2 min

Tags in this article

, ,

Rollbar today announced an upgrade of its error monitoring platform for developers, Adaptive Alerts. The upgrade adds real-time error alerts that warn users of unexpected issues in monitored apps using anomaly detection to lower false negatives and false-positive results.

Developers and operations teams would love to detect errors before deployment but that is more of an ideal than a feasible possibility at the moment.

So, error monitoring services come in handy to detect as much as possible before deployment. However, these services operate on threshold reporting metrics or levels, which can yield a lot of noise and increase the grunt work.

A next-gen upgrade

Brian Rue, co-founder, and chief executive of Rollbar said that Adaptive Alerts is the next-gen offering in trendline alerting. Compared to the last generation (10^nth Occurrence and High Occurrence Rate), Rue said that Adaptive Alerts sends 86% fewer notifications.

The reason? Adjusting thresholds and a broader exception-level view that effectively finds app-level trends.

Enterprise apps are hosted in complicated environments where serious errors can appear. Many of them are not a priority because they are not critical and have low occurrence rates. Developers want to know when significant trendline changes occur so they know what to prioritize.

Cutting through the noise

Rollbar’s algorithms leverage a continuous monitoring solution that learns from the past two weeks of data to create a trend for every app.

Adaptive Alerts then uses the baseline created to determine what is an anomaly and sends alerts in minutes. Users can adjust the minimum exception rates to alert and fine-tune anomaly detection to sift through the noise.

The new Adaptive Alerts feature will be available to customers of the Advanced and Enterprise plans.