Is it possible to have telegraf ONLY collect vm/host metrics from a single cluster in a vCenter?
Per the documentation , this does not work:
We can extend this to looking at a cluster level:
/DC0/host/Cluster1/*/hadoop*. This selects any VM matching “hadoop*” on any host in Cluster1.
this does not work:
Can you share what you tried?
Not sure what I may be doing wrong, I’ve tried:
cluster_include = [ "/DC01/host/clu01/**"]
cluster_include = [ "/DC02/host/clu02/*/*"]
cluster_include = [ "/*/host/clu01/*/*"]
cluster_include = [ "/**/host/clu01/**"]
cluster_include = [ "/*/*/host/clu01/**"]
cluster_include = [ "/*/*/host/clu01/*/*"]
Is that your entire vsphere configuration?
I would suggest looking at the
vsphere readme where it calls out a variety of
*_exclude options. If some of these options are not included, there is a default set of metrics that are collected, which may be causing you to collect more than you think.
Also take a look at the
inventory-paths of the readme.
If those sections do not help resolve your issue can you be sure to include your full vpshere config + some example output?