AdvisorAdvisor(Ignore all the errors and dead devices - this is intentional to help us build out dashboards and alert rules!!)
If it helps anyone visualise, this is what we’re working with for our structure. The very top _All Devices one is where every resource gets added. Then the _Internal group sub dynamic groups have no constraint for customers so bring in every resource matching the dynamic query (e.g. “hasCategory(“Windows_DNS”)).
Then we have dashboard groups matching the resource groups and are constrained to that resource group, and again for the _Internal dashboard group there is no customer constraint so on those dashboards we see everything across all customers together.
(Our customers will have logins but to obviously only their resource and dashboard groups.)
Then the only difference for the customer sub dynamic groups is they have the additional auto.organisation = “XYZ” constraint.
_Unknown just has query where auto.organisation property doesn’t exist.
At least so far this makes adding a new customers really easy as we just create the structure and then clone all the dashboards.
The Infrastructure group I will probably split to have Applications as well, and then put SQL for example in that instead. Not sure yet.
And, actually drawing and writing this out has made me think of something else; templating resource and dashboard group structures.
