Create a CI table or utilize an existing table

[Daiki Shinohara]

Hello everyone.

I would like to add a new CI.

However, there is no table in the existing cmdb_ci table that is suitable for the CI I want to add.

In this case, should I use the CI class manager to find a suitable hierarchy and create a new CI table?
(Is there any possibility that creating a new CI table will cause problems?)

Or should I customize the existing CI table and rebuild it?

How would you handle this?
Any best practices or success stories would be appreciated.

Thank you in advance.

Translated with www.DeepL.com/Translator (free version)

[Tom Sienkiewicz · Accepted solution]

Hi,

there will be absolutely no problems by creating a new CI Class in the Class Manager 🙂

Please just make sure to add this child class under the most suitable parent. Note that there are no licensing implications, CMDB tables can be freely extended.

Having dedicated classes for specific CIs can be very important, for example during Service Mapping (if we're thinking about Applications).

That being said, sometimes if you have a fitting class, but some fields are missing, you can just add the missing attributes also in Class Manager. But this should not be heavy customization of table, just adding of missing fields, does not sound like your case?

[Daniel Needles1]

Of course, proliferation of Classes should be resisted under the KISS principal.  That said, if it still makes sense, you can use this as an example of how to extend the Network class: https://www.youtube.com/watch?v=6ULRbpGOSB4

Of course you won't be able to reference the CIs in an Incident until you Navigate to cmdb_class_info.list and
-Create New Class
-Click drop down arrow on Class and select Item
-From cmdb_class_info.list set Principal Class to True