Question

The transform map "CsvComputerCustomFieldTransform" in the Snow Software CMDB Integration application within ServiceNow only has 1 coalesce and that is off of the Snow computer Id field. However, when we process our import, Snow is coalescing off of Name and/or Serial Number. Our Identification and Reconciliation in ServiceNow looks at Correlation ID, then Serial Number, then Name.  Why is Snow coalescing off of Name and/or Serial number for the ServiceNow integration?
 

Answer

A ServiceNow customer may have any number of CI classes in use for classifying their computers, including custom CI classes.

E.g. some would classify a laptop as "Personal Computer", others would not, and others have created custom CI classes to match their company processes. Some would want to identify their servers based on serial numbers, while others would use FQDN, etc.

The identification rules for each class is defined in the CI Class Manager editor:

When the ServiceNow transform process is running, it expects to evaluate the payload against the rules of a specific CI class in order to determine the identity. This means you need to pre-determine the class you want to evaluate a computer against when you submit the payload to the Identification and Reconciliation engine for identification.

In other words, if you want to match a computer against the Linux CI Class Identification rules, you need to tell the Identification and Reconciliation that your data payload is destined for the Linux CI Class (and its identification rules).

This pre-classifying step is performed in the CI Class Mapping editor, where Snow provides a default set of rules but which should be adapted/added to if required.

When the transform runs, there is an OnBefore script that uses this mapping table to direct the payload to the correct CI Class (and its identification rules), and the IRE engine returns a matching SysID if the computer is found.

This is all according to best practices from ServiceNow, and is something we are evaluated on each time we go through the certification process.

Below is a summary diagram to show how this works. Note that the process is different for the Product Model data - the IRE process only relates to the CMDB computers.

Note, the IRE process is only used for identifying the computer object. Any subsequent transforms, such as Network, Disk, CustomFields, etc, use the Snow Computer ID to link the data back to the correct computer.