Host Metrics K8s Dashboard

This dashboard provides comprehensive monitoring of host-level system metrics for Kubernetes environments, offering detailed visibility into CPU usage, memory utilization, disk I/O, network performance, and filesystem usage across all nodes in your cluster.

Dashboard Preview

Host Metrics K8s Dashboard
Host Metrics K8s Dashboard

Dashboards → + New dashboard → Import JSON

What This Dashboard Monitors

This dashboard tracks essential host system metrics to help you:

  • Monitor System Resources: Track CPU and memory usage across Kubernetes hosts
  • Disk Performance: Monitor disk I/O operations, usage, and performance metrics
  • Network Activity: Observe network traffic, bandwidth utilization, and interface statistics
  • Filesystem Management: Monitor disk space usage, available capacity, and filesystem health
  • Infrastructure Health: Identify performance bottlenecks and resource constraints at the host level

Metrics Included

Overview Section

  • CPU Used: Current CPU utilization value
  • Allocatable CPU: Total CPU resources available for allocation
  • Memory Used: Current memory consumption in GiB
  • Allocatable Memory: Total memory available for allocation in GiB

Resources Section

  • CPU Usage Chart: Real-time CPU utilization breakdown by type (user, idle, system, softirq, wait, steal, interrupt, irq, nice)
  • Memory Usage Chart: Memory consumption breakdown by type (used, free, cached, slab_reclaimable, buffered, slab_unreclaimable)
  • CPU Usage by Namespace: CPU utilization per Kubernetes namespace
  • Memory Usage by Namespace: Memory consumption per Kubernetes namespace
  • CPU Usage by Pod: CPU utilization per individual pod
  • Memory Usage by Pod: Memory consumption per individual pod

System Section

  • System Load Average: System load averages over 1min, 5min, and 15min intervals

Network Section

  • Network Usage (Bytes): Network traffic in bytes for transmit and receive operations
  • Network Usage (Packet/s): Network traffic in packets per second
  • Network Errors: Network error rates and failed transmissions
  • Network Drops: Dropped network packets
  • Network Connections: Active network connections by TCP state (TIME_WAIT, ESTABLISHED, LISTEN, etc.)

Disk Section

  • System Disk I/O (Bytes Transferred): Data transfer rates for read and write operations
  • System Disk Operations/s: Disk operations per second for various devices
  • Disk Operations Time: Time taken for disk operations
  • Queue Size: Disk I/O queue size

Filesystem Section

  • File System Usage Table: Tabular view showing device, mountpoint, state, and filesystem type
  • File System Inode Usage Table: Inode utilization showing device, mountpoint, state, and type

Dashboard Variables

This dashboard includes pre-configured variables for filtering:

  • k8s.cluster.name: Filter by Kubernetes cluster name
  • k8s.node.name: Filter by specific node names

Was this page helpful?