Managing Data Operating System
Also available as:
PDF
loading table of contents...

Monitoring Nodes

This page displays details about the nodes.

It contains three parts: Information, Node Status and Nodes Heatmap Chart.

Information

This tab displays all the node labels and their details including the rack of the node, node state, node address, node HTTP address, containers, memory used, memory available, VCores used, VCores available, last health update, and health report.

You can sort through any of the columns to view the details of the required nodes. You can also search to find the specific node labels from the entire list.



Node Manager Page

To view the Node Manager page of any node label, click on the Node HTTP address of the node.



The Node Manager page displays details about the node including the following:

  • Total virtual memory allocated for Containers.
  • Virtual memory enforcement enabled. It is displayed as a true or false value.
  • Total physical memory allocated for Containers.
  • Physical memory enforcement enabled. It is displayed as a true or false value.
  • Total VCores allocated for Containers.
  • Node Healthy Status
  • Last Node Health Report time
  • Link to the Node Health Report, if generated
  • Node Manager Start Time
  • Node Manager version
  • Hadoop version

You can also view the resource usage in the following categories in a pie-chart representation:

  1. Memory
  2. VCores
  3. yarn-io/GPU

Node Status

This tab displays the node managers in a pictorial representation. It displays details such as the number of active nodes, number of unhealthy nodes, decommissioning nodes, and the number of decommissioned node managers.

Nodes Heatmap Chart

This tab graphically displays nodes on the basis of their usage of memory. You can enter host/rack details in the search bar to filter nodes.