ServiceNow CIS-DF Exam Questions

The Foundation tab covers the shared reference and foundational data upon which the remaining CSDM domains depend. This includes organizational and reference structures such as companies, business units, locations, users, groups, product models, and related foundational records.

Product models establish standardized product definitions. Locations provide reusable geographical and physical context for CIs and services. Business units establish organizational context and accountability. If these records are duplicated, incomplete, or modeled inconsistently, later service modeling becomes unreliable.

The guide distinguishes Foundation from Design and Planning, Build and Integration, Service Consumption, and Service Delivery. It identifies Foundation as the area containing shared reference data such as companies, locations, and users.

Crawl, Walk, Run, and Fly are progressive implementation stages. They do not replace the Foundation dashboard category that directly assesses these shared reference-data domains.

=========

In the CMDB class hierarchy, Server is a generic parent class, while Linux Server and Windows Server are more specific child classes. When ServiceNow Discovery detects sufficient evidence (such as OS signatures) to move a CI from a generic class to a more specialized one, this action is called a Class Upgrade.

A Class Upgrade occurs when a CI is reclassified down the hierarchy into a more specific subclass, enriching the record with additional attributes, behaviors, and discovery patterns appropriate to that class. This is a standard and expected behavior in mature CMDB implementations and aligns with Data Foundations best practices.

A Class Switch would imply lateral movement between unrelated classes, which is not what happens here. A Class Downgrade would move a CI from a specific class back to a more generic one, typically when discovery confidence is reduced---not the case in this scenario.

By performing class upgrades automatically, Discovery improves CMDB accuracy, reporting precision, and service mapping quality without manual intervention.

Therefore, the correct answer is C -- Class Upgrade.

When using Import Sets and Transform Maps to ingest data into the CMDB, it is critical that records are processed through the Identification and Reconciliation Engine (IRE) to prevent duplicates and enforce source precedence. In ServiceNow, this is achieved by invoking the IRE after the transform logic has completed.

The onAfter transform script is the correct place to call the IRE API. At this stage, the transformed data has already been mapped and prepared, allowing the IRE to correctly identify whether a CI already exists and reconcile updates according to defined rules.

The onBefore and onStart scripts execute too early---before data mapping is complete---making them unsuitable for IRE processing. The onComplete script runs after the entire import job finishes and is not intended for per-record CI identification and reconciliation.

Because Import Sets can bypass IRE if not configured correctly, using an onAfter script is a critical Data Foundations safeguard when this ingestion method is chosen.

Therefore, the correct answer is C -- onAfter.

Comprehensive and Detailed Explanation (200--300 words)

Within ServiceNow Data Foundations, CMDB Data Manager provides multiple policy types to support governance, data quality, and lifecycle management of configuration items (CIs). The scenario described---confirming that servers recorded in the CMDB physically exist in the data center---is a classic example of existence validation and ownership confirmation, which is exactly the purpose of an Attestation policy.

An Attestation policy is designed to request a human validation from a responsible individual or group (such as a data center manager, platform owner, or infrastructure team). The policy generates attestation tasks that require reviewers to explicitly confirm whether a CI is valid, accurate, and still exists. This aligns directly with CMDB governance best practices and ITIL 4 Service Configuration Management, where periodic verification ensures trust in the CMDB as a system of record.

The other policy types do not meet this requirement:

Certification is typically used to validate compliance with defined data standards (e.g., mandatory fields populated), not physical existence.

Delete, Archive, and Retire are lifecycle actions, used after a CI has already been identified as obsolete or no longer required.

None of these options involve human confirmation of real-world existence.

From a CSDM and Data Foundations perspective, attestation supports:

CMDB accuracy and credibility

Audit and regulatory compliance (especially critical in financial services)

Clear accountability for CI ownership and validation

Therefore, when the goal is to confirm that servers actually exist, the correct and fully aligned CMDB Data Manager policy type is Attestation (E).

In Data Foundations (CMDB and CSDM), relationship modeling must reflect real operational dependency so that incident triage, change impact analysis, and service visibility remain accurate. When an Application is hosted on a Server, the standard hosting-style relationship used in CMDB relationship governance is expressed as ''Runs on::Runs''. This pairing represents the two directional descriptors of the same relationship type: from the application perspective it runs on the server, and from the server perspective it runs the application.

This matters because CMDB relationships are used by downstream capabilities (for example, dependency views, impact calculations, and governance rules). Using the correct out-of-box relationship descriptor pair ensures consistent reporting and prevents confusion when teams traverse relationships ''upstream'' and ''downstream.'' In addition, relationship governance rules and inheritance are commonly built around standard relationship types; using the correct ''Runs on::Runs'' semantics supports validation across subclasses (for example, specific application and server subclasses) without requiring custom relationship definitions.

The other options are either reversed (''Runs::Runs On''), represent different semantics (''Uses::Used by''), or do not align with the typical hosting relationship naming used for application-to-server hosting dependencies. Therefore, the correct relationship expression between an Application and a Server is Application > Runs on::Runs > Server.