Container Service for Kubernetes (ACK) regularly releases new OS image versions, introducing new features, performance optimizations, and bug fixes. We recommend that you update the OS image of your node pools promptly. Additionally, you can switch the operating system type as needed, such as replacing an end-of-life (EOL) operating system with a supported one. For more information about the operating systems supported by ACK, the latest versions of OS images, and the usage limitations for certain operating systems, see Release notes for OS images.
Usage notes
This operation updates the operating system in batches by replacing the node system disk. Important data should not be stored in the system disks, or should be backed up before the update.
Data disks are unaffected during the update. We recommend that you perform this operation during your off-peak hours.
When you update a node by replacing system disks, ACK drains the node and evicts the pods from the node to other available nodes based on PodDisruptionBudget (PDB). To ensure high service availability, we recommend that you use a multi-replica deployment strategy to distribute workloads across multiple nodes. You can also configure PDB for key services to control the number of pods that are interrupted at the same time.
The default timeout period for node draining is 30 minutes. If the pod migration fails to be completed within the timeout period, ACK terminates the update to ensure service stability.
By default, some ACK operating systems use cgroup v2. For more information about cgroup v2, see cgroup version description.
Nodes with custom images are not eligible for this operation.
For free nodes, which are worker nodes not managed by a node pool, refer to Add free nodes to a node pool for migration instructions.
If your node pools use custom NVIDIA driver versions specified by adding labels or OSS URLs, OS image upgrades may cause compatibility issues. Always select the latest drivers from the ACK-supported NVIDIA driver versions.
Procedure
To update the OS image to the latest version or to switch the operating system type, perform the following steps:
Log on to the ACK console. In the left-side navigation pane, click Clusters.
On the Clusters page, find the cluster to manage and click its name. In the left-side navigation pane, choose .
In the Node Pools list, find the node pool that you want to update, and click
> Change Operating System in the Actions column.Choose an available image version, select the nodes for update, and specify whether to create a snapshot for the node. Then, set the batch replacement policy, and click Start Replacement.
Update Method:
Create Node Snapshot before Update: If your system disk contains critical data, you can create a snapshot of the node before updating the operating system to facilitate data backup and recovery. For more information about the charges incurred by snapshots, see Snapshot billing. If you no longer need the snapshot after the update, delete it promptly.
Batch Replace:
Maximum Number of Nodes per Batch: The system sequentially updates nodes based on the configured maximum number per batch.
Automatic Pause Policy: The policy for pausing during the node update process.
Interval Between Batches: If the automatic pause policy is set to automatic, choose whether a time interval is needed between each update batch and specify the duration of the interval. The configurable range is 5 to 120 minutes.
References
For more information about the release notes of OS images supported by ACK, see Release notes for OS images.
For guidance on updating the kubelet and container runtime of a node pool, see Update a node pool.
For details about updating by replacing system disks, see Reference: In-place updates and updates by replacing system disks.