In this article, we will describe how Pings and Outages work in UNMS and how users can configure those features.
Table of Contents
There is a global ping setting in UNMS options but you can overrule it for an individual device if necessary. Go to the device -> Settings -> UNMS -> Device specific settings and turn them ON.
In this guide, we will talk in details about how exactly UNMS uses Ping for several different purposes and how Outages are configured.
Firstly let's talk about different PING tests in UNMS:
Latency PING: In order to measure a device's latency UNMS can instruct that device to send a ping. By default, such ping is targeted to UNMS server, but users can redirect it to anywhere else. When the route is set to a different location and ping target is unreachable, the device's status will remain "Active" because the Monitoring PING is still going through. The situation will be visibly highlighted in the device's Latency graph (a Latency outage).
Monitoring PING: There is a WebSocket channel opened between a device and UNMS and both sides are periodically checking if the connection is healthy. If they cannot see each other for a user-specified amount of time, an outage is created and that device status is changed to "Disconnected".
PING for 3rd party device: This ping is actively sending from UNMS in order to monitor the connection to a 3rd party device.
It is possible to further configure the Latency ping through Settings -> Devices -> Ping Settings.
Ping Address: By default, the value of this setting is your UNMS hostname. It is, however, possible to change this address and ping a different target. In order to measure the ping to the Internet precisely, we suggest using 'ping.ubnt.com' in this field.
Ping interval (default): This value represents a period of time between Latency ping attempts while a device is running normally.
Ping interval (outage): This value represents a period of time between Latency ping attempts while a device is reporting a Latency outage. The idea is that a user probably wants the ping to be more frequent in case of a Latency outage so that the Latency graph is filled with accurate latency values as soon as possible.
This section influences how the Monitoring Ping behaves. Each value defines a time interval after which a device is considered to be in an outage state after the Monitoring Ping failed. There are different times according to some situations as described below:
Running device reportable outage: This value represents a time interval after which a device is considered in an outage state if it was running normally before.
Restarting device reportable outage: This value represents a time interval after which a device is considered in an outage state if it was in the process of restarting before.
Upgrading device reportable outage: This value represents a time interval after which a device is considered in an outage state if it was being upgraded before.
NOTE: When a device is in the maintenance mode, no outage is reported and no email alerts are sent.