topic Re: Most of problems without root cause in Alerting

Most of problems without root cause

ajdimanuel — Fri, 24 Feb 2023 12:38:44 GMT

Hi,

I've noted that most of problems that one customer has, doesn't have root cause associated neither Evolution Path analysis. Most of problems are "Degradation Time or Failure Increase" without Infrastructure component/Users (RUM not configured yet) affected.

Do you know why IA is not getting the root cause of these issues?

Thanks in advance.

Alberto.

Re: Most of problems without root cause

rastislav_danis — Tue, 23 Oct 2018 09:19:47 GMT

maybe there is no real root cause out of monitored resources for these service related problems

Re: Most of problems without root cause

wolfgang_beer — Tue, 23 Oct 2018 14:20:22 GMT

Root-causes are determined by health state of related components along the transactions. So if there is no related unhealthy component there will be no root cause.

An example would be a RUM JavaScript error rate increase that is triggered by some users using exotic browser versions. This situation does not really have a root-cause on back-end side.

Also health state of components is defined as having an event open either through a metric baseline or threshold. If there is no event, there will be no root cause.

We are in the process of releasing our new AI causation engine that is able to capture root causes even if there is no event detected on a component.

Re: Most of problems without root cause

ajdimanuel — Tue, 23 Oct 2018 14:54:30 GMT

Ok Wolfgang, thanks for your answer.

Alberto.