job_controller.go

/*
Copyright 2015 The Kubernetes Authors.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

    http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/

package job

import (
	"fmt"
	"math"
	"reflect"
	"sort"
	"sync"
	"time"

	batch "k8s.io/api/batch/v1"
	"k8s.io/api/core/v1"
	"k8s.io/apimachinery/pkg/api/errors"
	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
	"k8s.io/apimachinery/pkg/labels"
	utilruntime "k8s.io/apimachinery/pkg/util/runtime"
	"k8s.io/apimachinery/pkg/util/wait"
	batchinformers "k8s.io/client-go/informers/batch/v1"
	coreinformers "k8s.io/client-go/informers/core/v1"
	clientset "k8s.io/client-go/kubernetes"
	"k8s.io/client-go/kubernetes/scheme"
	v1core "k8s.io/client-go/kubernetes/typed/core/v1"
	batchv1listers "k8s.io/client-go/listers/batch/v1"
	corelisters "k8s.io/client-go/listers/core/v1"
	"k8s.io/client-go/tools/cache"
	"k8s.io/client-go/tools/record"
	"k8s.io/client-go/util/integer"
	"k8s.io/client-go/util/workqueue"
	"k8s.io/kubernetes/pkg/controller"
	"k8s.io/kubernetes/pkg/util/metrics"

	"github.com/golang/glog"
)

// controllerKind contains the schema.GroupVersionKind for this controller type.
var controllerKind = batch.SchemeGroupVersion.WithKind("Job")

var (
	// DefaultJobBackOff is the max backoff period, exported for the e2e test
	DefaultJobBackOff = 10 * time.Second
	// MaxJobBackOff is the max backoff period, exported for the e2e test
	MaxJobBackOff = 360 * time.Second
)

type JobController struct {
	kubeClient clientset.Interface
	podControl controller.PodControlInterface

	// To allow injection of updateJobStatus for testing.
	updateHandler func(job *batch.Job) error
	syncHandler   func(jobKey string) (bool, error)
	// podStoreSynced returns true if the pod store has been synced at least once.
	// Added as a member to the struct to allow injection for testing.
	podStoreSynced cache.InformerSynced
	// jobStoreSynced returns true if the job store has been synced at least once.
	// Added as a member to the struct to allow injection for testing.
	jobStoreSynced cache.InformerSynced

	// A TTLCache of pod creates/deletes each rc expects to see
	expectations controller.ControllerExpectationsInterface

	// A store of jobs
	jobLister batchv1listers.JobLister

	// A store of pods, populated by the podController
	podStore corelisters.PodLister

	// Jobs that need to be updated
	queue workqueue.RateLimitingInterface

	recorder record.EventRecorder
}

func NewJobController(podInformer coreinformers.PodInformer, jobInformer batchinformers.JobInformer, kubeClient clientset.Interface) *JobController {
	eventBroadcaster := record.NewBroadcaster()
	eventBroadcaster.StartLogging(glog.Infof)
	eventBroadcaster.StartRecordingToSink(&v1core.EventSinkImpl{Interface: kubeClient.CoreV1().Events("")})

	if kubeClient != nil && kubeClient.CoreV1().RESTClient().GetRateLimiter() != nil {
		metrics.RegisterMetricAndTrackRateLimiterUsage("job_controller", kubeClient.CoreV1().RESTClient().GetRateLimiter())
	}

	jm := &JobController{
		kubeClient: kubeClient,
		podControl: controller.RealPodControl{
			KubeClient: kubeClient,
			Recorder:   eventBroadcaster.NewRecorder(scheme.Scheme, v1.EventSource{Component: "job-controller"}),
		},
		expectations: controller.NewControllerExpectations(),
		queue:        workqueue.NewNamedRateLimitingQueue(workqueue.NewItemExponentialFailureRateLimiter(DefaultJobBackOff, MaxJobBackOff), "job"),
		recorder:     eventBroadcaster.NewRecorder(scheme.Scheme, v1.EventSource{Component: "job-controller"}),
	}

	jobInformer.Informer().AddEventHandler(cache.ResourceEventHandlerFuncs{
		AddFunc: func(obj interface{}) {
			jm.enqueueController(obj, true)
		},
		UpdateFunc: jm.updateJob,
		DeleteFunc: func(obj interface{}) {
			jm.enqueueController(obj, true)
		},
	})
	jm.jobLister = jobInformer.Lister()
	jm.jobStoreSynced = jobInformer.Informer().HasSynced

	podInformer.Informer().AddEventHandler(cache.ResourceEventHandlerFuncs{
		AddFunc:    jm.addPod,
		UpdateFunc: jm.updatePod,
		DeleteFunc: jm.deletePod,
	})
	jm.podStore = podInformer.Lister()
	jm.podStoreSynced = podInformer.Informer().HasSynced

	jm.updateHandler = jm.updateJobStatus
	jm.syncHandler = jm.syncJob

	return jm
}

// Run the main goroutine responsible for watching and syncing jobs.
func (jm *JobController) Run(workers int, stopCh <-chan struct{}) {
	defer utilruntime.HandleCrash()
	defer jm.queue.ShutDown()

	glog.Infof("Starting job controller")
	defer glog.Infof("Shutting down job controller")

	if !controller.WaitForCacheSync("job", stopCh, jm.podStoreSynced, jm.jobStoreSynced) {
		return
	}

	for i := 0; i < workers; i++ {
		go wait.Until(jm.worker, time.Second, stopCh)
	}

	<-stopCh
}

// getPodJobs returns a list of Jobs that potentially match a Pod.
func (jm *JobController) getPodJobs(pod *v1.Pod) []*batch.Job {
	jobs, err := jm.jobLister.GetPodJobs(pod)
	if err != nil {
		return nil
	}
	if len(jobs) > 1 {
		// ControllerRef will ensure we don't do anything crazy, but more than one
		// item in this list nevertheless constitutes user error.
		utilruntime.HandleError(fmt.Errorf("user error! more than one job is selecting pods with labels: %+v", pod.Labels))
	}
	ret := make([]*batch.Job, 0, len(jobs))
	for i := range jobs {
		ret = append(ret, &jobs[i])
	}
	return ret
}

// resolveControllerRef returns the controller referenced by a ControllerRef,
// or nil if the ControllerRef could not be resolved to a matching controller
// of the correct Kind.
func (jm *JobController) resolveControllerRef(namespace string, controllerRef *metav1.OwnerReference) *batch.Job {
	// We can't look up by UID, so look up by Name and then verify UID.
	// Don't even try to look up by Name if it's the wrong Kind.
	if controllerRef.Kind != controllerKind.Kind {
		return nil
	}
	job, err := jm.jobLister.Jobs(namespace).Get(controllerRef.Name)
	if err != nil {
		return nil
	}
	if job.UID != controllerRef.UID {
		// The controller we found with this Name is not the same one that the
		// ControllerRef points to.
		return nil
	}
	return job
}

// When a pod is created, enqueue the controller that manages it and update it's expectations.
func (jm *JobController) addPod(obj interface{}) {
	pod := obj.(*v1.Pod)
	if pod.DeletionTimestamp != nil {
		// on a restart of the controller controller, it's possible a new pod shows up in a state that
		// is already pending deletion. Prevent the pod from being a creation observation.
		jm.deletePod(pod)
		return
	}

	// If it has a ControllerRef, that's all that matters.
	if controllerRef := metav1.GetControllerOf(pod); controllerRef != nil {
		job := jm.resolveControllerRef(pod.Namespace, controllerRef)
		if job == nil {
			return
		}
		jobKey, err := controller.KeyFunc(job)
		if err != nil {
			return
		}
		jm.expectations.CreationObserved(jobKey)
		jm.enqueueController(job, true)
		return
	}

	// Otherwise, it's an orphan. Get a list of all matching controllers and sync
	// them to see if anyone wants to adopt it.
	// DO NOT observe creation because no controller should be waiting for an
	// orphan.
	for _, job := range jm.getPodJobs(pod) {
		jm.enqueueController(job, true)
	}
}

// When a pod is updated, figure out what job/s manage it and wake them up.
// If the labels of the pod have changed we need to awaken both the old
// and new job. old and cur must be *v1.Pod types.
func (jm *JobController) updatePod(old, cur interface{}) {
	curPod := cur.(*v1.Pod)
	oldPod := old.(*v1.Pod)
	if curPod.ResourceVersion == oldPod.ResourceVersion {
		// Periodic resync will send update events for all known pods.
		// Two different versions of the same pod will always have different RVs.
		return
	}
	if curPod.DeletionTimestamp != nil {
		// when a pod is deleted gracefully it's deletion timestamp is first modified to reflect a grace period,
		// and after such time has passed, the kubelet actually deletes it from the store. We receive an update
		// for modification of the deletion timestamp and expect an job to create more pods asap, not wait
		// until the kubelet actually deletes the pod.
		jm.deletePod(curPod)
		return
	}

	// the only time we want the backoff to kick-in, is when the pod failed
	immediate := curPod.Status.Phase != v1.PodFailed

	curControllerRef := metav1.GetControllerOf(curPod)
	oldControllerRef := metav1.GetControllerOf(oldPod)
	controllerRefChanged := !reflect.DeepEqual(curControllerRef, oldControllerRef)
	if controllerRefChanged && oldControllerRef != nil {
		// The ControllerRef was changed. Sync the old controller, if any.
		if job := jm.resolveControllerRef(oldPod.Namespace, oldControllerRef); job != nil {
			jm.enqueueController(job, immediate)
		}
	}

	// If it has a ControllerRef, that's all that matters.
	if curControllerRef != nil {
		job := jm.resolveControllerRef(curPod.Namespace, curControllerRef)
		if job == nil {
			return
		}
		jm.enqueueController(job, immediate)
		return
	}

	// Otherwise, it's an orphan. If anything changed, sync matching controllers
	// to see if anyone wants to adopt it now.
	labelChanged := !reflect.DeepEqual(curPod.Labels, oldPod.Labels)
	if labelChanged || controllerRefChanged {
		for _, job := range jm.getPodJobs(curPod) {
			jm.enqueueController(job, immediate)
		}
	}
}

// When a pod is deleted, enqueue the job that manages the pod and update its expectations.
// obj could be an *v1.Pod, or a DeletionFinalStateUnknown marker item.
func (jm *JobController) deletePod(obj interface{}) {
	pod, ok := obj.(*v1.Pod)

	// When a delete is dropped, the relist will notice a pod in the store not
	// in the list, leading to the insertion of a tombstone object which contains
	// the deleted key/value. Note that this value might be stale. If the pod
	// changed labels the new job will not be woken up till the periodic resync.
	if !ok {
		tombstone, ok := obj.(cache.DeletedFinalStateUnknown)
		if !ok {
			utilruntime.HandleError(fmt.Errorf("couldn't get object from tombstone %+v", obj))
			return
		}
		pod, ok = tombstone.Obj.(*v1.Pod)
		if !ok {
			utilruntime.HandleError(fmt.Errorf("tombstone contained object that is not a pod %+v", obj))
			return
		}
	}

	controllerRef := metav1.GetControllerOf(pod)
	if controllerRef == nil {
		// No controller should care about orphans being deleted.
		return
	}
	job := jm.resolveControllerRef(pod.Namespace, controllerRef)
	if job == nil {
		return
	}
	jobKey, err := controller.KeyFunc(job)
	if err != nil {
		return
	}
	jm.expectations.DeletionObserved(jobKey)
	jm.enqueueController(job, true)
}

func (jm *JobController) updateJob(old, cur interface{}) {
	oldJob := cur.(*batch.Job)
	curJob := cur.(*batch.Job)

	// never return error
	key, err := controller.KeyFunc(curJob)
	if err != nil {
		return
	}
	jm.enqueueController(curJob, true)
	// check if need to add a new rsync for ActiveDeadlineSeconds
	if curJob.Status.StartTime != nil {
		curADS := curJob.Spec.ActiveDeadlineSeconds
		if curADS == nil {
			return
		}
		oldADS := oldJob.Spec.ActiveDeadlineSeconds
		if oldADS == nil || *oldADS != *curADS {
			now := metav1.Now()
			start := curJob.Status.StartTime.Time
			passed := now.Time.Sub(start)
			total := time.Duration(*curADS) * time.Second
			// AddAfter will handle total < passed
			jm.queue.AddAfter(key, total-passed)
			glog.V(4).Infof("job ActiveDeadlineSeconds updated, will rsync after %d seconds", total-passed)
		}
	}
}

// obj could be an *batch.Job, or a DeletionFinalStateUnknown marker item,
// immediate tells the controller to update the status right away, and should
// happen ONLY when there was a successful pod run.
func (jm *JobController) enqueueController(obj interface{}, immediate bool) {
	key, err := controller.KeyFunc(obj)
	if err != nil {
		utilruntime.HandleError(fmt.Errorf("Couldn't get key for object %+v: %v", obj, err))
		return
	}

	if immediate {
		jm.queue.Forget(key)
	}
	backoff := getBackoff(jm.queue, key)

	// TODO: Handle overlapping controllers better. Either disallow them at admission time or
	// deterministically avoid syncing controllers that fight over pods. Currently, we only
	// ensure that the same controller is synced for a given pod. When we periodically relist
	// all controllers there will still be some replica instability. One way to handle this is
	// by querying the store for all controllers that this rc overlaps, as well as all
	// controllers that overlap this rc, and sorting them.
	jm.queue.AddAfter(key, backoff)
}

// worker runs a worker thread that just dequeues items, processes them, and marks them done.
// It enforces that the syncHandler is never invoked concurrently with the same key.
func (jm *JobController) worker() {
	for jm.processNextWorkItem() {
	}
}

func (jm *JobController) processNextWorkItem() bool {
	key, quit := jm.queue.Get()
	if quit {
		return false
	}
	defer jm.queue.Done(key)

	forget, err := jm.syncHandler(key.(string))
	if err == nil {
		if forget {
			jm.queue.Forget(key)
		}
		return true
	}

	utilruntime.HandleError(fmt.Errorf("Error syncing job: %v", err))
	jm.queue.AddRateLimited(key)

	return true
}

// getPodsForJob returns the set of pods that this Job should manage.
// It also reconciles ControllerRef by adopting/orphaning.
// Note that the returned Pods are pointers into the cache.
func (jm *JobController) getPodsForJob(j *batch.Job) ([]*v1.Pod, error) {
	selector, err := metav1.LabelSelectorAsSelector(j.Spec.Selector)
	if err != nil {
		return nil, fmt.Errorf("couldn't convert Job selector: %v", err)
	}
	// List all pods to include those that don't match the selector anymore
	// but have a ControllerRef pointing to this controller.
	pods, err := jm.podStore.Pods(j.Namespace).List(labels.Everything())
	if err != nil {
		return nil, err
	}
	// If any adoptions are attempted, we should first recheck for deletion
	// with an uncached quorum read sometime after listing Pods (see #42639).
	canAdoptFunc := controller.RecheckDeletionTimestamp(func() (metav1.Object, error) {
		fresh, err := jm.kubeClient.BatchV1().Jobs(j.Namespace).Get(j.Name, metav1.GetOptions{})
		if err != nil {
			return nil, err
		}
		if fresh.UID != j.UID {
			return nil, fmt.Errorf("original Job %v/%v is gone: got uid %v, wanted %v", j.Namespace, j.Name, fresh.UID, j.UID)
		}
		return fresh, nil
	})
	cm := controller.NewPodControllerRefManager(jm.podControl, j, selector, controllerKind, canAdoptFunc)
	return cm.ClaimPods(pods)
}

// syncJob will sync the job with the given key if it has had its expectations fulfilled, meaning
// it did not expect to see any more of its pods created or deleted. This function is not meant to be invoked
// concurrently with the same key.
func (jm *JobController) syncJob(key string) (bool, error) {
	startTime := time.Now()
	defer func() {
		glog.V(4).Infof("Finished syncing job %q (%v)", key, time.Since(startTime))
	}()

	ns, name, err := cache.SplitMetaNamespaceKey(key)
	if err != nil {
		return false, err
	}
	if len(ns) == 0 || len(name) == 0 {
		return false, fmt.Errorf("invalid job key %q: either namespace or name is missing", key)
	}
	sharedJob, err := jm.jobLister.Jobs(ns).Get(name)
	if err != nil {
		if errors.IsNotFound(err) {
			glog.V(4).Infof("Job has been deleted: %v", key)
			jm.expectations.DeleteExpectations(key)
			return true, nil
		}
		return false, err
	}
	job := *sharedJob

	// if job was finished previously, we don't want to redo the termination
	if IsJobFinished(&job) {
		return true, nil
	}

	// retrieve the previous number of retry
	previousRetry := jm.queue.NumRequeues(key)

	// Check the expectations of the job before counting active pods, otherwise a new pod can sneak in
	// and update the expectations after we've retrieved active pods from the store. If a new pod enters
	// the store after we've checked the expectation, the job sync is just deferred till the next relist.
	jobNeedsSync := jm.expectations.SatisfiedExpectations(key)

	pods, err := jm.getPodsForJob(&job)
	if err != nil {
		return false, err
	}

	activePods := controller.FilterActivePods(pods)
	active := int32(len(activePods))
	succeeded, failed := getStatus(pods)
	conditions := len(job.Status.Conditions)
	// job first start
	if job.Status.StartTime == nil {
		now := metav1.Now()
		job.Status.StartTime = &now
		// enqueue a sync to check if job past ActiveDeadlineSeconds
		if job.Spec.ActiveDeadlineSeconds != nil {
			glog.V(4).Infof("Job %s have ActiveDeadlineSeconds will sync after %d seconds",
				key, *job.Spec.ActiveDeadlineSeconds)
			jm.queue.AddAfter(key, time.Duration(*job.Spec.ActiveDeadlineSeconds)*time.Second)
		}
	}

	var manageJobErr error
	jobFailed := false
	var failureReason string
	var failureMessage string

	jobHaveNewFailure := failed > job.Status.Failed
	exceedsBackoffLimit := jobHaveNewFailure && (int32(previousRetry)+1 > *job.Spec.BackoffLimit)

	if exceedsBackoffLimit || pastBackoffLimitOnFailure(&job, pods) {
		// check if the number of pod restart exceeds backoff (for restart OnFailure only)
		// OR if the number of failed jobs increased since the last syncJob
		jobFailed = true
		failureReason = "BackoffLimitExceeded"
		failureMessage = "Job has reached the specified backoff limit"
	} else if pastActiveDeadline(&job) {
		jobFailed = true
		failureReason = "DeadlineExceeded"
		failureMessage = "Job was active longer than specified deadline"
	}

	if jobFailed {
		errCh := make(chan error, active)
		jm.deleteJobPods(&job, activePods, errCh)
		select {
		case manageJobErr = <-errCh:
			if manageJobErr != nil {
				break
			}
		default:
		}

		// update status values accordingly
		failed += active
		active = 0
		job.Status.Conditions = append(job.Status.Conditions, newCondition(batch.JobFailed, failureReason, failureMessage))
		jm.recorder.Event(&job, v1.EventTypeWarning, failureReason, failureMessage)
	} else {
		if jobNeedsSync && job.DeletionTimestamp == nil {
			active, manageJobErr = jm.manageJob(activePods, succeeded, &job)
		}
		completions := succeeded
		complete := false
		if job.Spec.Completions == nil {
			// This type of job is complete when any pod exits with success.
			// Each pod is capable of
			// determining whether or not the entire Job is done.  Subsequent pods are
			// not expected to fail, but if they do, the failure is ignored.  Once any
			// pod succeeds, the controller waits for remaining pods to finish, and
			// then the job is complete.
			if succeeded > 0 && active == 0 {
				complete = true
			}
		} else {
			// Job specifies a number of completions.  This type of job signals
			// success by having that number of successes.  Since we do not
			// start more pods than there are remaining completions, there should
			// not be any remaining active pods once this count is reached.
			if completions >= *job.Spec.Completions {
				complete = true
				if active > 0 {
					jm.recorder.Event(&job, v1.EventTypeWarning, "TooManyActivePods", "Too many active pods running after completion count reached")
				}
				if completions > *job.Spec.Completions {
					jm.recorder.Event(&job, v1.EventTypeWarning, "TooManySucceededPods", "Too many succeeded pods running after completion count reached")
				}
			}
		}
		if complete {
			job.Status.Conditions = append(job.Status.Conditions, newCondition(batch.JobComplete, "", ""))
			now := metav1.Now()
			job.Status.CompletionTime = &now
		}
	}

	forget := false
	// Check if the number of jobs succeeded increased since the last check. If yes "forget" should be true
	// This logic is linked to the issue: https://github.com/kubernetes/kubernetes/issues/56853 that aims to
	// improve the Job backoff policy when parallelism > 1 and few Jobs failed but others succeed.
	// In this case, we should clear the backoff delay.
	if job.Status.Succeeded < succeeded {
		forget = true
	}

	// no need to update the job if the status hasn't changed since last time
	if job.Status.Active != active || job.Status.Succeeded != succeeded || job.Status.Failed != failed || len(job.Status.Conditions) != conditions {
		job.Status.Active = active
		job.Status.Succeeded = succeeded
		job.Status.Failed = failed

		if err := jm.updateHandler(&job); err != nil {
			return forget, err
		}

		if jobHaveNewFailure && !IsJobFinished(&job) {
			// returning an error will re-enqueue Job after the backoff period
			return forget, fmt.Errorf("failed pod(s) detected for job key %q", key)
		}

		forget = true
	}

	return forget, manageJobErr
}

func (jm *JobController) deleteJobPods(job *batch.Job, pods []*v1.Pod, errCh chan<- error) {
	// TODO: below code should be replaced with pod termination resulting in
	// pod failures, rather than killing pods. Unfortunately none such solution
	// exists ATM. There's an open discussion in the topic in
	// https://github.com/kubernetes/kubernetes/issues/14602 which might give
	// some sort of solution to above problem.
	// kill remaining active pods
	wait := sync.WaitGroup{}
	nbPods := len(pods)
	wait.Add(nbPods)
	for i := int32(0); i < int32(nbPods); i++ {
		go func(ix int32) {
			defer wait.Done()
			if err := jm.podControl.DeletePod(job.Namespace, pods[ix].Name, job); err != nil {
				defer utilruntime.HandleError(err)
				glog.V(2).Infof("Failed to delete %v, job %q/%q deadline exceeded", pods[ix].Name, job.Namespace, job.Name)
				errCh <- err
			}
		}(i)
	}
	wait.Wait()
}

// pastBackoffLimitOnFailure checks if container restartCounts sum exceeds BackoffLimit
// this method applies only to pods with restartPolicy == OnFailure
func pastBackoffLimitOnFailure(job *batch.Job, pods []*v1.Pod) bool {
	if job.Spec.Template.Spec.RestartPolicy != v1.RestartPolicyOnFailure {
		return false
	}
	result := int32(0)
	for i := range pods {
		po := pods[i]
		if po.Status.Phase != v1.PodRunning {
			continue
		}
		for j := range po.Status.InitContainerStatuses {
			stat := po.Status.InitContainerStatuses[j]
			result += stat.RestartCount
		}
		for j := range po.Status.ContainerStatuses {
			stat := po.Status.ContainerStatuses[j]
			result += stat.RestartCount
		}
	}
	return result >= *job.Spec.BackoffLimit
}

// pastActiveDeadline checks if job has ActiveDeadlineSeconds field set and if it is exceeded.
func pastActiveDeadline(job *batch.Job) bool {
	if job.Spec.ActiveDeadlineSeconds == nil || job.Status.StartTime == nil {
		return false
	}
	now := metav1.Now()
	start := job.Status.StartTime.Time
	duration := now.Time.Sub(start)
	allowedDuration := time.Duration(*job.Spec.ActiveDeadlineSeconds) * time.Second
	return duration >= allowedDuration
}

func newCondition(conditionType batch.JobConditionType, reason, message string) batch.JobCondition {
	return batch.JobCondition{
		Type:               conditionType,
		Status:             v1.ConditionTrue,
		LastProbeTime:      metav1.Now(),
		LastTransitionTime: metav1.Now(),
		Reason:             reason,
		Message:            message,
	}
}

// getStatus returns no of succeeded and failed pods running a job
func getStatus(pods []*v1.Pod) (succeeded, failed int32) {
	succeeded = int32(filterPods(pods, v1.PodSucceeded))
	failed = int32(filterPods(pods, v1.PodFailed))
	return
}

// manageJob is the core method responsible for managing the number of running
// pods according to what is specified in the job.Spec.
// Does NOT modify <activePods>.
func (jm *JobController) manageJob(activePods []*v1.Pod, succeeded int32, job *batch.Job) (int32, error) {
	var activeLock sync.Mutex
	active := int32(len(activePods))
	parallelism := *job.Spec.Parallelism
	jobKey, err := controller.KeyFunc(job)
	if err != nil {
		utilruntime.HandleError(fmt.Errorf("Couldn't get key for job %#v: %v", job, err))
		return 0, nil
	}

	var errCh chan error
	if active > parallelism {
		diff := active - parallelism
		errCh = make(chan error, diff)
		jm.expectations.ExpectDeletions(jobKey, int(diff))
		glog.V(4).Infof("Too many pods running job %q, need %d, deleting %d", jobKey, parallelism, diff)
		// Sort the pods in the order such that not-ready < ready, unscheduled
		// < scheduled, and pending < running. This ensures that we delete pods
		// in the earlier stages whenever possible.
		sort.Sort(controller.ActivePods(activePods))

		active -= diff
		wait := sync.WaitGroup{}
		wait.Add(int(diff))
		for i := int32(0); i < diff; i++ {
			go func(ix int32) {
				defer wait.Done()
				if err := jm.podControl.DeletePod(job.Namespace, activePods[ix].Name, job); err != nil {
					defer utilruntime.HandleError(err)
					// Decrement the expected number of deletes because the informer won't observe this deletion
					glog.V(2).Infof("Failed to delete %v, decrementing expectations for job %q/%q", activePods[ix].Name, job.Namespace, job.Name)
					jm.expectations.DeletionObserved(jobKey)
					activeLock.Lock()
					active++
					activeLock.Unlock()
					errCh <- err
				}
			}(i)
		}
		wait.Wait()

	} else if active < parallelism {
		wantActive := int32(0)
		if job.Spec.Completions == nil {
			// Job does not specify a number of completions.  Therefore, number active
			// should be equal to parallelism, unless the job has seen at least
			// once success, in which leave whatever is running, running.
			if succeeded > 0 {
				wantActive = active
			} else {
				wantActive = parallelism
			}
		} else {
			// Job specifies a specific number of completions.  Therefore, number
			// active should not ever exceed number of remaining completions.
			wantActive = *job.Spec.Completions - succeeded
			if wantActive > parallelism {
				wantActive = parallelism
			}
		}
		diff := wantActive - active
		if diff < 0 {
			utilruntime.HandleError(fmt.Errorf("More active than wanted: job %q, want %d, have %d", jobKey, wantActive, active))
			diff = 0
		}
		jm.expectations.ExpectCreations(jobKey, int(diff))
		errCh = make(chan error, diff)
		glog.V(4).Infof("Too few pods running job %q, need %d, creating %d", jobKey, wantActive, diff)

		active += diff
		wait := sync.WaitGroup{}

		// Batch the pod creates. Batch sizes start at SlowStartInitialBatchSize
		// and double with each successful iteration in a kind of "slow start".
		// This handles attempts to start large numbers of pods that would
		// likely all fail with the same error. For example a project with a
		// low quota that attempts to create a large number of pods will be
		// prevented from spamming the API service with the pod create requests
		// after one of its pods fails.  Conveniently, this also prevents the
		// event spam that those failures would generate.
		for batchSize := int32(integer.IntMin(int(diff), controller.SlowStartInitialBatchSize)); diff > 0; batchSize = integer.Int32Min(2*batchSize, diff) {
			errorCount := len(errCh)
			wait.Add(int(batchSize))
			for i := int32(0); i < batchSize; i++ {
				go func() {
					defer wait.Done()
					err := jm.podControl.CreatePodsWithControllerRef(job.Namespace, &job.Spec.Template, job, metav1.NewControllerRef(job, controllerKind))
					if err != nil && errors.IsTimeout(err) {
						// Pod is created but its initialization has timed out.
						// If the initialization is successful eventually, the
						// controller will observe the creation via the informer.
						// If the initialization fails, or if the pod keeps
						// uninitialized for a long time, the informer will not
						// receive any update, and the controller will create a new
						// pod when the expectation expires.
						return
					}
					if err != nil {
						defer utilruntime.HandleError(err)
						// Decrement the expected number of creates because the informer won't observe this pod
						glog.V(2).Infof("Failed creation, decrementing expectations for job %q/%q", job.Namespace, job.Name)
						jm.expectations.CreationObserved(jobKey)
						activeLock.Lock()
						active--
						activeLock.Unlock()
						errCh <- err
					}
				}()
			}
			wait.Wait()
			// any skipped pods that we never attempted to start shouldn't be expected.
			skippedPods := diff - batchSize
			if errorCount < len(errCh) && skippedPods > 0 {
				glog.V(2).Infof("Slow-start failure. Skipping creation of %d pods, decrementing expectations for job %q/%q", skippedPods, job.Namespace, job.Name)
				active -= skippedPods
				for i := int32(0); i < skippedPods; i++ {
					// Decrement the expected number of creates because the informer won't observe this pod
					jm.expectations.CreationObserved(jobKey)
				}
				// The skipped pods will be retried later. The next controller resync will
				// retry the slow start process.
				break
			}
			diff -= batchSize
		}
	}

	select {
	case err := <-errCh:
		// all errors have been reported before, we only need to inform the controller that there was an error and it should re-try this job once more next time.
		if err != nil {
			return active, err
		}
	default:
	}

	return active, nil
}

func (jm *JobController) updateJobStatus(job *batch.Job) error {
	_, err := jm.kubeClient.BatchV1().Jobs(job.Namespace).UpdateStatus(job)
	return err
}

func getBackoff(queue workqueue.RateLimitingInterface, key interface{}) time.Duration {
	exp := queue.NumRequeues(key)

	if exp <= 0 {
		return time.Duration(0)
	}

	// The backoff is capped such that 'calculated' value never overflows.
	backoff := float64(DefaultJobBackOff.Nanoseconds()) * math.Pow(2, float64(exp-1))
	if backoff > math.MaxInt64 {
		return MaxJobBackOff
	}

	calculated := time.Duration(backoff)
	if calculated > MaxJobBackOff {
		return MaxJobBackOff
	}
	return calculated
}

// filterPods returns pods based on their phase.
func filterPods(pods []*v1.Pod, phase v1.PodPhase) int {
	result := 0
	for i := range pods {
		if phase == pods[i].Status.Phase {
			result++
		}
	}
	return result
}

// byCreationTimestamp sorts a list by creation timestamp, using their names as a tie breaker.
type byCreationTimestamp []batch.Job

func (o byCreationTimestamp) Len() int      { return len(o) }
func (o byCreationTimestamp) Swap(i, j int) { o[i], o[j] = o[j], o[i] }

func (o byCreationTimestamp) Less(i, j int) bool {
	if o[i].CreationTimestamp.Equal(&o[j].CreationTimestamp) {
		return o[i].Name < o[j].Name
	}
	return o[i].CreationTimestamp.Before(&o[j].CreationTimestamp)
}