b9d00841a6
Recording the expected and actual checksum in the error makes it possible to provide that information, for example in a failed test like the ones for DRA. Otherwise developers have to manually step through the test with a debugger to figure out what the new checksum is.
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, dockershim, devicemanager, pods
and cpumanager, 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 |
|---|---|---|
| kubelet/dockershim | Feb 3, 2017 | [CRI] Implement Dockershim Checkpoint |
| devicemanager | Sep 6, 2017 | Deviceplugin checkpoint |
| kubelet/pod | Nov 22, 2017 | Initial basic bootstrap-checkpoint support |
| cpumanager | Oct 27, 2017 | Add file backed state to cpu manager |