#Monitoring and Alerts

Monitor and alert on virtual machines in terms of CPU, memory, storage, and network. To facilitate timely alerts, notification policies can also be configured.

The intuitively presented monitoring data can be used to provide decision-making support for operations inspection or performance tuning, while the comprehensive alerting and notification mechanism will help ensure the stable operation of virtual machines.

#Monitoring

By default, the platform collects commonly used performance monitoring metrics for virtual machines, including CPU, memory, storage, and network. Navigate to Virtualization > Virtual Machines, and on the Monitoring tab in the virtual machine details, you can view real-time monitoring data for the metrics.

#Alerts

#Configuring Alert Policies

To enable alerts, you must first create an alert policy. An alert policy describes the objects you wish to monitor, the conditions under which you wish to be alerted, and how you will be notified of relevant alerts. Navigate to Container Platform > Virtualization > Virtual Machines, and in the virtual machine details, click Create Alert Policy on the Alerts tab to complete the configuration.

Tip: The virtual machine alerting function is similar to the platform's general alerting function. For more detailed configuration guidance, refer to the platform's general alerting documentation.

#Handling Alerts

Navigate to the Alerts tab, and if there are alert status strategies indicated, please address them promptly.

#Binding Notification Policies

In addition to real-time alerts on the Alerts tab, the platform also supports sending alert information via email, SMS, and other means to relevant personnel, notifying them to take necessary measures to resolve issues or prevent failures. The notification policy needs to be set up by contacting the administrator.

#Using the API

Alert policies are backed by Prometheus rules. You can also define an alert directly as a PrometheusRule (monitoring.coreos.com/v1):

apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
  name: vm-high-memory
  namespace: demo
spec:
  groups:
    - name: vm.rules
      rules:
        - alert: VMHighMemoryUsage
          expr: kubevirt_vmi_memory_used_bytes / kubevirt_vmi_memory_available_bytes > 0.8
          for: 3m
          labels:
            severity: warning
          annotations:
            summary: "VMI {{ $labels.name }} memory usage above 80%"

KubeVirt emits per-virtual-machine-instance metrics (prefix kubevirt_vmi_*, with namespace and name labels) that you can query in Prometheus or use in your own dashboards — for example kubevirt_vmi_memory_used_bytes, kubevirt_vmi_network_receive_bytes_total, kubevirt_vmi_storage_read_traffic_bytes_total, and kubevirt_vmi_migration_data_processed_bytes. Exact metric names vary by KubeVirt version; list the full set with the query {__name__=~"kubevirt_vmi_.+"}.