Container Intelligence Service (CIS) provides the cluster check feature. Before major operations like upgrading or migrating a cluster, run a cluster check to ensure all requirements are met. This topic describes how to use the cluster check feature.
Prerequisites
-
A Container Service for Kubernetes (ACK) cluster is created. For more information, see Create an ACK managed cluster.
-
Ensure your cluster is in the Running state. You can verify the status on the Clusters page of the ACK console.
Cluster check
Container Intelligence Service (CIS) provides a wide range of cluster check capabilities, including cluster upgrade checks, cluster migration checks, component installation checks, component upgrade checks, and node pool checks. Before you upgrade or migrate a cluster, install or upgrade a component, or upgrade a node pool, CIS automatically triggers a precheck. You can proceed only if the cluster passes the check. You can also manually run a comprehensive check on your cluster resources, cluster components, and cluster configurations at any time to get optimization suggestions. We recommend regularly checking and maintaining your cluster to avoid security risks.
Procedure
The procedures for running cluster upgrade checks, component checks, cluster migration checks, and node pool checks are similar. The following steps use a cluster upgrade check as an example.
Log on to the ACK console. In the left navigation pane, click Clusters.
-
On the Clusters page, click the name of your target cluster. In the left-side navigation pane, choose .
-
In the left-side navigation pane of the CIS console, choose , and then click Run Upgrade Check.
-
In the Upgrade Check panel, read the notes, select the I know and agree checkbox, and then click Run Check.
On the Upgrade Check page, you can view a check overview. You can also click View Details to view the progress of the cluster upgrade check on the Report page.
Check results
After the check is complete, review the Report page and address any issues found.
Check results are categorized as follows:
-
Normal: No action is required.
-
Warning: Confirmation is required. You must address issues that could cause cluster exceptions.
-
Error: You must resolve these issues immediately to prevent cluster failure.
For each check item, click its status indicator to view details, including a description of the issue, its potential impact, and a recommended solution.
A cluster upgrade check flags the use of deprecated APIs as errors. We recommend fixing them to avoid security risks. For more information, see Upgrade a cluster and separately upgrade its control plane and node pools.
The report page categorizes results by cluster resources, cluster components, and cluster configurations. The check results for cluster components might show that even if individual components like Kube Proxy Worker, API Service, and each instance are Normal, a related check item still reports an Error. This error is indicated by a red square icon.