topic Re: Classical Dashboards in Dashboarding

Classical Dashboards

MarwanC — Fri, 19 Jan 2024 18:28:25 GMT

Dear Dynatrace Gurus,

before you say move to the new dasboards with DQL, I would like t keep the old classical dashboards for now.

I have a couple of questions, I hope someone can help me?

1. I need to modify this classical dashboard query to display two important metrics (see below),

In this I use tech.generic.cpu.usage

builtin:tech.generic.cpu.usage:filter(eq("dt.entity.process_group_instance",PROCESS_GROUP_INSTANCE-2F6198CAE881F38A)):splitBy("dt.entity.process_group_instance"):max

1. Process Group -> Tecnology Specific Metrics -> GC Time

2. Process Group -> Tecnology Specific Metrics -> GC Time

Where can I find the documentation for tech.generic.?.? (use to obtain more out of the box metrics)

2. I have a dashboard that uses the same query above, but if the process is down it says no data is available and you would see an empty no name chart. How can I show on the same dashboard to say process is down and not available and may be show as red - like Availability button.

Can someone help?

I appreciate your support.

Marwan...

Re: Classical Dashboards

MarwanC — Fri, 19 Jan 2024 18:29:14 GMT

Correction:

2. Process Group -> Tecnology Specific Metrics -> Response Time

Re: Classical Dashboards

dannemca — Fri, 19 Jan 2024 20:38:49 GMT

The doc for the tech generic metrics is here: https://docs.dynatrace.com/docs/observe-and-explore/metrics/built-in-metrics#generic

You can use the metric builtin:tech.generic.count to show if your process is Up (>1) or down (0). Try the single value visualization and mark 0 as red, >1 as green.

Re: Classical Dashboards

MarwanC — Sat, 20 Jan 2024 10:27:07 GMT

Thank you Dann, much appreciated it!

Re: Classical Dashboards

MarwanC — Sat, 20 Jan 2024 11:12:43 GMT

I used this for the CPU of that process like this

builtin:tech.generic.cpu.usage:filter(eq("dt.entity.process_group_ins

tance",PROCESS_GROUP_INSTANCE-98CB935943677AB7)):splitBy("dt.entity.process_group_instance"):max

Then If I use the single count with the metric that you recomened

builtin:tech.generic.count:filter(eq("dt.entity.process_group_instance",PROCESS_GROUP_INSTANCE-98CB935943677AB7)):splitBy("dt.entity.process_group_instance")

I get this error

The dimension key `dt.entity.process_group_instance` has been referenced, but the metric has no such key.

If I remove the split

builtin:tech.generic.count:filter(eq("dt.entity.process_group_instance",PROCESS_GROUP_INSTANCE-98CB935943677AB7))

I get this error

The dimension key `dt.entity.process_group_instance` has been referenced, but the metric has no such key.

I need to monior availability this particular process instance

PROCESS_GROUP_INSTANCE-98CB935943677AB7

Thanks for any feedback

Re: Classical Dashboards

dannemca — Mon, 22 Jan 2024 12:29:11 GMT

How many Process Instances you do have for this process Process Group? This metric will look for Process Instances in your Process Group. The Process Group Instance is not a valid dimension for this metric. You should use the Process Group as filter/split by.

Re: Classical Dashboards

MarwanC — Thu, 25 Jan 2024 09:51:12 GMT

There are two instances one each host and the code above works like a charm. I filter by host and technology to find availability of each process in one dashboard. Do you another way to do this?