How to exclude some specific VMs from discovery?

[Aleksey4]

Hi,

we use Databricks in Azure and it creates a large number of the VM's and corresponding resources (disks, NIC's etc) every day, that live for a couple of hours only and then get terminated and deleted from azure, that generates a lot of "noise" in CMDB - in few months it grew from 200k to 2.5mln (VM's, disks, VNIC, endpoint etc)

We created autoflush ( sys_auto_flush) rules to get rid of old records in different tables, specifically used "databricks" in object ID to identify and delete the old VM.

It would be easier if we could exclude databricks objects from Discovery at the very beginning rather than cleaning it afterwards - is there a way to do that? Unfortunately, databrics are in the same subscriptions as other resources, but have a specific object ID and resource groups.

[ServiceNow Tec2 · Accepted solution]

This has been resolved by ServiceNow Technical Support. Please refer to KB0755878 (HI login required) for more information.

[Jonah Piascik]

The link you included is an article for using Idea Management for customer enhancement requests, which does not provide a solution for the initial question being asked.  Did you have a solution to the original post in another KB article perhaps?

[Aleksey4]

I am not sure what has happened to the discussion and how KB became an asnwer, there were some replies from the product manager. 

Eventually, we managed to achieve (almost) what we needed by a combination of:

1) suppressing Azure alerts related to specific VM's to disable Alert driven discovery

2) those that come through via regular scheduled discovery, got instantly deleted by table clean-up (sys_auto_flush)

See screenshot, hope you can use something like that as well.

[NickHopper78]

Were there assets created? If so, were they left orphaned when the CIs were deleted through the auto flush?