Method for restoring compute nodes in a kubernetes cluster and related apparatus

By backing up the mount information of compute nodes in the Kubernetes cluster and using native extended resources to monitor the management flag, the problem of difficulty in remanaging compute nodes and restoring data disks after reinstallation is solved, achieving high-reliability recovery of compute nodes.

CN122240235APending Publication Date: 2026-06-19ZHEJIANG DAHUA TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
ZHEJIANG DAHUA TECH CO LTD
Filing Date
2026-02-03
Publication Date
2026-06-19

AI Technical Summary

Technical Problem

In Kubernetes clusters, it is difficult to accurately reinstate compute nodes to the cluster after an operating system is reinstalled following an anomaly, and it is also difficult to recover the data disks on the compute nodes, resulting in low reliability.

Method used

Obtain the mounting information of the data disk matching the compute node and back it up to the distributed key-value store database. Use the management flag in the native extended resources of Kubernetes to monitor whether remanagement is required after reinstalling the operating system. Obtain the mounting information through the distributed key-value store database to restore the data disk and remanage the compute node.

Benefits of technology

It improves the reliability of compute nodes in Kubernetes clusters, ensuring accurate recovery of data disks and stable operation of compute nodes without the need for third-party tools.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122240235A_ABST
    Figure CN122240235A_ABST
Patent Text Reader

Abstract

This application discloses a method and related apparatus for restoring compute nodes in a Kubernetes cluster. The method includes: obtaining mount information matching the data disk on a compute node managed to the cluster, and backing up the mount information to a distributed key-value store database; wherein the compute node managed to the cluster is configured with extended resources, the extended resources include a management identifier for indicating whether to perform management, and the extended resources are stored in the distributed key-value store database; in response to the compute node reinstalling its operating system and detecting the management identifier in the extended resources indicating that management should be performed, obtaining the mount information of the compute node from the distributed key-value store database, restoring the data disk on the compute node using the mount information, and re-managing the compute node to the cluster. Through the above method, this application can improve the reliability of compute nodes in a Kubernetes cluster.
Need to check novelty before this filing date? Find Prior Art