Cloud Native Cluster Monitoring Is Compatible with Self-Built Prometheus
Compatibility Mode of Cloud Native Cluster Monitoring
- If you have deployed Prometheus and your Prometheus is open-source and has not been deeply customized or integrated with your monitoring system, you are advised to uninstall it and use the Cloud Native Cluster Monitoring add-on instead. By doing so, you do not need to enable Compatibility Mode.
To uninstall your self-built Prometheus, ensure that all workloads of your self-built Prometheus and all CRDs ending with monitoring.coreos.com are deleted. For details, see How Do I Delete Self-Built Prometheus?.
- If your self-built Prometheus cannot be uninstalled and you need to use functions such as Cost Insights and Monitoring Center, you can enable Compatibility Mode when your self-built Prometheus meets Compatibility Requirements. For details, see Enabling Cluster Monitoring.
The compatibility mode does not support all functions of the Cloud Native Cluster Monitoring add-on. For details, see Restrictions on the Compatibility Mode.
Compatibility Requirements
Ensure that your self-built Prometheus meets the following requirements, or it cannot run normally.
- If your Prometheus is not based on kube-prometheus of the Operator community, the compatibility requirements are met.
- If your Prometheus is based on kube-prometheus of the Operator community, the prometheuses.monitoring.coreos.com version must be 0.8.0 or later.
You can run the following command in the cluster to check the version:
kubectl get crd prometheuses.monitoring.coreos.com -oyaml | grep controller-gen.kubebuilder.io/version
If the following information is displayed, the compatibility requirements are met:
Error from server (NotFound): customresourcedefinitions.apiextensions.k8s.io "prometheuses.monitoring.coreos.com" not found
If the following information is displayed and the version is 0.8.0 or later, the compatibility requirements are met:
controller-gen.kubebuilder.io/version: v0.9.2
Restrictions on the Compatibility Mode
- In compatibility mode, the Cloud Native Cluster Monitoring add-on is installed in the cce-monitoring namespace. By default, the ServiceMonitor and PodMonitor of another namespace are not identified.
- The Cloud Native Cluster Monitoring add-on can only be deployed without local data storage.
- You cannot disable compatibility mode once it is enabled. To use the normal mode, you can uninstall the Cloud Native Cluster Monitoring add-on first and install it again.
- Currently, cost optimization is not supported.
- Currently, metrics discarding on the AOM page is not supported.
- Currently, the ServiceMonitor and PodMonitor cannot be started or stopped on the AOM page.
How Do I Delete Self-Built Prometheus?
- Delete all workloads, services, and other resources related to self-built Prometheus, such as DaemonSets, Deployments, StatefulSets, and Services. Generally, these resources are in the monitoring namespace. Replace it with the actual namespace used during the add-on installation.
- Query all CRDs ending with monitoring.coreos.com.
kubectl get crd | grep monitoring.coreos.com | awk '{print $1}' | xargs
If the command output is empty, no further action is required.
- Delete all CRDs ending with monitoring.coreos.com.
kubectl delete crd $(kubectl get crd | grep monitoring.coreos.com | awk '{print $1}' | xargs)
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot