checkpointmanager

README

DISCLAIMER

• Sig-Node community has reached a general consensus, as a best practice, to avoid introducing any new checkpointing support. We reached this understanding after struggling with some hard-to-debug issues in the production environments caused by the checkpointing.
• Any changes to the checkpointed data structure would be considered incompatible and a component should add its own handling if it needs to ensure backward compatibility of reading old-format checkpoint files.

Introduction

This folder contains a framework & primitives, Checkpointing Manager, which is used by several other Kubelet submodules, devicemanager, pods, cpumanager, memorymanager, dra and allocation, to implement checkpointing at each submodule level. As already explained in above Disclaimer section, think twice before introducing any further checkpointing in Kubelet. If still checkpointing is required, then this folder provides the common APIs and the framework for implementing checkpointing. Using same APIs across all the submodules will help maintaining consistency at Kubelet level.

Below is the history of checkpointing support in Kubelet.

Package	First checkpointing support merged on	PR link	Status
kubelet/dockershim	Feb 3, 2017	[CRI] Implement Dockershim Checkpoint	Removed
devicemanager	Sep 6, 2017	Deviceplugin checkpoint	In use
kubelet/pod	Nov 22, 2017	Initial basic bootstrap-checkpoint support	In use
cpumanager	Oct 27, 2017	Add file backed state to cpu manager	In use
memorymanager	Feb 9, 2021	Memory manager	In use
dra	Mar 12, 2023	kubelet dra: Add checkpointing mechanism in the DRA Manager	In use
dra	Jul 24, 2024	DRA:refactor checkpointing	In use
allocation	Feb 28, 2025	Move pod resource allocation management out of the status manager	In use

Note: Dockershim has been removed from the Kubernetes project as of release 1.24, as a consequence dockershim checkpoint support has also been removed.

Documentation

Index

• type Checkpoint
• type CheckpointManager
  • func NewCheckpointManager(checkpointDir string) (CheckpointManager, error)

Types

type Checkpoint

type Checkpoint interface {
	MarshalCheckpoint() ([]byte, error)
	UnmarshalCheckpoint(blob []byte) error
	VerifyChecksum() error
}

Checkpoint provides the process checkpoint data

type CheckpointManager

type CheckpointManager interface {
	// CreateCheckpoint persists checkpoint in CheckpointStore. checkpointKey is the key for utilstore to locate checkpoint.
	// For file backed utilstore, checkpointKey is the file name to write the checkpoint data.
	CreateCheckpoint(checkpointKey string, checkpoint Checkpoint) error
	// GetCheckpoint retrieves checkpoint from CheckpointStore.
	GetCheckpoint(checkpointKey string, checkpoint Checkpoint) error
	// WARNING: RemoveCheckpoint will not return error if checkpoint does not exist.
	RemoveCheckpoint(checkpointKey string) error
	// ListCheckpoint returns the list of existing checkpoints.
	ListCheckpoints() ([]string, error)
}

CheckpointManager provides the interface to manage checkpoint

func NewCheckpointManager

func NewCheckpointManager(checkpointDir string) (CheckpointManager, error)

NewCheckpointManager returns a new instance of a checkpoint manager