As organizations move towards cloud monitoring solutions like AWS, IKS, AKS Azure and so on. How are you handling hosts being spun down per defined rules in the cloud platform.
For Example. Lets say Your Azure Platform is set to have 5 PaaS hosts supporting your EasyTravel Platform. One of these PaaS hosts becomes unhealthy, as a result azure spins up a replacement. Once that replacement Host is running and taking load, Azure spins down that other host that had become unhealthy. When Azure spins down the host it is not detected as graceful by Dynatrace, rather its more abrupt, which is why we get an alert on it.
So how are other organizations handling this noise? Are you using the solution of setting a tag that a host is slated for Decommission/spin down? How is that working for you? With Azure you can set a Min and Max range the the amount of instances. Since we are able to connect to Azure, why cant Dynatrace capture that value and apply it.
I just don't see how setting a tag is a viable solution, as at any given time Azure automatically spins up and spins down devices per load and health.
Please share your solutions that you have implemented to mitigate this issue.