Alert > Connectivity - arcgis_server_metrics (%) always maxed or nearly maxed out

[Brian_McLeer]

ArcGIS Monitor was deployed to monitor one of our ArcGIS Server machines about 3 months ago. There is one alert that has been maxed out or nearly maxed out since deployment titled "Connectivity - arcgis_server_metrics (%)". We deployed on 4/24/24, and it has been consistent since. Does anyone else have this appear as an alert consistently or have any tips? The metrics reference did not have much info on this. We are at Monitor version 2023.3.1. 

[Brian_McLeer]

In addition, there is a similar pattern with the metric "connectivity - arcgis_server_discover (%)" 

[GeoJosh]

Hi @Brian_McLeer,

The connectivity metrics measure the ability of ArcGIS Monitor to communicate with the component. A value of 100 means that ArcGIS Monitor can successfully communicate, whereas a value of 0 means communication has failed and the component is down / unreachable.

By default, the connectivity metrics open an alert if their value falls below 100 over two or more samples. Looking at your screenshots, there has been some interruptions in connectivity but monitor has generally been able to communicate with the component just fine.

Let me know if I can clarify further or if you have other questions.

Josh

[Brian_McLeer]

Thank you @GeoJosh , that helps explain things. You are right, our metrics are not that low from 100. 

[AndrewSakowicz]

@Brian_McLeer , to clarify, the connectivity has only 2 values: 100 and 0.  That's what you would see when zoomed in to a raw data time span.  When you zoom out beyond that, the default charts will show averages.  So anything less than 100, indicates there was a connectivity 0 in this range.  You can verify the duration by zooming in.

You can fix that "averaging" impact by building analysis and selecting min, e.g. Solved: Analysis from ESRI Developer Summit 2024 - Esri Community