Network Management

False Node Down Alerts Due to High Network Latency

This article describes the behavior of node alerts when nodes are flagged as down due to high ICMP latency. .

First published date

10/22/2018 9:29 PM

Last published date

10/10/2022 10:02 PM

Overview

When polling for a node's status via ICMP, by default, once an ICMP request exceeds 1000ms the device will be considered to have not responded. In charts, the availability of the device will be zero during this time and the latency will not be recorded 

Product section

Network Performance Monitor

Cause

Latency.

Resolution

Resolve this issue in either of the following ways:

 

  1. Address the high latency. If the high latency was a result of a network event, addressing the cause of such events will prevent this issue.
  2. If the high ICMP latency is expected, you can increase the default 1000ms timeout. To do this:
    1. In the Orion Web Console, choose Settings > All settings.
    2. Click Polling Settings.
    3. Adjust the ICMP timeout value.