We deployed Dynatrace OneAgent on several Openshift nodes on which are deployed all the applications environments (dev, preprod, prod). As the OpenShift hosts are installed in our Production tenant, Dynatrace detects problems on all the applications environments and we receive alerts for applications that are not production.
Is there a simple way to disabled problem detection on non-prod environments? I created management zones to distinguish the various environments but it does not help for this kind of issue.
Solved! Go to Solution.
Management Zones and Tags are the right way to go in combination with alerting profiles.
You can configure your alerting profiles so you are only notifited about, and see the problem indicator in the UI for, problems from production systems.
Thanks for your quick answer.
What I did is:
- created a "Production" auto tag on all non-Openshift processgroups and services
- created a "Openshift Prod" auto tag on all Openshift Prod processgroupes and services
- created auto tag for other Openshift instances
- created a new alerting profil based on "Production" and Openshift Prod" tags
That way, I should not receive alerts for non-production entities.
Is there a way to exclude some entities from problem detection the same way?
Yes, you can combine tags for alert rules with either any or all tags.
If you don't want alerts for all production machines for example, you could create an alert tag, and match for entities containing the alert and production tag.
Also note that if all of your process groups are being detected together (i.e. big groups that contain the same application in dev, QA, prod etc...) you may need to look at splitting those up so that the process groups are split by by dev, QA, prod. Otherwise even with management zones you may have issues with troubles in pre-prod environments creating problems tagged as prod. Hostgroups are usually helpful in doing that.
No, it's OK, my pods are all separated based on the Kubernates Namespace that contains the environment.