topic Re: DQL - Metric in dashboard to host in DQL

DQL - Metric in dashboard to host

Fast — Mon, 10 Feb 2025 22:49:54 GMT

Hi there.

I am new in DQL, and I need help to have and view metrics (CPU usage by host / Memory usage by host) in the dashboard, those hosts have an specific tag.

Therefore, I have been trying to get the metrics since days ago; however, I have not been able to do it. I was wondering if you know how to segment this by the tag and bring the metric.

I was checking many cases in Dynatrace Community and I tried those queries, and the queries work, but they mix the host.

Grateful if anyone can help on the this issue.

Thanks!

Re: DQL - Metric in dashboard to host

p_devulapalli — Tue, 11 Feb 2025 00:45:20 GMT

@Fast Please try similar to below

Re: DQL - Metric in dashboard to host

zaid-bashir — Tue, 11 Feb 2025 06:37:18 GMT

Hi @Fast
I am posting here a similar solution but by searching hostname, you can find the query below.

timeseries {avgCpu = avg(dt.host.cpu.usage),avgMem = avg(dt.host.memory.usage)}, by:{dt.entity.host}, interval:30m | fieldsRename id = dt.entity.host | join [fetch dt.entity.host | fieldsAdd entity.name],on: {id} | filter right.entity.name == $Search_Host

But You have to first create a variable for Search_Host which will take your hostname and checks the hostname that is equal to resulted hostname and filters the result. You can also use contains() here to get the results matching with Search_Host Variable.

You can modify the above query to get remaining metrics and add other variables to furthur extend the query.

If you need furthur help in this, reply.

Re: DQL - Metric in dashboard to host

Fin_Ubels — Tue, 11 Feb 2025 09:12:20 GMT

Just wanted to add on here another way to get entity names as a tip!

timeseries {avgCpu = avg(dt.host.cpu.usage),avgMem = avg(dt.host.memory.usage)}, by:{dt.entity.host}, interval:30m | fieldsAdd entity.name = entityName(dt.entity.host) | filter entity.name == $Search_Host

You can also use the entityAttr() to get a specific attribute removing the need to write out and do a full join/lookup and then cleanup any of the stuff that comes with that.