Merge pull request kubernetes-retired#584 from k82cn/fifo_2_wq

Replaced FIFO by workqueue.

pkg/scheduler/cache/cache.go

@@ -26,6 +26,7 @@ import (
 
 	"k8s.io/api/core/v1"
 	metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
+	"k8s.io/apimachinery/pkg/util/wait"
 	"k8s.io/client-go/informers"
 	infov1 "k8s.io/client-go/informers/core/v1"
 	policyv1 "k8s.io/client-go/informers/policy/v1beta1"
@@ -35,6 +36,7 @@ import (
 	"k8s.io/client-go/rest"
 	"k8s.io/client-go/tools/cache"
 	"k8s.io/client-go/tools/record"
+	"k8s.io/client-go/util/workqueue"
 	podutil "k8s.io/kubernetes/pkg/api/v1/pod"
 	"k8s.io/kubernetes/pkg/scheduler/volumebinder"
 
@@ -79,8 +81,8 @@ type SchedulerCache struct {
 	Nodes  map[string]*kbapi.NodeInfo
 	Queues map[kbapi.QueueID]*kbapi.QueueInfo
 
-	errTasks    *cache.FIFO
-	deletedJobs *cache.FIFO
+	errTasks    workqueue.RateLimitingInterface
+	deletedJobs workqueue.RateLimitingInterface
 
 	namespaceAsQueue bool
 }
@@ -164,41 +166,13 @@ func (dvb *defaultVolumeBinder) BindVolumes(task *api.TaskInfo) error {
 	return dvb.volumeBinder.Binder.BindPodVolumes(task.Pod)
 }
 
-func taskKey(obj interface{}) (string, error) {
-	if obj == nil {
-		return "", fmt.Errorf("the object is nil")
-	}
-
-	task, ok := obj.(*kbapi.TaskInfo)
-
-	if !ok {
-		return "", fmt.Errorf("failed to convert %v to TaskInfo", obj)
-	}
-
-	return string(task.UID), nil
-}
-
-func jobKey(obj interface{}) (string, error) {
-	if obj == nil {
-		return "", fmt.Errorf("the object is nil")
-	}
-
-	job, ok := obj.(*kbapi.JobInfo)
-
-	if !ok {
-		return "", fmt.Errorf("failed to convert %v to TaskInfo", obj)
-	}
-
-	return string(job.UID), nil
-}
-
 func newSchedulerCache(config *rest.Config, schedulerName string, nsAsQueue bool) *SchedulerCache {
 	sc := &SchedulerCache{
 		Jobs:             make(map[kbapi.JobID]*kbapi.JobInfo),
 		Nodes:            make(map[string]*kbapi.NodeInfo),
 		Queues:           make(map[kbapi.QueueID]*kbapi.QueueInfo),
-		errTasks:         cache.NewFIFO(taskKey),
-		deletedJobs:      cache.NewFIFO(jobKey),
+		errTasks:         workqueue.NewRateLimitingQueue(workqueue.DefaultControllerRateLimiter()),
+		deletedJobs:      workqueue.NewRateLimitingQueue(workqueue.DefaultControllerRateLimiter()),
 		kubeclient:       kubernetes.NewForConfigOrDie(config),
 		kbclient:         kbver.NewForConfigOrDie(config),
 		namespaceAsQueue: nsAsQueue,
@@ -324,10 +298,10 @@ func (sc *SchedulerCache) Run(stopCh <-chan struct{}) {
 	}
 
 	// Re-sync error tasks.
-	go sc.resync()
+	go wait.Until(sc.processResyncTask, 0, stopCh)
 
 	// Cleanup jobs.
-	go sc.cleanupJobs()
+	go wait.Until(sc.processCleanupJob, 0, stopCh)
 }
 
 func (sc *SchedulerCache) WaitForCacheSync(stopCh <-chan struct{}) bool {
@@ -388,7 +362,9 @@ func (sc *SchedulerCache) Evict(taskInfo *kbapi.TaskInfo, reason string) error {
 	}
 
 	// Add new task to node.
-	node.UpdateTask(task)
+	if err := node.UpdateTask(task); err != nil {
+		return err
+	}
 
 	p := task.Pod
 
@@ -430,7 +406,9 @@ func (sc *SchedulerCache) Bind(taskInfo *kbapi.TaskInfo, hostname string) error
 	task.NodeName = hostname
 
 	// Add task to the node.
-	node.AddTask(task)
+	if err := node.AddTask(task); err != nil {
+		return err
+	}
 
 	p := task.Pod
 
@@ -476,77 +454,52 @@ func (sc *SchedulerCache) taskUnschedulable(task *api.TaskInfo, message string)
 func (sc *SchedulerCache) deleteJob(job *kbapi.JobInfo) {
 	glog.V(3).Infof("Try to delete Job <%v:%v/%v>", job.UID, job.Namespace, job.Name)
 
-	time.AfterFunc(5*time.Second, func() {
-		sc.deletedJobs.AddIfNotPresent(job)
-	})
+	sc.deletedJobs.AddRateLimited(job)
 }
 
-func (sc *SchedulerCache) processCleanupJob() error {
-	_, err := sc.deletedJobs.Pop(func(obj interface{}) error {
-		job, ok := obj.(*kbapi.JobInfo)
-		if !ok {
-			return fmt.Errorf("failed to convert %v to *v1.Pod", obj)
-		}
-
-		func() {
-			sc.Mutex.Lock()
-			defer sc.Mutex.Unlock()
-
-			if kbapi.JobTerminated(job) {
-				delete(sc.Jobs, job.UID)
-				glog.V(3).Infof("Job <%v:%v/%v> was deleted.", job.UID, job.Namespace, job.Name)
-			} else {
-				// Retry
-				sc.deleteJob(job)
-			}
-		}()
+func (sc *SchedulerCache) processCleanupJob() {
+	obj, shutdown := sc.deletedJobs.Get()
+	if shutdown {
+		return
+	}
 
-		return nil
-	})
+	job, found := obj.(*kbapi.JobInfo)
+	if !found {
+		glog.Errorf("Failed to convert <%v> to *JobInfo", obj)
+		return
+	}
 
-	return err
-}
+	sc.Mutex.Lock()
+	defer sc.Mutex.Unlock()
 
-func (sc *SchedulerCache) cleanupJobs() {
-	for {
-		err := sc.processCleanupJob()
-		if err != nil {
-			glog.Errorf("Failed to process job clean up: %v", err)
-		}
+	if kbapi.JobTerminated(job) {
+		delete(sc.Jobs, job.UID)
+		glog.V(3).Infof("Job <%v:%v/%v> was deleted.", job.UID, job.Namespace, job.Name)
+	} else {
+		// Retry
+		sc.deleteJob(job)
 	}
 }
 
 func (sc *SchedulerCache) resyncTask(task *kbapi.TaskInfo) {
-	if err := sc.errTasks.AddIfNotPresent(task); err != nil {
-		glog.Errorf("Failed to re-sync tasks <%v/%v>: %v",
-			task.Namespace, task.Name, err)
-	}
+	sc.errTasks.AddRateLimited(task)
 }
 
-func (sc *SchedulerCache) resync() {
-	for {
-		err := sc.processResyncTask()
-		if err != nil {
-			glog.Errorf("Failed to process resync: %v", err)
-		}
+func (sc *SchedulerCache) processResyncTask() {
+	obj, shutdown := sc.errTasks.Get()
+	if shutdown {
+		return
+	}
+	task, ok := obj.(*kbapi.TaskInfo)
+	if !ok {
+		glog.Errorf("failed to convert %v to *v1.Pod", obj)
+		return
 	}
-}
-
-func (sc *SchedulerCache) processResyncTask() error {
-	_, err := sc.errTasks.Pop(func(obj interface{}) error {
-		task, ok := obj.(*kbapi.TaskInfo)
-		if !ok {
-			return fmt.Errorf("failed to convert %v to *v1.Pod", obj)
-		}
-
-		if err := sc.syncTask(task); err != nil {
-			glog.Errorf("Failed to sync pod <%v/%v>", task.Namespace, task.Name)
-			return err
-		}
-		return nil
-	})
 
-	return err
+	if err := sc.syncTask(task); err != nil {
+		glog.Errorf("Failed to sync pod <%v/%v>, retry it.", task.Namespace, task.Name)
+		sc.resyncTask(task)
+	}
 }
 
 func (sc *SchedulerCache) Snapshot() *kbapi.ClusterInfo {