topic Monitoring Mode Count in Dashboarding

Monitoring Mode Count

MikeL — Thu, 21 Aug 2025 07:25:52 GMT

Is their a way in DT Dashboard to create a single Tile or Table showing the count of Host set in Full Stack & Infra mode. Then if that count changes, the tiles changes color based on simple threshold, simply way to check dynamically.
Script example or Demo Dashboard we can clone to grab that tile.
We have some scenarios where some of our hundreds of VM Monitoring Mode get changed without notice.

Re: Monitoring Mode Count

dannemca — Wed, 20 Aug 2025 17:00:54 GMT

Yes, you can

Using this simple DQL you can achieve this:

fetch dt.entity.host | fields entity.name, monitoringMode | filterOut isNull(monitoringMode) | summarize Total = count(),by:{monitoringMode}

Then you play around with visualizations and thresholds to achieve what you need.

Regards.

Re: Monitoring Mode Count

MikeL — Mon, 22 Sep 2025 12:45:16 GMT

Much appreciated, I will try to get the DT techs to create us a tile we can close to our dashboards.

Would you also know why we cannot get any results when trying to do similar via Data Explorer > Adv Metric Explorer:
When we try this...

builtin:host.splitBy(monitoringMode):count

builtin:host.availability:splitBy(monitoringMode):count

after we type "builtin:", we only get Apps.*, but not host.*?

Re: Monitoring Mode Count

dannemca — Mon, 22 Sep 2025 17:51:07 GMT

You should have builtin:host.* metrics there. Have you tried to type builtin:host. and see if the list appear?

But, you will not get any metric that gives you the number of hosts by monitoring mode, since it is not a dimension for any metric.

https://docs.dynatrace.com/docs/analyze-explore-automate/metrics-classic/built-in-metrics

https://docs.dynatrace.com/docs/analyze-explore-automate/metrics/built-in-metrics-on-grail

Re: Monitoring Mode Count

MikeL — Mon, 22 Sep 2025 19:44:05 GMT

Thx, yes I have, just like the screenshot, if I keep typing.... builtin:host... I get blank...
Must be some permission somewhere?
But seems that might be a non-starter anyways per your URL. thx for that.

Re: Monitoring Mode Count

t_pawlak — Tue, 23 Sep 2025 07:18:13 GMT

Hi, try use this:
builtin:billing.full_stack_monitoring.usage_per_host:splitBy("dt.entity.host"):sort(value(auto,descending)):limit(20)

and builtin:billing.infrastructure_monitoring.usage_per_host:splitBy("dt.entity.host"):sort(value(auto,descending)):limit(20)

You will have list of hosts.
I don’t recall there being a metric in Data Explorer that lists only hosts by monitoring type. But you could create a custom metric, or use the API to pull the monitoring mode and push it into Dynatrace as a metric.

Re: Monitoring Mode Count

MikeL — Tue, 23 Sep 2025 13:18:20 GMT

thx Tomaxp,
Not sure what the "Usage Billing" data gets interpreted as, but I can indeed see those metrics, see the screen grab, are you are hinting at the unit cost (Full Stack technically cost more to run), anything over 5 = Full Stack & anything <2 = Infra Mode? if yes you might be onto something. 🙂

Re: Monitoring Mode Count

t_pawlak — Tue, 23 Sep 2025 14:04:43 GMT

That metric isn’t directly “host monitoring mode,” it’s the billing usage per host. What you see in the table is the consumption value for each host in each mode.
So the metric gives you the list of hosts and their usage values, which you can then interpret as the monitoring mode.

If you want a single tile showing just the count of hosts per mode, you’d need to either:

pull the host monitoring mode via the Dynatrace API, then push it back into Dynatrace as a metric for easy thresholding on a dashboard.

Re: Monitoring Mode Count

MikeL — Tue, 23 Sep 2025 14:20:08 GMT

Thx Tomaxp,
I think we have something we can use now, thx for that info..
Breaking it up in two (one for Count & one for the VM list, labeling what the Count 'should' be in the title, if those changes, it should be noticed.

To bad we can't set to a # like 147, and anything higher or lower than that 147 count would trigger the threshold to change to Red (can't change the < or > for the colors)...

But this is usable. much appreciated.