Database service alerting - show failed services calling database in problem/root cause

john_martin — Tue, 25 Mar 2025 15:32:14 GMT

We're seeing a recurring issue where Dynatrace detects failed connections to oracle database instances, however Davis is only showing the number of failed connections, not which services calling that database that are failing.

In the very rare instance where a database is totally unavailable, this is sufficient. However the vast majority of issues are specific to only a single service failing, not all services. Without this specific information in the Problem and root cause, we have to click within the problem to get into the backtraces and then filter only on failed requests to identify which service(s) is failing.

Is there a good work around for this? Would like to automate these problems to get to the correct app groups rather than dumping all failed database connects over to the database engineers.

Re: Database service alerting - show failed services calling database in problem/root cause

marco_irmer — Thu, 27 Mar 2025 23:17:57 GMT

This sounds like a good idea for a possible enhancement. We see these problems in our environment also from time to time. It would seem that the calling service is only ever mentioned right in the problem when its own health is impacted sufficiently to trip the anomaly detection. One possible workaround might be to look for ways to generate davis events whenever an app service experiences database connection failures, in the hopes that Davis would then correlate both events into a single problem.

topic Database service alerting - show failed services calling database in problem/root cause in Alerting

Database service alerting - show failed services calling database in problem/root cause

Re: Database service alerting - show failed services calling database in problem/root cause