Call Insights enables customers to evaluate a call efficiently without reviewing all the various call metrics. The highlevel information is available on the Call Debug Level dashboard , indicated by the following panel visible above the map:
Clicking on the panel opens the Single Call Insights Dashboard.
On the dashboard there are several tables that give information on the types of insights we've found, the users impacted by them and the networks the users who experienced them were working on. Insights are comprised of an analysis of several factors occuring during the call in certain combinations. They generally point at unusual or potentially problematic situations and can help administrators in determining the causes of issues more quickly. The list of Insights as determined so far is listed below and this can be exapanded with more insights as we go forward.
Elements:
- KPI's: Provide general overview information about the users for which insights were registered (excluding users in the call without insights!)
- Insights table: Lists each individual Insight per user with additional details about the users networking situation during the call. A single insight can occur multiple times and users can have multiple insights but an insight can not occur multiple times for the same user.
- Device Platform(s) Used table: Shows each user with the platforms he/she used during the call and indicates when changes took place. Use the Information text (
icon) in the top to see which number corresponds to which platform.
List of Insights
Current list of possible Insights
| Insight Message | Details |
|---|---|
| Potential Firewall Port Exhaustion Detected | Conditions:
|
| Audio stream was transferred between mobile and desktop versions of Teams | This can point at the user having problems with a device and switching to see if it improves. However, this can of course also happen for other reasons. |
| More than 5 BSSIDs were used | The user's device connected to more than 5 different (distinct) Wi-Fi AccessPoints during the call. This can occur when users move during a call but can also point at users who are in an environment where there are access points that service overlapping areas. This can cause unwanted/unnecessary switching between access point and have a detrimental effect on the user's call experience. |
| TCP was used for Audio traffic | UDP is always recommended for performance and preferred by Microsoft Teams, TCP is a fail back for unreliable connections or cases where UDP is blocked. Users using TCP during calls can point at configuration and or network policies that are misconfigured |
| Non-default Audio Ports Detected | Audio should use the local ports of 50000-50019. Other ports could result in QoS policies failing to identify and prioritize the traffic. |
| User Experienced Mid-Call Failure(s) | As reported by teams client and/or call records |
| Mid-Call Network Disconnect/Reconnect | Windows OS detects a network change event during the call |
| Audio switched between wired and Wi-Fi In-Call | Teams Audio stream changes between Wi-FI and Wired connections during a call, and this change was not the result of a network change event |
| Multiple Audio Segments were used | More than 1 distinct audio segment AND no signs of previously listed network or teams issues. (disconnect, midcall failure, etc) |