OS images supported by ACK

This topic describes the OS images supported by Container Service for Kubernetes (ACK). This topic also describes the filter rules and limits on OS images.

When you create an ACK cluster or a node pool, you must select an OS image for nodes. By default, ACK provides only the latest versions of different OS images. The following table describes the OS images supported by ACK and the Kubernetes versions supported by each image.

Important

The following image list is maintained by ACK and may be different from the latest versions of images supported by Elastic Compute Service (ECS) due to update delays. We recommend that you use the OS images in the following table to deploy nodes.
In ACK clusters, the default operating system is cgroup v2 for ContainerOS 3.3 or later, RHEL 9 or later, and Ubuntu 22 or later. For more information about cgroup v2, see cgroup version description.
Alibaba Cloud Linux 2 was discontinued (end of life) at 00:00:00 (UTC+8) on March 31, 2024. CentOS was discontinued (end of life) at 00:00:00 (UTC+8) on June 30, 2024, at 00:00:00. Alibaba Cloud stops providing technical support for ECS instances that run these OS versions. To quickly obtain technical support and ensure business continuity and stability, we commend that you use Alibaba Cloud Linux 3 or ContainerOS. For more information about the changes and solutions, see [Product Changes] EOL of Alibaba Cloud Linux 2 and CentOS 7.

Image Type	Image name (including the image version)	Image ID	OS	Architecture	Startup mode	OS size	Supported Kubernetes version
ContainerOS	ContainerOS 3	lifsea_3_x64_10G_containerd_1_6_28_alibase_20240705.vhd	Aliyun	x86_64	BIOS	10 GiB	1.24 and later
Alibaba Cloud Linux 3	Alibaba Cloud Linux 3.2104 LTS 64 bit	aliyun_3_x64_20G_alibase_20240528.vhd	Aliyun	x86_64	BIOS	20 GiB	1.18 and later
Alibaba Cloud Linux 3 for ARM.	Alibaba Cloud Linux 3.2104 LTS 64 bit ARM Edition	aliyun_3_arm64_20G_alibase_20240528.vhd	Aliyun	arm64	UEFI	20 GiB	1.20 and later
Alibaba Cloud Linux UEFI 3	Alibaba Cloud Linux UEFI 3.2104 Security Enhanced	aliyun_3_x64_20G_uefi_alibase_20230727.vhd	Aliyun	x86_64	UEFI	20 GiB	1.18 and later
Red Hat	Red Hat Enterprise Linux (RHEL) 9.3 64-bit	Red Hat Enterprise Linux 9.3 64bit	Red Hat	x86_64	BIOS	20 GiB	1.20 and later
Ubuntu	Ubuntu 22.04	ubuntu_22_04_x64_20G_alibase_20240508.vhd	Ubuntu	x86_64	BIOS	20 GiB	1.30 and later
Windows	Windows Server 2022 (20240220)	win2022_21H2_x64_dtc_en-us_40G_container_alibase_20240220.vhd	WindowsServer2022	x86_64	BIOS	40 GiB	1.18 and later
Windows	Windows Server 2019 (20240220)	win2019_1809_x64_dtc_en-us_40G_container_alibase_20240220.vhd	WindowsServer2019	x86_64	BIOS	40 GiB	1.18 and later
Windows Core	Windows Server Core, version 2022 (20240223)	wincore_2022_x64_dtc_en-us_40G_container_alibase_20240223.vhd	WindowsServer2022	x86_64	BIOS	40 GiB	1.18 and later
Alibaba Cloud Linux 2 (EOL)	Alibaba Cloud Linux 2.1903 LTS 64 bit	aliyun_2_1903_x64_20G_alibase_20231221.vhd	Aliyun	x86_64	BIOS	20 GiB	1.30 and earlier
CentOS (EOL)	CentOS 7.9 64 bit	centos_7_9_x64_20G_alibase_20230718.vhd	CentOS	x86_64	BIOS	20 GiB	1.30 and earlier

Usage notes

Limits on OS images

OS images that are not supported by the instance types you specify are automatically filtered out. If you specify multiple instance types, only the OS images that are supported by all of the instance types are displayed.

OS type	Limits
Windows	You cannot create Windows images when you create an ACK managed cluster. You must create an ACK managed cluster first. You cannot use Windows images in the following scenarios: Your cluster uses node pools other than managed node pools. Your cluster uses the Terway network plug-in. Nodes in your cluster use containerd as the container runtime. Your instance meets the requirement, such as a non-GPU-accelerated instance with at least 4 vCPUs and 8 GB of memory is used. For more information about limits and precautions when you use Windows images, see Limits.
Alibaba Cloud Linux	You can use only Alibaba Cloud Linux images in the following scenarios: Your cluster uses the Terway-Eniip network plug-in and IPVLAN or DataPath V2 is selected. The container runtime is Sandboxed-Container. You want to create an ACK managed cluster for confidential computing.
Alibaba Cloud Linux 3 for ARM	Only instance types that use the ARM architecture are supported.
Ubuntu	The Kubernetes version of the cluster is 1.30 or above. For more information, see Manually upgrade ACK clusters. Cloud Parallel File Storage (CPFS) volume, image acceleration plugin, and security hardening are not supported.
RHEL 9.3	You can install the RHEL operating system in an ACK cluster by using a custom image. For more information, see Create a custom image from an instance. After a custom image is created, you can use the image to create a node pool. All nodes in the node pool are deployed based on the custom image. View the operations Log on to the ACK console. In the left-side navigation pane, click Clusters. On the Clusters page, find the cluster that you want to manage and click its name. In the left-side navigation pane, choose Nodes > Node Pools. In the upper-right corner of the page, click Create Node Pool. Find Custom Image below advanced options and select the image that you created. After you confirm the configuration, the node pool is created. Note This feature is only available to whitelist users. submit an application in the Quota Center console. For more information about how to create a node pool, see Create a node pool.
Custom OS images	Create a custom image based on the operating system supported by the ACK cluster. Do not create custom images from ECS instances running in the ACK cluster. Remove them from the cluster first if necessary. For more information, see Remove a node. The predefined behavior logic in a custom image may affect operations such as cluster node initialization, container launching, node updates, and automatic recovery of nodes in a managed node pool. Before you use it in a production environment, ensure that the custom image has been tested and validated. This feature is only available to whitelist users. submit an application in the Quota Center console.

cgroup version description

Linux provides two versions of control groups: cgroup v1 and cgroup v2, to limit, record, and isolate the physical resources such as CPU, memory, and I/O used by process groups. cgroup v2 is the optimized newer version of the cgroup API.

There are differences between the common interfaces and subsystem interfaces of cgroup v2 and cgroup v1. For more information, see Differences between cgroup v1 and cgroup v2.

cgroup v1 and cgroup v2 use different APIs. If your Java applications are deployed on an operating system that supports cgroup v2 and require access to the cgroup file system, make sure that the tool versions are compatible. For more information, see Migrate to cgroup v2.

Category	Description
cAdvisor	If you use cAdvisor as an independent DaemonSet to monitor pods and containers, you must update its version to 0.43.0 or later.
Java applications	DragonWell: 11.0.16.12, 8.15.16-GA or later versions. OpenJDK/HotSpot: jdk8u372, 11.0.16, 15 or later versions. IBM Semeru Runtimes: 8.0.382.0, 11.0.20.0, 17.0.8.0 or later versions. IBM Java: 8.0.8.6 or later versions.
Go applications	Update uber-go/automaxprocs to 1.5.1 or later versions.

References

You can use node pools to facilitate node management. For example, you can schedule, configure, manage, or maintain nodes by node pool, and enable auto scaling for a node pool.
Managed node pools, introduced by ACK, are capable of automating O&M tasks for specific nodes.
A node pool consists of nodes that have the same configuration or nodes that are used for the same purpose. You can create node pools to easily manage and maintain nodes. For more information, see Create a node pool.
You only need to create nodes when you create an ACK managed cluster. The control planes of the cluster are created and managed by ACK, which reduces O&M costs and allows you to focus on application development.