fix: Can not sync job status correctly when upgrading from v1.5

v1.5 changed the naming logics of pod group by adding UID into the name: volcano-sh#2140, and there is also another fix regarding handling the already created pod group without UID in create or update: volcano-sh#2400. But a similar fix does not exist in the syncJob function.

Fixes volcano-sh#3640

pkg/controllers/job/helpers/helpers.go

@@ -146,3 +146,13 @@ func GetPodsNameUnderTask(taskName string, job *batch.Job) []string {
 	}
 	return res
 }
+
+// GetRelatedPodGroupName returns the name of podgroup related to a job
+func GetRelatedPodGroupName(job *batch.Job) string {
+	return fmt.Sprintf("%s-%s", job.Name, string(job.UID))
+}
+
+// GetRelatedPodGroupLegacyName returns the legacy name of podgroup related to a job
+func GetRelatedPodGroupLegacyName(job *batch.Job) string {
+	return job.Name
+}

pkg/controllers/job/job_controller_actions.go

@@ -152,11 +152,18 @@ func (cc *jobcontroller) killJob(jobInfo *apis.JobInfo, podRetainPhase state.Pha
 	}
 
 	// Delete PodGroup
-	pgName := job.Name + "-" + string(job.UID)
-	if err := cc.vcClient.SchedulingV1beta1().PodGroups(job.Namespace).Delete(context.TODO(), pgName, metav1.DeleteOptions{}); err != nil {
+	pgName := jobhelpers.GetRelatedPodGroupName(job)
+	pgIface := cc.vcClient.SchedulingV1beta1().PodGroups(job.Namespace)
+	if err := pgIface.Delete(context.TODO(), pgName, metav1.DeleteOptions{}); err != nil {
 		if !apierrors.IsNotFound(err) {
-			klog.Errorf("Failed to delete PodGroup of Job %v/%v: %v",
-				job.Namespace, job.Name, err)
+			klog.Errorf("Failed to delete PodGroup of Job %s/%s: %v", job.Namespace, job.Name, err)
+			return err
+		}
+
+		pgLegacyName := jobhelpers.GetRelatedPodGroupLegacyName(job)
+		err := pgIface.Delete(context.TODO(), pgLegacyName, metav1.DeleteOptions{})
+		if err != nil && !apierrors.IsNotFound(err) {
+			klog.Errorf("Failed to delete legacy PodGroup of Job %s/%s: %v", job.Namespace, job.Name, err)
 			return err
 		}
 	}
@@ -281,8 +288,14 @@ func (cc *jobcontroller) syncJob(jobInfo *apis.JobInfo, updateStatus state.Updat
 	}
 
 	var syncTask bool
-	pgName := job.Name + "-" + string(job.UID)
-	if pg, _ := cc.pgLister.PodGroups(job.Namespace).Get(pgName); pg != nil {
+	pgLister := cc.pgLister.PodGroups(job.Namespace)
+	pgName := jobhelpers.GetRelatedPodGroupName(job)
+	pg, err := pgLister.Get(pgName)
+	if err != nil && apierrors.IsNotFound(err) {
+		pgLegacyName := jobhelpers.GetRelatedPodGroupLegacyName(job)
+		pg, _ = pgLister.Get(pgLegacyName)
+	}
+	if pg != nil {
 		if pg.Status.Phase != "" && pg.Status.Phase != scheduling.PodGroupPending {
 			syncTask = true
 		}
@@ -662,18 +675,20 @@ func (cc *jobcontroller) createPVC(job *batch.Job, vcName string, volumeClaim *v
 
 func (cc *jobcontroller) createOrUpdatePodGroup(job *batch.Job) error {
 	// If PodGroup does not exist, create one for Job.
-	pgName := job.Name + "-" + string(job.UID)
+	pgLister := cc.pgLister.PodGroups(job.Namespace)
 	var pg *scheduling.PodGroup
 	var err error
-	pg, err = cc.pgLister.PodGroups(job.Namespace).Get(pgName)
+	pgName := jobhelpers.GetRelatedPodGroupName(job)
+	pg, err = pgLister.Get(pgName)
 	if err != nil {
 		if !apierrors.IsNotFound(err) {
 			klog.Errorf("Failed to get PodGroup for Job <%s/%s>: %v",
 				job.Namespace, job.Name, err)
 			return err
 		}
 		// try to get old pg if new pg not exist
-		pg, err = cc.pgLister.PodGroups(job.Namespace).Get(job.Name)
+		pgLegacyName := jobhelpers.GetRelatedPodGroupLegacyName(job)
+		pg, err = pgLister.Get(pgLegacyName)
 		if err != nil {
 			if !apierrors.IsNotFound(err) {
 				klog.Errorf("Failed to get PodGroup for Job <%s/%s>: %v",