Multi-environment service dashboarding

[swin12]

Hi There,

I am looking for advice related to dashboarding service data for an application that is sending OneAgent data from multiple environments. So far, I have created management zones that allow me to filter the Dynatrace SaaS tenant such that I can see only service data for a single environment at a time.

Then, I put together a dashboard that visualizes several things:

1. Host and Service health honeycomb tiles.
2. Top 10 lists of slow, erroring, or high count services.
3. Metrics for a specific service.
4. Metrics for a specific key request of a service.

Using a single dashboard, I can see tiles #1 and #2 above dynamically change when I select a different management zone. However, because the data explorer used for tiles #3 and #4 involves me selecting an environment-specific service or key request ID, I do not see the same effect when I change management zones (those tiles don't show data when the zone that doesn't contain that service ID is chosen).

If I were to export the codified version of this dashboard as a baseline for another application environment, these hard-coded service IDs would cause some issues. This has lead me down the path of creating multiple dashboards, which I feel is also not ideal. So, given the following scenario... What would be the best way of displaying this data on a single dashboard (if possible)?

- Average response time for SERVICE-62CB109C62F0206B (service named "foobar" in UAT environment, visible in management zone named "UAT").

- Average response time for SERVICE-3E48C5B068474940 (service named "foobar" in PROD environment, visible in management zone named "PROD").

[Fin_Ubels · Accepted solution]

Hey swin12,

The easiest method you could use to resolve this is by putting all of the services or key requests on the one tile. Then, when you filter by your management zones it will include only the services or key requests that can be seen within that management zone. Below I have attached an example. You can see I have the same metric specified twice for the same graph but each one is filtered for a different request. Assuming each request is in a different management zone then it will display only the metric for requests you can see in the management zone you're in. To set this up you need to be in the "All" management zone but then afterwards you can filter by management zone to see the results.

This method is limited to 10 metrics so as long as you don't have more than 10 services/requests that you wish to dashboard then it should be fine.

A more flexible and better long term solution is having the metric defined and then split by service or split by request instead of filtering for a specific one. Now I understand this is not ideal if you have more than a couple in each management zone but it is more flexible as my first solution offered does not adapt to changes within your environment. To make this work you can reduce the number of requests or services that show up by using the entity selector to filter for only requests or services that match certain criteria. Below I have an example which filters out all requests with a name containing "/recipe/mushroom". To make this work within your environment you can use naming rules for the requests and services you want to dashboard to make them all have an identifier that you could then filter for using the entity selector.

Using this method allows you to easily add more services or requests in the future by configuring your naming rule.

Docs for the entity selector
https://www.dynatrace.com/support/help/dynatrace-api/environment-api/entity-v2/entity-selector

Docs for naming rules
https://www.dynatrace.com/support/help/how-to-use-dynatrace/services/service-detection-and-naming/cu...
https://www.dynatrace.com/support/help/how-to-use-dynatrace/services/service-monitoring-settings/set...