[Kubernetes] Eviction Manager工作原理分析

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11

thresholds, err := eviction.ParseThresholdConfig(enforceNodeAllocatable, kubeCfg.EvictionHard, kubeCfg.EvictionSoft, kubeCfg.EvictionSoftGracePeriod, kubeCfg.EvictionMinimumReclaim)
if err != nil {
  return nil, err
}
evictionConfig := eviction.Config{
  PressureTransitionPeriod: kubeCfg.EvictionPressureTransitionPeriod.Duration,
  MaxPodGracePeriodSeconds: int64(kubeCfg.EvictionMaxPodGracePeriod),
  Thresholds:               thresholds,
  KernelMemcgNotification:  experimentalKernelMemcgNotification,
  PodCgroupRoot:            kubeDeps.ContainerManager.GetPodCgroupRoot(),
}

1
2
3
4
5

evictionManager, evictionAdmitHandler := eviction.NewManager(klet.resourceAnalyzer, evictionConfig, killPodNow(klet.podWorkers, kubeDeps.Recorder),  
 klet.podManager.GetMirrorPodByPod, klet.imageManager, klet.containerGC, kubeDeps.Recorder, nodeRef, klet.clock, etcHostsPathFunc)

	klet.evictionManager = evictionManager
	klet.admitHandlers.AddPodAdmitHandler(evictionAdmitHandler)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33

// NewResourceAnalyzer returns a new ResourceAnalyzer
func NewResourceAnalyzer(statsProvider Provider, calVolumeFrequency time.Duration) ResourceAnalyzer {
	fsAnalyzer := newFsResourceAnalyzer(statsProvider, calVolumeFrequency)
	summaryProvider := NewSummaryProvider(statsProvider)
	return &resourceAnalyzer{fsAnalyzer, summaryProvider}
}
//其中fsAnalyzer 初始化实现为
// newFsResourceAnalyzer returns a new fsResourceAnalyzer implementation
func newFsResourceAnalyzer(statsProvider Provider, calcVolumePeriod time.Duration) *fsResourceAnalyzer {
	r := &fsResourceAnalyzer{
		statsProvider: statsProvider,
		calcPeriod:    calcVolumePeriod,
	}
	r.cachedVolumeStats.Store(make(statCache))
	return r
}
//summaryProvider实现
// NewSummaryProvider returns a SummaryProvider using the stats provided by the
// specified statsProvider.
func NewSummaryProvider(statsProvider Provider) SummaryProvider {
	kubeletCreationTime := metav1.Now()
	bootTime, err := util.GetBootTime()
	if err != nil {
		// bootTime will be zero if we encounter an error getting the boot time.
		klog.Warningf("Error getting system boot time.  Node metrics will have an incorrect start time: %v", err)
	}

	return &summaryProviderImpl{
		kubeletCreationTime: kubeletCreationTime,
		systemBootTime:      metav1.NewTime(bootTime),
		provider:            statsProvider,
	}
}

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39

// Admit rejects a pod if its not safe to admit for node stability.
func (m *managerImpl) Admit(attrs *lifecycle.PodAdmitAttributes) lifecycle.PodAdmitResult {
	m.RLock()
	defer m.RUnlock()
	if len(m.nodeConditions) == 0 {
		return lifecycle.PodAdmitResult{Admit: true}
	}
	// Admit Critical pods even under resource pressure since they are required for system stability.
	// https://github.com/kubernetes/kubernetes/issues/40573 has more details.
	if kubelettypes.IsCriticalPod(attrs.Pod) {
		return lifecycle.PodAdmitResult{Admit: true}
	}

	// Conditions other than memory pressure reject all pods
	nodeOnlyHasMemoryPressureCondition := hasNodeCondition(m.nodeConditions, v1.NodeMemoryPressure) && len(m.nodeConditions) == 1
	if nodeOnlyHasMemoryPressureCondition {
		notBestEffort := v1.PodQOSBestEffort != v1qos.GetPodQOS(attrs.Pod)
		if notBestEffort {
			return lifecycle.PodAdmitResult{Admit: true}
		}

		// When node has memory pressure, check BestEffort Pod's toleration:
		// admit it if tolerates memory pressure taint, fail for other tolerations, e.g. DiskPressure.
		if v1helper.TolerationsTolerateTaint(attrs.Pod.Spec.Tolerations, &v1.Taint{
			Key:    v1.TaintNodeMemoryPressure,
			Effect: v1.TaintEffectNoSchedule,
		}) {
			return lifecycle.PodAdmitResult{Admit: true}
		}
	}

	// reject pods when under memory pressure (if pod is best effort), or if under disk pressure.
	klog.Warningf("Failed to admit pod %s - node has conditions: %v", format.Pod(attrs.Pod), m.nodeConditions)
	return lifecycle.PodAdmitResult{
		Admit:   false,
		Reason:  Reason,
		Message: fmt.Sprintf(nodeConditionMessageFmt, m.nodeConditions),
	}
}

1

kl.evictionManager.Start(kl.StatsProvider, kl.GetActivePods, kl.podResourcesAreReclaimed, evictionMonitoringPeriod)

1
2
3
4
5
6
7

klet.StatsProvider = stats.NewCadvisorStatsProvider(
			klet.cadvisor,
			klet.resourceAnalyzer,
			klet.podManager,
			klet.runtimeCache,
			klet.containerRuntime,
			klet.statusManager)

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31

// Start starts the control loop to observe and response to low compute resources.
func (m *managerImpl) Start(diskInfoProvider DiskInfoProvider, podFunc ActivePodsFunc, podCleanedUpFunc PodCleanedUpFunc, monitoringInterval time.Duration) {
	thresholdHandler := func(message string) {
		klog.Infof(message)
		m.synchronize(diskInfoProvider, podFunc)
	}
	if m.config.KernelMemcgNotification {
		for _, threshold := range m.config.Thresholds {
			if threshold.Signal == evictionapi.SignalMemoryAvailable || threshold.Signal == evictionapi.SignalAllocatableMemoryAvailable {
				notifier, err := NewMemoryThresholdNotifier(threshold, m.config.PodCgroupRoot, &CgroupNotifierFactory{}, thresholdHandler)
				if err != nil {
					klog.Warningf("eviction manager: failed to create memory threshold notifier: %v", err)
				} else {
					go notifier.Start()
					m.thresholdNotifiers = append(m.thresholdNotifiers, notifier)
				}
			}
		}
	}
	// start the eviction manager monitoring
	go func() {
		for {
			if evictedPods := m.synchronize(diskInfoProvider, podFunc); evictedPods != nil {
				klog.Infof("eviction manager: pods %s evicted, waiting for pod to be cleaned up", format.Pods(evictedPods))
				m.waitForPodsCleanup(podCleanedUpFunc, evictedPods)
			} else {
				time.Sleep(monitoringInterval)
			}
		}
	}()
}

  1
  2
  3
  4
  5
  6
  7
  8
  9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24
 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165

// synchronize is the main control loop that enforces eviction thresholds.
// Returns the pod that was killed, or nil if no pod was killed.
func (m *managerImpl) synchronize(diskInfoProvider DiskInfoProvider, podFunc ActivePodsFunc) []*v1.Pod {
	// if we have nothing to do, just return
	thresholds := m.config.Thresholds
	if len(thresholds) == 0 && !utilfeature.DefaultFeatureGate.Enabled(features.LocalStorageCapacityIsolation) {
		return nil
	}

	klog.V(3).Infof("eviction manager: synchronize housekeeping")
	// build the ranking functions (if not yet known)
	// TODO: have a function in cadvisor that lets us know if global housekeeping has completed
	if m.dedicatedImageFs == nil {
		hasImageFs, ok := diskInfoProvider.HasDedicatedImageFs()
		if ok != nil {
			return nil
		}
		m.dedicatedImageFs = &hasImageFs
		m.signalToRankFunc = buildSignalToRankFunc(hasImageFs)
		m.signalToNodeReclaimFuncs = buildSignalToNodeReclaimFuncs(m.imageGC, m.containerGC, hasImageFs)
	}

	activePods := podFunc()
	updateStats := true
	summary, err := m.summaryProvider.Get(updateStats)
	if err != nil {
		klog.Errorf("eviction manager: failed to get summary stats: %v", err)
		return nil
	}

	if m.clock.Since(m.thresholdsLastUpdated) > notifierRefreshInterval {
		m.thresholdsLastUpdated = m.clock.Now()
		for _, notifier := range m.thresholdNotifiers {
			if err := notifier.UpdateThreshold(summary); err != nil {
				klog.Warningf("eviction manager: failed to update %s: %v", notifier.Description(), err)
			}
		}
	}

	// make observations and get a function to derive pod usage stats relative to those observations.
	observations, statsFunc := makeSignalObservations(summary)
	debugLogObservations("observations", observations)

	// determine the set of thresholds met independent of grace period
	thresholds = thresholdsMet(thresholds, observations, false)
	debugLogThresholdsWithObservation("thresholds - ignoring grace period", thresholds, observations)

	// determine the set of thresholds previously met that have not yet satisfied the associated min-reclaim
	if len(m.thresholdsMet) > 0 {
		thresholdsNotYetResolved := thresholdsMet(m.thresholdsMet, observations, true)
		thresholds = mergeThresholds(thresholds, thresholdsNotYetResolved)
	}
	debugLogThresholdsWithObservation("thresholds - reclaim not satisfied", thresholds, observations)

	// track when a threshold was first observed
	now := m.clock.Now()
	thresholdsFirstObservedAt := thresholdsFirstObservedAt(thresholds, m.thresholdsFirstObservedAt, now)

	// the set of node conditions that are triggered by currently observed thresholds
	nodeConditions := nodeConditions(thresholds)
	if len(nodeConditions) > 0 {
		klog.V(3).Infof("eviction manager: node conditions - observed: %v", nodeConditions)
	}

	// track when a node condition was last observed
	nodeConditionsLastObservedAt := nodeConditionsLastObservedAt(nodeConditions, m.nodeConditionsLastObservedAt, now)

	// node conditions report true if it has been observed within the transition period window
	nodeConditions = nodeConditionsObservedSince(nodeConditionsLastObservedAt, m.config.PressureTransitionPeriod, now)
	if len(nodeConditions) > 0 {
		klog.V(3).Infof("eviction manager: node conditions - transition period not met: %v", nodeConditions)
	}

	// determine the set of thresholds we need to drive eviction behavior (i.e. all grace periods are met)
	thresholds = thresholdsMetGracePeriod(thresholdsFirstObservedAt, now)
	debugLogThresholdsWithObservation("thresholds - grace periods satisfied", thresholds, observations)

	// update internal state
	m.Lock()
	m.nodeConditions = nodeConditions
	m.thresholdsFirstObservedAt = thresholdsFirstObservedAt
	m.nodeConditionsLastObservedAt = nodeConditionsLastObservedAt
	m.thresholdsMet = thresholds

	// determine the set of thresholds whose stats have been updated since the last sync
	thresholds = thresholdsUpdatedStats(thresholds, observations, m.lastObservations)
	debugLogThresholdsWithObservation("thresholds - updated stats", thresholds, observations)

	m.lastObservations = observations
	m.Unlock()

	// evict pods if there is a resource usage violation from local volume temporary storage
	// If eviction happens in localStorageEviction function, skip the rest of eviction action
	if utilfeature.DefaultFeatureGate.Enabled(features.LocalStorageCapacityIsolation) {
		if evictedPods := m.localStorageEviction(summary, activePods); len(evictedPods) > 0 {
			return evictedPods
		}
	}

	if len(thresholds) == 0 {
		klog.V(3).Infof("eviction manager: no resources are starved")
		return nil
	}

	// rank the thresholds by eviction priority
	sort.Sort(byEvictionPriority(thresholds))
	thresholdToReclaim, resourceToReclaim, foundAny := getReclaimableThreshold(thresholds)
	if !foundAny {
		return nil
	}
	klog.Warningf("eviction manager: attempting to reclaim %v", resourceToReclaim)

	// record an event about the resources we are now attempting to reclaim via eviction
	m.recorder.Eventf(m.nodeRef, v1.EventTypeWarning, "EvictionThresholdMet", "Attempting to reclaim %s", resourceToReclaim)

	// check if there are node-level resources we can reclaim to reduce pressure before evicting end-user pods.
	if m.reclaimNodeLevelResources(thresholdToReclaim.Signal, resourceToReclaim) {
		klog.Infof("eviction manager: able to reduce %v pressure without evicting pods.", resourceToReclaim)
		return nil
	}

	klog.Infof("eviction manager: must evict pod(s) to reclaim %v", resourceToReclaim)

	// rank the pods for eviction
	rank, ok := m.signalToRankFunc[thresholdToReclaim.Signal]
	if !ok {
		klog.Errorf("eviction manager: no ranking function for signal %s", thresholdToReclaim.Signal)
		return nil
	}

	// the only candidates viable for eviction are those pods that had anything running.
	if len(activePods) == 0 {
		klog.Errorf("eviction manager: eviction thresholds have been met, but no pods are active to evict")
		return nil
	}

	// rank the running pods for eviction for the specified resource
	rank(activePods, statsFunc)

	klog.Infof("eviction manager: pods ranked for eviction: %s", format.Pods(activePods))

	//record age of metrics for met thresholds that we are using for evictions.
	for _, t := range thresholds {
		timeObserved := observations[t.Signal].time
		if !timeObserved.IsZero() {
			metrics.EvictionStatsAge.WithLabelValues(string(t.Signal)).Observe(metrics.SinceInSeconds(timeObserved.Time))
		}
	}

	// we kill at most a single pod during each eviction interval
	for i := range activePods {
		pod := activePods[i]
		gracePeriodOverride := int64(0)
		if !isHardEvictionThreshold(thresholdToReclaim) {
			gracePeriodOverride = m.config.MaxPodGracePeriodSeconds
		}
		message, annotations := evictionMessage(resourceToReclaim, pod, statsFunc)
		if m.evictPod(pod, gracePeriodOverride, message, annotations) {
			metrics.Evictions.WithLabelValues(string(thresholdToReclaim.Signal)).Inc()
			return []*v1.Pod{pod}
		}
	}
	klog.Infof("eviction manager: unable to evict any pods from the node")
	return nil
}

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50

// killPodNow returns a KillPodFunc that can be used to kill a pod.
// It is intended to be injected into other modules that need to kill a pod.
func killPodNow(podWorkers PodWorkers, recorder record.EventRecorder) eviction.KillPodFunc {
	return func(pod *v1.Pod, status v1.PodStatus, gracePeriodOverride *int64) error {
		// determine the grace period to use when killing the pod
		gracePeriod := int64(0)
		if gracePeriodOverride != nil {
			gracePeriod = *gracePeriodOverride
		} else if pod.Spec.TerminationGracePeriodSeconds != nil {
			gracePeriod = *pod.Spec.TerminationGracePeriodSeconds
		}

		// we timeout and return an error if we don't get a callback within a reasonable time.
		// the default timeout is relative to the grace period (we settle on 10s to wait for kubelet->runtime traffic to complete in sigkill)
		timeout := int64(gracePeriod + (gracePeriod / 2))
		minTimeout := int64(10)
		if timeout < minTimeout {
			timeout = minTimeout
		}
		timeoutDuration := time.Duration(timeout) * time.Second

		// open a channel we block against until we get a result
		type response struct {
			err error
		}
		ch := make(chan response, 1)
		podWorkers.UpdatePod(&UpdatePodOptions{
			Pod:        pod,
			UpdateType: kubetypes.SyncPodKill,
			OnCompleteFunc: func(err error) {
				ch <- response{err: err}
			},
			KillPodOptions: &KillPodOptions{
				PodStatusFunc: func(p *v1.Pod, podStatus *kubecontainer.PodStatus) v1.PodStatus {
					return status
				},
				PodTerminationGracePeriodSecondsOverride: gracePeriodOverride,
			},
		})

		// wait for either a response, or a timeout
		select {
		case r := <-ch:
			return r.err
		case <-time.After(timeoutDuration):
			recorder.Eventf(pod, v1.EventTypeWarning, events.ExceededGracePeriod, "Container runtime did not kill the pod within specified grace period.")
			return fmt.Errorf("timeout waiting to kill pod")
		}
	}
}