KubeVirt

Metrics, Dashboards, Alerts and more for KubeVirt Integration in Sysdig Monitor.

This integration is enabled by default.

This integration is out-of-the-box, so it doesn’t require any exporter.

This integration has 31 metrics.

List of Alerts

Alert	Description	Format
[KubeVirt] Virtual Machine Instances Not Running	Virtual Machine Instances are not running	Prometheus
[KubeVirt] High vCPU Wait Time	vCPU delay is too high	Prometheus
[KubeVirt] Memory Pressure Low Available Memory	Available memory is below 15%	Prometheus
[KubeVirt] High Network Error Rate	Network error rate is above 1%	Prometheus
[KubeVirt] Storage I/O Latency High	Average storage I/O latency is above 50ms	Prometheus

List of Dashboards

KubeVirt Virtualization Overview

The dashboard provides information on the status and resource consumption of KubeVirt virtual machines. KubeVirt Virtualization Overview

OpenShift Virtualization VM Details

The dashboard provides information on the details of OpenShift VMs. OpenShift Virtualization VM Details

List of Metrics

Metric name
kubevirt_info
kubevirt_vm_info
kubevirt_vmi_cpu_system_usage_seconds_total
kubevirt_vmi_cpu_usage_seconds_total
kubevirt_vmi_cpu_user_usage_seconds_total
kubevirt_vmi_filesystem_capacity_bytes
kubevirt_vmi_filesystem_used_bytes
kubevirt_vmi_info
kubevirt_vmi_memory_available_bytes
kubevirt_vmi_memory_cached_bytes
kubevirt_vmi_memory_domain_bytes
kubevirt_vmi_memory_pgmajfault_total
kubevirt_vmi_memory_pgminfault_total
kubevirt_vmi_memory_resident_bytes
kubevirt_vmi_memory_unused_bytes
kubevirt_vmi_network_receive_bytes_total
kubevirt_vmi_network_receive_errors_total
kubevirt_vmi_network_receive_packets_dropped_total
kubevirt_vmi_network_receive_packets_total
kubevirt_vmi_network_transmit_bytes_total
kubevirt_vmi_network_transmit_errors_total
kubevirt_vmi_network_transmit_packets_dropped_total
kubevirt_vmi_network_transmit_packets_total
kubevirt_vmi_storage_iops_read_total
kubevirt_vmi_storage_iops_write_total
kubevirt_vmi_storage_read_times_seconds_total
kubevirt_vmi_storage_read_traffic_bytes_total
kubevirt_vmi_storage_write_times_seconds_total
kubevirt_vmi_storage_write_traffic_bytes_total
kubevirt_vmi_vcpu_delay_seconds_total
kubevirt_vmi_vcpu_wait_seconds_total

Prerequisites

None.

Installation

Installing an exporter is not required for this integration.

Agent Configuration

The default agent job for this integration is as follows:

- job_name: "kubevirt-metrics"
  scrape_interval: 30s
  scheme: https
  tls_config:
    insecure_skip_verify: true
  bearer_token_file: /var/run/secrets/kubernetes.io/serviceaccount/token

  kubernetes_sd_configs:
    - role: pod

  relabel_configs:
    - action: keep
      source_labels: [__meta_kubernetes_pod_host_ip]
      regex: __HOSTIPS__

    - action: keep
      source_labels: [__meta_kubernetes_pod_label_prometheus_kubevirt_io]
      regex: true

    - action: replace
      source_labels: [__meta_kubernetes_pod_name]
      target_label: kube_pod_name

    - action: replace
      source_labels: [__meta_kubernetes_pod_uid]
      target_label: sysdig_k8s_pod_uid

    - action: replace
      source_labels: [__meta_kubernetes_pod_container_name]
      target_label: sysdig_k8s_pod_container_name

    - source_labels: [__meta_kubernetes_pod_container_port_name]
      action: keep
      regex: "metrics"

  metric_relabel_configs:
    # Drop the kubevirt REST client histograms. They are high-cardinality
    # (hundreds of series per virt-* pod) and are not used by any KubeVirt
    # dashboard or alert. On busy nodes they consume the agent's Prometheus
    # metric budget and can cause the per-VMI metrics (kubevirt_vmi_*) from
    # the local virt-handler to be dropped, leaving the dashboards empty.
    - action: drop
      source_labels: [__name__]
      regex: "kubevirt_rest_client_.*"