In recent months Sippy has gained the ability to publish metrics and fire alerts to specific parties when problems are detected. Today this capability is mainly used for notifying specific teams who have a vested interest in a particular job when it's pass rate drops below some threshold.

TRT believes this capability can be significantly expanded to detect and notify us of a number of classes of problems which are difficult for humans to notice, or sometimes are missed simply because nobody happened to look at the right place.

This related to the 4.Next effort as one of the primary wins here would be to automatically detect problems that are specifically showing up on the build cluster that has been upgraded to the sprintly release.