Dynamic metrics with Elasticsearch for a virtualized environnement

Support Elasticsearch
#1

Hi,

I’m monitoring a virtual environnement and I’m trying to put into grafana the performance data we get from our VMs. Since the VMs change quite often and that they don’t all have the same number of CPUs, is there a way I could dynamically create graphs for this in Grafana, with the right metrics for each VMs?

I’ve been searching the forum and I haven’t found anything conclusive.

Thanks in advance.

#2

If anyone ever asks himself, it’s actually pretty easy to do, when you have the right document configuration.

I was mapping every metrics in a document per VM and time. But, if you create a document for every metric/VM/time, it becomes pretty easy to template your metrics and your VM names a graph for cpu and for memory for each VM.

Sorry if I disturbed anyone with my newbie question!