ApsaraDB for SelectDB allows you to update the kernel version to enhance performance, add new features, or fix known issues. Using an outdated kernel version can lead to service interruptions and data loss. This topic describes how to update the kernel version of an instance in the ApsaraDB for SelectDB.
Usage notes
When a new kernel version is released, ApsaraDB for SelectDB sends a notification to the contacts configured for your Alibaba Cloud account. Update the kernel version promptly.
If you do not update to the latest kernel version within six months of receiving the notification, you are responsible for any resulting service interruptions, data loss, or other consequences.
After you update the kernel version, you cannot roll it back to a previous version.
During a kernel version update, you can only view cluster information. You cannot perform operations such as scaling, pausing, or deleting a cluster.
Impacts
A minor version update causes transient connections on the instance, and some read and write requests may fail briefly.
A major version upgrade causes downtime, and the instance may be unavailable for about 10 minutes. We recommend that you perform this upgrade during off-peak hours.
Rolling update: Nodes within a cluster are updated sequentially. The cluster remains available, but transient connections may occur.
Parallel update: All nodes within a cluster are updated simultaneously. This causes the cluster to be unavailable for 1 to 5 minutes. We recommend that you perform this operation during off-peak hours.
Procedure
Log on to the SelectDB console.
In the upper-left corner of the page, select the region where your instance is located.
On the Instances page, click the Instance ID of the target instance to open the Instance Details page.
In the Configuration Information section, next to Version Upgrade, click Minor Engine Version.
In the dialog box that appears, configure the following parameters.
Parameter
Description
target version
Select the version to which you want to update the instance.
update method
ApsaraDB for SelectDB instances support the following update methods:
Upgrade Now: The instance is updated immediately.
Update Within Maintenance Window: The instance is updated within the maintenance window that you have configured. For more information about how to set a maintenance window, see Set a maintenance window.
Parallel Operation Node
Select whether to perform a parallel update. By default, a rolling update is performed.
ImportantFor a major version upgrade, we strongly recommend that you select Parallel Operation Node. The default rolling update may cause protocol incompatibility issues between versions.
If you do not select Parallel Operation Node, a rolling update is performed. It has the following characteristics:
Nodes within a cluster are updated sequentially. The cluster remains available, but transient connections may occur. This is the recommended method for minor version updates.
If you select Parallel Operation Node, a parallel update is performed. This method has the following characteristics:
All nodes in a single cluster are updated simultaneously, which makes the cluster unavailable for 1 to 5 minutes.
The cluster is briefly unavailable during the update, but the process is faster. We recommend that you perform this operation during off-peak hours.
Click OK.