Skip to main content

Using Unit Dashboard

Introduction

Unit Dashboard designed to show full Unit status and all its monitoring. Here you can see all running services on nodes, their resource usage. Also, here you may see living service migration between nodes.

How to open Unit dashboard

  1. Login as OEM or FleetOwner user.
  2. Got to Unit Details Page and click Dashboard button. Link to unit dashboard

Overview

use-unit-dashboard-overview.png

Show all nodes or single node

All nodes view

This mode is default. Here all nodes are displayed simultaneously but only with CPU or MEM monitoring. use-unit-dashboard-all-node.png To change mode, click on the highlighted button.

Single node view

In this mode, only one selected node is displayed with all available monitoring. use-unit-dashboard-one-node.png To change displayed node, click on green node name like on previous image.

What you see on dashboard

use-unit-dashboard-description.png

  • On the top of the node, you may see available resources on the node and its status.
  • Next is a timeline graph with consumed resources per service over time.
  • At the bottom, you may see a list of installed Services with their monitoring. The Status of the service is displayed as green or red dot to the left service name.
  • System row represent resource usage by OS and Core services.
  • Total is the sum of all above.

Settings

To open setting click on settings icon the top the Page. Refer to Overview chapter where to find it.

use-unit-dashboard-settings.png

FieldSizeDefaultDescription
Refresh Unit TimeoutSeconds5How often refreshes Unit status (online, offline), services status (running, installing, error, moved to another node). The less period, the sooner changes will be reflected
Refresh Monitoring TimeoutSeconds30How often monitoring information if pulled from server. The best parameter here is how often Unit is sending monitoring to the server. Default unit will send monitoring every 30 seconds.
Graph durationMinutes3Period represented on timeline graph
Graph delaySeconds30Pull timeline to the right for n seconds. This is usefull to hide empty chart on time while monitoring not received
Aggregate monitoringBooleanTrueShould periods be aggregated
Aggregate everySeconds10Duration of time between windows Influx docs
Aggregate periodSeconds10Duration of windows Influx docs