Proactive detection of problems with Application Insights

Surely you must be tired of hearing to be proactive is better than reactive – especially when it comes to problems in the user application, right?

The Application has a very nice feature Insights of proactive detection of problems. Want to know a little about it?

Proactive detection monitors the data sent by your site or application to Application Insights from machine learning algorithms and data mining, attempts to identify patterns of behavior that may indicate problems that would be difficult to diagnose manually.

According to the documentation on the site:

Proactive Detection performance discovers anomalies in your app by analyzing the telemetry it sends to Application Insights.

In particular, it finds performance issues that only affect some of your users, or only affect your users in some cases.

For example, it can notify you if your app pages load much more slowly on one type of browser than others, or if requests are served more slowly from a particular server. It can also discover problems associated with combinations of properties, such as slow page loads in one geographical area at particular times of day.

In other words, performance problems can be correlated automatically from common standards, such as the location of users (can indicate problem in a regional CDN, for example) or the devices that are using (for example, a new JS library not working right in the browser of those devices).

When a situation is detected, you receive an email like below. In fact, this is a real email we receive warning of problems at one of our sites:

Email notification of proactive detection
Email notification of proactive detection

To enable proactive detection for a site that is already being monitored by the Application Insights, open the "Blade" (blade) to Application Insights of your site, click the tile of proactive detection (1), then click on Settings and enable the detection (2). Don't forget to inform the emails of people who must be notified (3).

Proactive Detection configuration in Azure
Proactive Detection configuration in Azure


And then, what did you think? Be sure to comment!


A hug,

Author: Igor Abade

Igor Abade V. Leite ([email protected]) is a Visual Studio ALM MVP (Microsoft Most Valuable Professional) since 2006. Speaker at various Software Development community events (TechEd Brasil, The Developers’ Conference, DevOps Summit Brasil, Agile Brazil, Visual Studio Summit, QCON among others), has also written articles in magazines and websites such as MSDN Brazil. Since March/2011 is one of the owners of Lambda3, a Brazilian consulting company specialized in ALM, software development and training. Visit his blog about VS ALM at and follow him on Twitter @igorabade.

Leave your comment!