d8dfun/virtual-kubelet
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 1 Wiki Activity

Remove sync provider support

Browse Source 
This removes the legacy sync provider interface. All new providers
are expected to implement the async NotifyPods interface.

The legacy sync provider interface creates complexities around
how the deletion flow works, and the mixed sync and async APIs
block us from evolving functionality.

This collapses in the NotifyPods interface into the PodLifecycleHandler
interface.
This commit is contained in:
Sargun Dhillon
2019-10-02 09:28:09 -07:00
parent b3aa0f577b
commit 4202b03cda
8 changed files with 41 additions and 441 deletions
Download Patch File Download Diff File Expand all files Collapse all files

14
node/podcontroller.go
View File

@@ -49,7 +49,9 @@ type PodLifecycleHandler interface {
// UpdatePod takes a Kubernetes Pod and updates it within the provider.
UpdatePod(ctx context.Context, pod *corev1.Pod) error
// DeletePod takes a Kubernetes Pod and deletes it from the provider.
// DeletePod takes a Kubernetes Pod and deletes it from the provider. Once a pod is deleted, the provider is
// expected to call the NotifyPods callback with a terminal pod status where all the containers are in a terminal
// state, as well as the pod.
DeletePod(ctx context.Context, pod *corev1.Pod) error
// GetPod retrieves a pod by name from the provider (can be cached).
@@ -69,12 +71,7 @@ type PodLifecycleHandler interface {
// concurrently outside of the calling goroutine. Therefore it is recommended
// to return a version after DeepCopy.
GetPods(context.Context) ([]*corev1.Pod, error)
}
// PodNotifier notifies callers of pod changes.
// Providers should implement this interface to enable callers to be notified
// of pod status updates asynchronously.
type PodNotifier interface {
// NotifyPods instructs the notifier to call the passed in function when
// the pod status changes. It should be called when a pod's status changes.
//
@@ -218,7 +215,10 @@ func (pc *PodController) Run(ctx context.Context, podSyncWorkers int) (retErr er
}()
podStatusQueue := workqueue.NewNamedRateLimitingQueue(workqueue.DefaultControllerRateLimiter(), "syncPodStatusFromProvider")
pc.runSyncFromProvider(ctx, podStatusQueue)
pc.provider.NotifyPods(ctx, func(pod *corev1.Pod) {
pc.enqueuePodStatusUpdate(ctx, podStatusQueue, pod.DeepCopy())
})
defer podStatusQueue.ShutDown()
// Wait for the caches to be synced *before* starting to do work.