Clayton Coleman 8722c834e5 kubelet: Never restart containers in deleting pods ...

When constructing the API status of a pod, if the pod is marked for
deletion no containers should be started. Previously, if a container
inside of a terminating pod failed to start due to a container
runtime error (that populates reasonCache) the reasonCache would
remain populated (it is only updated by syncPod for non-terminating
pods) and the delete action on the pod would be delayed until the
reasonCache entry expired due to other pods.

This dramatically reduces the amount of time the Kubelet waits to
delete pods that are terminating and encountered a container runtime
error.

2020-03-04 13:34:25 -05:00

..

testing

…

BUILD

…

cache_test.go

…

cache.go

…

container_gc.go

…

container_hash_test.go

…

container_reference_manager.go

…

helpers_test.go

…

helpers.go

…

os.go

…

ref_test.go

…

ref.go

…

resize.go

…

runtime_cache_fake.go

…

runtime_cache_test.go

…

runtime_cache.go

…

runtime.go

…

sync_result_test.go

…

sync_result.go

…