input:
-              description: The volume mount for input of Job
-              properties:
-                volumeClaim:
-                  description: VolumeClaim defines the PVC used by the VolumeMount.
-                  type: object
-                mountPath:
-                  description: Path within the container at which the volume should
-                    be mounted.  Must not contain ':'.
-                  type: string
-              required:
-              - mountPath
-              type: object
+            volumes:
+              description: The volumes for Job
+              items:
+                properties:
+                  volumeClaim:
+                    description: VolumeClaim defines the PVC used by the VolumeMount.
+                    type: object
+                  mountPath:
+                    description: Path within the container at which the volume should be mounted.
+                      Must not contain ':'.
+                    type: string
+                  volumeClaimName:
+                    description: The name of the volume claim.
+                type: object
+                required:
+                  - mountPath
+              type: array
             minAvailable:
               description: The minimal available pods to run for this Job
               format: int32
               type: integer
-            output:
-              description: The volume mount for output of Job
-              properties:
-                volumeClaim:
-                  description: VolumeClaim defines the PVC used by the VolumeMount.
-                  type: object
-                mountPath:
-                  description: Path within the container at which the volume should
-                    be mounted.  Must not contain ':'.
-                  type: string
-              required:
-              - mountPath
-              type: object
             policies:
               description: Specifies the default lifecycle of tasks
               items:
@@ -83,6 +74,11 @@ spec:
             schedulerName:
               description: SchedulerName is the default value of `tasks.template.spec.schedulerName`.
               type: string
+            plugins:
+              description: Enabled task plugins when creating job.
+              type: object
+              additionalProperties:
+                type: array
             tasks:
               description: Tasks specifies the task specification of Job
               items:
@@ -120,6 +116,13 @@ spec:
                     type: object
                 type: object
               type: array
+            queue:
+              description: The name of the queue on which job should been created
+              type: string
+            maxRetry:
+              description: The limit for retrying submiting job, default is 3
+              format: int32
+              type: integer
           type: object
         status:
           description: Current status of Job
@@ -148,6 +151,15 @@ spec:
               description: Job's current version
               format: int32
               type: integer
+            retryCount:
+              description: The number that volcano retried to submit the job.
+              format: int32
+              type: integer
+            ControlledResources:
+              description: All of the resources that are controlled by this job.
+              type: object
+              additionalProperties:
+                type: string
             state:
               description: Current state of Job.
               properties:

pkg/apis/batch/v1alpha1/job.go

@@ -90,6 +90,8 @@ type JobEvent string
 const (
 	CommandIssued JobEvent = "CommandIssued"
 	PluginError   JobEvent = "PluginError"
+	PVCError      JobEvent = "PVCError"
+	PodGroupError JobEvent = "PodGroupError"
 )
 
 // Event represent the phase of Job, e.g. pod-failed.