Merge pull request #1561 from brendandburns/sleep

Add some backoff to the scheduler to prevent tight-looping.
This commit is contained in:
bgrant0607 2014-10-03 15:29:12 -07:00
commit fea58c9b88
2 changed files with 134 additions and 4 deletions

View File

@ -20,6 +20,7 @@ package factory
import (
"math/rand"
"sync"
"time"
"github.com/GoogleCloudPlatform/kubernetes/pkg/api"
@ -79,6 +80,11 @@ func (factory *ConfigFactory) Create() *scheduler.Config {
algorithm.EqualPriority,
&storeToPodLister{podCache}, r)
podBackoff := podBackoff{
perPodBackoff: map[string]*backoffEntry{},
clock: realClock{},
}
return &scheduler.Config{
MinionLister: &storeToMinionLister{minionCache},
Algorithm: algo,
@ -91,7 +97,7 @@ func (factory *ConfigFactory) Create() *scheduler.Config {
pod.ID, minionCache.Contains(), podCache.Contains())
return pod
},
Error: factory.makeDefaultErrorFunc(podQueue),
Error: factory.makeDefaultErrorFunc(&podBackoff, podQueue),
}
}
@ -167,15 +173,16 @@ func (factory *ConfigFactory) pollMinions() (cache.Enumerator, error) {
return &minionEnumerator{list}, nil
}
func (factory *ConfigFactory) makeDefaultErrorFunc(podQueue *cache.FIFO) func(pod *api.Pod, err error) {
func (factory *ConfigFactory) makeDefaultErrorFunc(backoff *podBackoff, podQueue *cache.FIFO) func(pod *api.Pod, err error) {
return func(pod *api.Pod, err error) {
glog.Errorf("Error scheduling %v: %v; retrying", pod.ID, err)
backoff.gc()
// Retry asynchronously.
// Note that this is extremely rudimentary and we need a more real error handling path.
go func() {
defer util.HandleCrash()
podID := pod.ID
backoff.wait(podID)
// Get the pod again; it may have changed/been scheduled already.
pod = &api.Pod{}
err := factory.Client.Get().Path("pods").Path(podID).Do().Into(pod)
@ -244,3 +251,62 @@ func (b *binder) Bind(binding *api.Binding) error {
glog.V(2).Infof("Attempting to bind %v to %v", binding.PodID, binding.Host)
return b.Post().Path("bindings").Body(binding).Do().Error()
}
type clock interface {
Now() time.Time
}
type realClock struct{}
func (realClock) Now() time.Time {
return time.Now()
}
type backoffEntry struct {
backoff time.Duration
lastUpdate time.Time
}
type podBackoff struct {
perPodBackoff map[string]*backoffEntry
lock sync.Mutex
clock clock
}
func (p *podBackoff) getEntry(podID string) *backoffEntry {
p.lock.Lock()
defer p.lock.Unlock()
entry, ok := p.perPodBackoff[podID]
if !ok {
entry = &backoffEntry{backoff: 1 * time.Second}
p.perPodBackoff[podID] = entry
}
entry.lastUpdate = p.clock.Now()
return entry
}
func (p *podBackoff) getBackoff(podID string) time.Duration {
entry := p.getEntry(podID)
duration := entry.backoff
entry.backoff *= 2
if entry.backoff > 60*time.Second {
entry.backoff = 60 * time.Second
}
glog.V(4).Infof("Backing off %s for pod %s", duration.String(), podID)
return duration
}
func (p *podBackoff) wait(podID string) {
time.Sleep(p.getBackoff(podID))
}
func (p *podBackoff) gc() {
p.lock.Lock()
defer p.lock.Unlock()
now := p.clock.Now()
for podID, entry := range p.perPodBackoff {
if now.Sub(entry.lastUpdate) > 60*time.Second {
delete(p.perPodBackoff, podID)
}
}
}

View File

@ -187,7 +187,11 @@ func TestDefaultErrorFunc(t *testing.T) {
server := httptest.NewServer(mux)
factory := ConfigFactory{client.NewOrDie(&client.Config{Host: server.URL, Version: testapi.Version()})}
queue := cache.NewFIFO()
errFunc := factory.makeDefaultErrorFunc(queue)
podBackoff := podBackoff{
perPodBackoff: map[string]*backoffEntry{},
clock: &fakeClock{},
}
errFunc := factory.makeDefaultErrorFunc(&podBackoff, queue)
errFunc(testPod, nil)
for {
@ -276,6 +280,14 @@ func TestMinionEnumerator(t *testing.T) {
}
}
type fakeClock struct {
t time.Time
}
func (f *fakeClock) Now() time.Time {
return f.t
}
func TestBind(t *testing.T) {
table := []struct {
binding *api.Binding
@ -301,3 +313,55 @@ func TestBind(t *testing.T) {
handler.ValidateRequest(t, "/api/"+testapi.Version()+"/bindings", "POST", &expectedBody)
}
}
func TestBackoff(t *testing.T) {
clock := fakeClock{}
backoff := podBackoff{
perPodBackoff: map[string]*backoffEntry{},
clock: &clock,
}
tests := []struct {
podID string
expectedDuration time.Duration
advanceClock time.Duration
}{
{
podID: "foo",
expectedDuration: 1 * time.Second,
},
{
podID: "foo",
expectedDuration: 2 * time.Second,
},
{
podID: "foo",
expectedDuration: 4 * time.Second,
},
{
podID: "bar",
expectedDuration: 1 * time.Second,
advanceClock: 120 * time.Second,
},
// 'foo' should have been gc'd here.
{
podID: "foo",
expectedDuration: 1 * time.Second,
},
}
for _, test := range tests {
duration := backoff.getBackoff(test.podID)
if duration != test.expectedDuration {
t.Errorf("expected: %s, got %s for %s", test.expectedDuration.String(), duration.String(), test.podID)
}
clock.t = clock.t.Add(test.advanceClock)
backoff.gc()
}
backoff.perPodBackoff["foo"].backoff = 60 * time.Second
duration := backoff.getBackoff("foo")
if duration != 60*time.Second {
t.Errorf("expected: 60, got %s", duration.String())
}
}