topic Re: Number of data points exceeded the limit of 20000000 in Dynatrace API

Number of data points exceeded the limit of 20000000

AlanFoley — Fri, 15 Oct 2021 09:08:36 GMT

Hi all

I am trying to extract some api/v2 metrics via the API , I also apply filtering by tags,osType and mzName to reduce output however I am still getting the following message:

"Number of data points exceeded the limit of 20000000" -

The .csv that is returned contains only 881 rows of data - see sample of a single row below

"builtin:host.disk.usedPct:names,Host001,HOST-FE7A6A123B192776,D:\,DISK-1E2CC761ED90F1C,2021-10-01 00:00:00,54.4775917176209"

How is it possible to get to 20m data points with only 881 rows and with 7 columns of data ? It just seems impossible to get to 20m data points

I am on version 1.226.116.20211006-102024

I would be keen to understand how the data points are calculated based on the sample above

Any ideas,workarounds or suggestions welcome

Apprecaite the help

Thanks

Alan

Re: Number of data points exceeded the limit of 20000000

david_lopes — Wed, 27 Oct 2021 12:44:24 GMT

You have 1 datapoint per host per disk per minute

Let's say you are looking at 2 weeks of data

And assuming you have 100 hosts, every host has 3 disks

100 hosts * 3 disks * 60 datapoints/hour * 336 hours/week = 6,048,000 datapoints

As you can see you can quickly get millions of datapoints, the error means that the limit has been reached, whatever response you got won't include the millions of datapoints.

My suggestion: have your script make one request per host (or a couple of hosts) at a time, or use Inf resolution depending on your use case

Re: Number of data points exceeded the limit of 20000000

AlanFoley — Wed, 27 Oct 2021 12:57:32 GMT

Hi David

Appreciate the detailed response - I have set my collections for every 4 hours and for those metrics which are a fit I set resolution-Inf - This way I have been able to avoid the data limits and have useful data

Thanks a mil

Alan