Cpu manager reconclie loop can restore state

pkg/kubelet/cm/cpumanager/cpu_manager.go

@@ -160,8 +160,8 @@ func NewManager(
 }
 
 func (m *manager) Start(activePods ActivePodsFunc, podStatusProvider status.PodStatusProvider, containerRuntime runtimeService) {
-	glog.Infof("[cpumanger] starting with %s policy", m.policy.Name())
-	glog.Infof("[cpumanger] reconciling every %v", m.reconcilePeriod)
+	glog.Infof("[cpumanager] starting with %s policy", m.policy.Name())
+	glog.Infof("[cpumanager] reconciling every %v", m.reconcilePeriod)
 
 	m.activePods = activePods
 	m.podStatusProvider = podStatusProvider
@@ -242,6 +242,25 @@ func (m *manager) reconcileState() (success []reconciledContainer, failure []rec
 				continue
 			}
 
+			// Check whether container is present in state, there may be 3 reasons why it's not present:
+			// - policy does not want to track the container
+			// - kubelet has just been restarted - and there is no previous state file
+			// - container has been removed from state by RemoveContainer call (DeletionTimestamp is set)
+			if _, ok := m.state.GetCPUSet(containerID); !ok {
+				if status.Phase == v1.PodRunning && pod.DeletionTimestamp == nil {
+					glog.V(4).Infof("[cpumanager] reconcileState: container is not present in state - trying to add (pod: %s, container: %s, container id: %s)", pod.Name, container.Name, containerID)
+					err := m.AddContainer(pod, &container, containerID)
+					if err != nil {
+						glog.Errorf("[cpumanager] reconcileState: failed to add container (pod: %s, container: %s, container id: %s, error: %v)", pod.Name, container.Name, containerID, err)
+						failure = append(failure, reconciledContainer{pod.Name, container.Name, containerID})
+					}
+				} else {
+					// if DeletionTimestamp is set, pod has already been removed from state
+					// skip the pod/container since it's not running and will be deleted soon
+					continue
+				}
+			}
+
 			cset := m.state.GetCPUSetOrDefault(containerID)
 			if cset.IsEmpty() {
 				// NOTE: This should not happen outside of tests.

pkg/kubelet/cm/cpumanager/policy.go

@@ -25,6 +25,8 @@ import (
 type Policy interface {
 	Name() string
 	Start(s state.State)
+	// AddContainer call is idempotent
 	AddContainer(s state.State, pod *v1.Pod, container *v1.Container, containerID string) error
+	// RemoveContainer call is idempotent
 	RemoveContainer(s state.State, containerID string) error
 }

pkg/kubelet/cm/cpumanager/policy_static.go

@@ -156,9 +156,15 @@ func (p *staticPolicy) assignableCPUs(s state.State) cpuset.CPUSet {
 }
 
 func (p *staticPolicy) AddContainer(s state.State, pod *v1.Pod, container *v1.Container, containerID string) error {
-	glog.Infof("[cpumanager] static policy: AddContainer (pod: %s, container: %s, container id: %s)", pod.Name, container.Name, containerID)
 	if numCPUs := guaranteedCPUs(pod, container); numCPUs != 0 {
+		glog.Infof("[cpumanager] static policy: AddContainer (pod: %s, container: %s, container id: %s)", pod.Name, container.Name, containerID)
 		// container belongs in an exclusively allocated pool
+
+		if _, ok := s.GetCPUSet(containerID); ok {
+			glog.Infof("[cpumanager] static policy: container already present in state, skipping (container: %s, container id: %s)", container.Name, containerID)
+			return nil
+		}
+
 		cpuset, err := p.allocateCPUs(s, numCPUs)
 		if err != nil {
 			glog.Errorf("[cpumanager] unable to allocate %d CPUs (container id: %s, error: %v)", numCPUs, containerID, err)