Limit of number of metric dimension values

Platform Alerting
,
1

Hi,

I am working with metrics event and metric selector.

I don’t understand why there is so much dimensions with this simple request :

  • sql-server.host.cpus:splitBy(“dt.entity.sql:sql_server_host”):max:rollup(avg,5m)

jegron_0-1688638617091.png
jegron_0-1688638617091.png801×256 16.8 KB

In Data Explorer there is only 76 “dimensions”

jegron_1-1688639001923.png
jegron_1-1688639001923.png1236×661 27.2 KB

Also the documentation is unclear about the behavior once the limit has been reached :

  • THROTTLED: The configuration reached the query limits of the metric selector.

Does the alert stop working for the next 24 hours?

Thanks for your help :slightly_smiling_face:

2

Hi,

I would split it per host groups or hosts, if that is possible. It means, creating more than one and filtering entities.

Best regards

3

Thanks @AntonPineiro . Sadly adding splitby fields does not change the number of dimension in my case.

4

Hi,

I do not see another option than reduce scope. Maybe, you can ask Dynatrace ONE or raising some product idea. It is design limitation.

Best regards

5

Hi,
We have the exact same problem here , and we also have the same questions unanswered

sql-server.databases.state:splitBy(“dt.entity.sql:sql_server_database”,instance,server,database,“database.state”,“database.recovery_model”):filter(in(“dt.entity.sql:sql_server_database”, entitySelector(“type(sql:sql_server_database),tag(INOP-Plateforme-Bd:SB)”)),and(ne(“database.state”,RESTORING)))

6

I am also facing the same… in my case, I am testing the Davis Anomaly Detection app, with good results.