cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

Dynatrace Problems affected Entities only show Service Level, but not Request.

Santhosh_1393
Frequent Guest

Hi There, Whenever we query for Davis problems, in the affected_entity_ids there are only, Services, hosts, or Applications, custom devices, etc. But there is no Request or endpoints affected. This is a much needed feature to pinpoint the problem. 
I'm surprised why this is not yet implemented?

In dynatrace, all the endpoints like /abc/cde/login will all be saved/referred to as SERVICE_METHOD-23K455..... right?
Why not DAVIS AI show the affected entity type as Service Method, instead of just service.

Please share insights.

3 REPLIES 3

Theodore_x86
Helper

Hello @Santhosh_1393 

I am pretty sure that affected service methods are described on a problem. I' ve seen it plenty of times.

However, it is possible that non of the service methods breach any baseline, but the Service all together operates out of expected behavior and so the problem shows affected only the service itself. Hope I am not confusing you but this is the case more or less.

BR

Theodore

Houston, we have a problem.

You see, The Service methods you saw are only displayed in the UI page, but not queryable as a field. They dont show up in affected_entity_ids. 
Which is what we needed, the only thing remote is it shows up somewhere in description field, and it gets burried, in some event description, where it says, Multiple service Problems detected, service A, endpoint B.. and 11 more..

had it been queryable, in affected_entity_ids or some other new field, It would prove a huge value. 

mgome
Advisor

SERVICE_METHOD entities are the product of marking key requests in a service.  

https://docs.dynatrace.com/docs/observe/applications-and-microservices/services/analysis/monitor-key...

There's a limit to the number key requests you can have per environment (500) and per service (100). Alternatively, you can use custom service metrics to get around this limitation. Tracking all requests/endpoints in all services individually would put significant load on the metrics database, especially in the case of services with a large number of unique endpoints. 

Featured Posts