Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

try get get old pg when new pg not exist #2400

Merged
merged 2 commits into from
Aug 1, 2022
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
82 changes: 47 additions & 35 deletions pkg/controllers/job/job_controller_actions.go
Original file line number Diff line number Diff line change
Expand Up @@ -641,51 +641,63 @@ 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)
pg, err := cc.pgLister.PodGroups(job.Namespace).Get(pgName)
var pg *scheduling.PodGroup
var err error
pg, err = cc.pgLister.PodGroups(job.Namespace).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
}
} else {
// try to get old pg if new pg not exist
pg, err = cc.pgLister.PodGroups(job.Namespace).Get(job.Name)
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
}

minTaskMember := map[string]int32{}
for _, task := range job.Spec.Tasks {
if task.MinAvailable != nil {
minTaskMember[task.Name] = *task.MinAvailable
} else {
minTaskMember[task.Name] = task.Replicas
}
}
minTaskMember := map[string]int32{}
for _, task := range job.Spec.Tasks {
if task.MinAvailable != nil {
minTaskMember[task.Name] = *task.MinAvailable
} else {
minTaskMember[task.Name] = task.Replicas
}
}

pg := &scheduling.PodGroup{
ObjectMeta: metav1.ObjectMeta{
Namespace: job.Namespace,
//add job.UID into its name when create new PodGroup
Name: pgName,
Annotations: job.Annotations,
Labels: job.Labels,
OwnerReferences: []metav1.OwnerReference{
*metav1.NewControllerRef(job, helpers.JobKind),
},
},
Spec: scheduling.PodGroupSpec{
MinMember: job.Spec.MinAvailable,
MinTaskMember: minTaskMember,
Queue: job.Spec.Queue,
MinResources: cc.calcPGMinResources(job),
PriorityClassName: job.Spec.PriorityClassName,
},
}
pg := &scheduling.PodGroup{
ObjectMeta: metav1.ObjectMeta{
Namespace: job.Namespace,
//add job.UID into its name when create new PodGroup
Name: pgName,
Annotations: job.Annotations,
Labels: job.Labels,
OwnerReferences: []metav1.OwnerReference{
*metav1.NewControllerRef(job, helpers.JobKind),
},
},
Spec: scheduling.PodGroupSpec{
MinMember: job.Spec.MinAvailable,
MinTaskMember: minTaskMember,
Queue: job.Spec.Queue,
MinResources: cc.calcPGMinResources(job),
PriorityClassName: job.Spec.PriorityClassName,
},
}

if _, err = cc.vcClient.SchedulingV1beta1().PodGroups(job.Namespace).Create(context.TODO(), pg, metav1.CreateOptions{}); err != nil {
if !apierrors.IsAlreadyExists(err) {
klog.Errorf("Failed to create PodGroup for Job <%s/%s>: %v",
job.Namespace, job.Name, err)
return err
if _, err = cc.vcClient.SchedulingV1beta1().PodGroups(job.Namespace).Create(context.TODO(), pg, metav1.CreateOptions{}); err != nil {
if !apierrors.IsAlreadyExists(err) {
klog.Errorf("Failed to create PodGroup for Job <%s/%s>: %v",
job.Namespace, job.Name, err)
return err
}
}
return nil
}
}
return nil
}

pgShouldUpdate := false
Expand Down