alicloud-monitor-controller is a system component that is provided by Container Service for Kubernetes (ACK). This system component is integrated with Cloud Monitor. When an application is created, modified, or deleted, alicloud-monitor-controller automatically updates the metadata to Cloud Monitor. This allows you to monitor containers that are provisioned for an application. In addition, alicloud-monitor-controller provides alert settings. Developers can use this feature in the Cloud Monitor console.

Release notes

v1.4.0-49ff2362-aliyun
  • Metric collection is supported in an ACK cluster where both nodes that run Windows and nodes that run Linux are deployed.
  • The following scaling periods are supported: 15 seconds, 20 seconds, 30 seconds, and 60 seconds.
  • The following issue is fixed: Horizontal Pod Autoscaler (HPA) is activated during rolling updates of applications.

Troubleshooting

No application group is found in the Cloud Monitor console

Perform the following checks:
  • Check whether the pod where alicloud-monitor-controller is deployed runs as normal in the kube-system namespace.
  • Check whether alicloud-monitor-controller is upgraded to the latest version. We recommend that you upgrade the component to the latest version.
  • Check the logs of alicloud-monitor-controller to fix issues such as connection timeouts and SDK errors.

If the issue persists, use the following ticket template to submit a ticket.

Ticket template
  1. Is alicloud-monitor-controller upgraded to the latest version?

    Yes, alicloud-monitor-controller is upgraded to the latest version.

  2. Are SDK errors or connection timeouts found in the logs of alicloud-monitor-controller? If one of the preceding errors is found, delete the pod.

    No application group is found in the Cloud Monitor console after the pod is restarted.

  3. Provide a complete log file of alicloud-monitor-controller in the ticket.

    Upload a compressed package of the log file.

No data is found in the application group in the Cloud Monitor console

Perform the following checks:

Check whether the name of the instance in the application group is the same as that of the pod where alicloud-monitor-controller is deployed.
  • If the instance name is not the same as the pod name, perform the steps that are provided in the preceding section.
  • If the instance name is the same as the pod name, check whether the pod where the metrics-server component is deployed runs as normal. Check whether logs can be printed as normal. If the Successful write 164190 bytes metrics to monitor server error message is found in the log file, this indicates that the log can be printed as normal.

If the issue persists, use the following ticket template to submit a ticket.

Ticket template
  1. Is the name of the instance in the application group the same as that of the pod where alicloud-monitor-controller is deployed?

    Yes, both names are the same.

  2. Can the logs of metrics-server in the kube-system namespace be printed as normal?

    Yes, the logs of metrics-server in the kube-system namespace can be printed as normal.

  3. Provide the cluster ID, the name of the application, and the name of the pod in the ticket.