tencent cloud

Tencent Kubernetes Engine

소식 및 공지 사항
릴리스 노트
제품 릴리스 기록
제품 소개
제품 장점
제품 아키텍처
시나리오
제품 기능
리전 및 가용존
빠른 시작
신규 사용자 가이드
표준 클러스터를 빠르게 생성
Demo
클라우드에서 컨테이너화된 애플리케이션 배포 Check List
TKE 표준 클러스터 가이드
Tencent Kubernetes Engine(TKE)
클러스터 관리
네트워크 관리
스토리지 관리
Worker 노드 소개
Kubernetes Object Management
워크로드
클라우드 네이티브 서비스 가이드
Tencent Managed Service for Prometheus
TKE Serverless 클러스터 가이드
TKE 클러스터 등록 가이드
실습 튜토리얼
Serverless 클러스터
네트워크
로그
모니터링
유지보수
DevOps
탄력적 스케일링
자주 묻는 질문
클러스터
TKE Serverless 클러스터
유지보수
서비스
이미지 레지스트리
원격 터미널
문서Tencent Kubernetes Engine

Host Operation System Release for Super Node Pods (Mitigated NodeLost Issue)

포커스 모드
폰트 크기
마지막 업데이트 시간: 2025-02-11 15:42:18
To enrich the features, Tencent Kubernetes Engine (TKE) plans to release a new host OS version for super nodes on March 10, 2025 (UTC+8). The incremental pods on super nodes will gradually use the new OS version.

Time

The operating system will gradually apply to the incremental pods starting from March 10, 2025 (UTC+8).

Impact Scope

This change will apply to the pods that are rebuilt or newly added on super nodes in TKE Standard Cluster and TKE Serverless Cluster. The pods currently running on super nodes will not be affected.

About the New Version of Host OS

The new version of the host OS is based on Tencent's commercial TencentOS distribution, with an upgraded kernel on the original OS, supporting a wider range of storage types. It greatly optimizes memory usage of system components, reducing their memory occupation.

Version Check Methods

After logging into the pod, execute the following command to view the host OS kernel version(5.4.241-1.0017):
# uname -a
Linux <pod-name> 5.4.241-1.0017.10.eks.8.tl4

New Features

Mitigated unresponsiveness of the machine that is caused by memory inadequacy. It will trigger cgroup OOM kill timely to prevent the machine from entering the unresponsive state (nodelost state) when the memory in the pod is nearly reached.
Added the support for Cloud Object Storage (COS). For the operation guide, see Using COS.
Added the support for Ceph File System (CephFS).

Optimized Features

The kernel version was upgraded from 5.4.119-1.0009 to 5.4.241-1.0017.
containerd was upgraded from v1.4.3 to v1.6.9.
The GPU driver was upgraded from 470 to 535, supporting CUDA version upgrade from up to 11.4 to 12.2.
The systemd software package version was updated from 239 to 255.
The dbus software package version was updated from 1.12.8 to 1.14.8.

Other Notes

To try the operating system in advance, you can submit a ticket for application.
To upgrade a running pod to the new OS version, you need to rebuild the pod first.
If you have specified a specific OS version through the annotation (eks.tke.cloud.tencent.com/eklet-version), you need to remove that annotation to make the new OS version take effect.


도움말 및 지원

문제 해결에 도움이 되었나요?

피드백