I've been collecting data from over +/-400 with extension snmp 2.0
I configured the yaml only to collect the metric and nothing more.
Now i have 2 questions about Anomaly detetion - metric events:
2. Having this problems i tried to use metric key and not metric selector. Now the problem is that i dont have alert on missing data. My question is how can i have this functionality? Scripting 😥
Thanks for your time 🤔
Solved! Go to Solution.
we have similar problem on our dynatrace managed cluster: metric with other 1000 dimensions.
we had conversation with support ppl and they said the 1000 dimension limit is configurable. Same for retention of dimensions.
however, are you sure you really need all 200k dimensions to be monitored?
You write, you have 400+ objects. Where all 200k dims come from? Maybe you can reduce unnecessary dims or split them in groups.
I understand its not a solution, bur sometimes to redefine problem helps too.
Thanks for the answer @Ingrida,
My metric selector was like this: metric_key:splitby(snmp:custom_device):avg:rollup(avg,10m)
The problem is using the rollup, i dont know why but increase a lot the number of dimensions....
Hi @rdinis ,
Not sure my answer was an solution. Maybe something obviosu, maybe you already did it, but to answer "...i dont know why but increase a lot the number of dimensions" you may tray to display this metric in data explorer table view (make sure you remove limit(20) before). In this case you could see what dimensions are you are "collecting". For me sounds like unique id's...