CIS-DF · Domain 2 of 5

CIS-DF Domain 2: Ingest

Ingest is about getting trustworthy data into the CMDB: picking the right ingestion method, populating relationships, handling non-discoverable CIs, tracking compliance identifiers and keeping assets aligned with configuration items.

Exam weight estimate
19%
Official objectives estimate
7
Estimated questions estimate
~14
Suggested study time estimate
12–14 hrs

About 14 min read Reviewed October 2026

Overview

Domain 2 has seven blueprint objectives, all about how data enters and stays aligned in the CMDB. You need to know how relationships are populated automatically by Discovery and Service Mapping or manually with suggested relationships, which automation options exist for CMDB integration, and which ingestion method suits a source: ServiceNow Discovery with a MID Server, a Service Graph Connector from the ServiceNow Store, IntegrationHub ETL for a custom source, or import sets for simple loads.

The domain also tests upgrade-safe design, such as using certified connectors and IRE rather than custom transform scripts. Further objectives cover non-discoverable CIs and attributes like ownership and support group, security and regulatory compliance identifiers on CSDM objects, and the alignment between hardware assets in alm_asset and CIs through model categories and field mappings.

How it's tested

Ingest questions describe a data source, its constraints and a goal, then ask for the best method or configuration. Expect phrases about third-party tools, cloud accounts, spreadsheets, manual data, or asset records not matching CIs. The best answer usually routes data through IRE with the least custom work.

  • Pick the ingestion method

    A source is named, such as Microsoft Intune, an on-premises data center or an HR spreadsheet, and you choose Discovery, a connector, ETL or import sets.

  • Non-discoverable data

    Attributes like owner or support group are missing, and you pick a maintenance approach such as attestation, an authoritative import or CI owner updates.

  • Asset versus CI

    A hardware asset and its CI disagree or one is missing, and you identify model category or field mapping settings as the cause.

Reasoning it through

A stem says a company manages laptops in Microsoft Intune and wants them in the CMDB quickly without custom code. Options include Discovery with a MID Server, a scripted REST integration, a Service Graph Connector for Intune, and monthly import sets. The certified connector uses IntegrationHub ETL and IRE, is maintained by ServiceNow and fits the no-code goal, so it is the best choice.

Objectives 7

Official objectives from the ServiceNow outline, in outline order.

2.1 Identify how relationships are auto or manually populatedWhere CI relationships come from.
What it means
Relationships link CIs with a typed pair such as Runs on::Runs or Depends on::Used by. They are created automatically by Discovery, Service Mapping and connectors, or added manually by users.
What to understand
Discovery patterns build infrastructure relationships, such as an application running on a server, while Service Mapping and tag-based mapping build service-level dependencies. Connectors map relationships in their ETL definitions. For manual entry, suggested relationships defined in Class Manager limit users to sensible relationship types between two classes, which keeps manual data consistent.
How it shows up
A scenario asks how to make manual relationship entry consistent, expecting suggested relationships.
Common confusion
Assuming suggested relationships create links automatically rather than guiding manual creation.
2.2 Identify the automation opportunities that are available to automate CMDB integrationAutomation options that keep the CMDB current.
What it means
ServiceNow offers several automation layers for CMDB data: Discovery schedules, Service Graph Connectors, IntegrationHub ETL, scheduled imports and Flow Designer or Workflow Studio flows that call IRE.
What to understand
Scheduling keeps data current without manual effort, and connectors include their own scheduled data sources. Automation should still pass through IRE so identification and reconciliation apply. Health jobs and Data Manager policies automate maintenance after data arrives, which links ingestion to governance.
How it shows up
You are asked how to keep a third-party source synchronised daily with minimal effort.
Common confusion
Choosing manual exports and imports because they look cheaper in the short term.
2.3 Identify which data ingestion method to useMatching sources to Discovery, connectors, ETL or import sets.
What it means
The main ingestion methods are ServiceNow Discovery (agentless, MID Server based), Agent Client Collector, Service Graph Connectors, IntegrationHub ETL for custom sources, import sets with transform maps, and manual entry.
What to understand
Use Discovery for infrastructure the instance can reach through a MID Server. Use a Service Graph Connector when a certified one exists for the tool, such as Intune, Jamf, AWS or Azure. Use IntegrationHub ETL to map a custom source to CMDB classes through IRE. Use plain import sets only for simple, low-frequency loads, and keep manual entry for data no tool can provide.
How it shows up
A list of sources is given and you match each to the most suitable ingestion method.
Common confusion
Using import sets with transform scripts for sources that already have a certified connector.
2.4 Identify how to minimize the risk of technical debt and ensure upgradeabilityDesign choices that keep the CMDB upgrade-safe.
What it means
Technical debt in the CMDB comes from custom tables, custom fields that duplicate out-of-box ones, modified base patterns, and scripts that write directly to CI tables.
What to understand
Prefer out-of-box classes and attributes, Store connectors maintained by ServiceNow, and IRE for every write. Copy a pattern before changing it so upgrades can still update the original. Document each customisation and its owner, and check whether new releases add the field you created.
How it shows up
Options include direct table writes, custom transform scripts and certified connectors, and you pick the upgrade-safe one.
Common confusion
Believing that a working custom script is free of debt because it causes no errors today.
2.5 Identify how to populate and maintain non-discoverable/manual CIs or non-discoverable attributesKeeping manual data accurate.
What it means
Some CIs, such as racks or legacy devices, and some attributes, such as owned by, support group or business criticality, cannot be discovered and must come from people or authoritative systems.
What to understand
Load them from authoritative sources through IRE where possible, assign CI owners, and use Data Manager attestation or certification policies so owners confirm the CI still exists and that key values are right. Manual CI creation through the CMDB Workspace or forms should still follow class and relationship guidance.
How it shows up
A stem asks how to keep support group values accurate for CIs that Discovery cannot populate.
Common confusion
Expecting Discovery to fill organisational fields such as owner or support group.
2.6 Identify how to track security and regulatory compliance identifiers against CSDM objectsRecording compliance context on services and applications.
What it means
Security and regulatory data, such as data classification, regulated scope or compliance identifiers, should be stored against CSDM objects like business applications and service instances rather than scattered across individual servers.
What to understand
Holding this data on CSDM objects lets security and risk teams see which services handle regulated data and which infrastructure supports them through relationships. Risk and compliance applications and vulnerability response can then use the same records. Keep the identifiers in out-of-box fields where they exist instead of adding parallel custom fields.
How it shows up
A scenario asks where to record that an application processes payment card data so all supporting CIs inherit that context.
Common confusion
Recording compliance tags only on servers, which loses the service context when infrastructure changes.
2.7 Identify how Asset and CI alignThe link between alm_asset records and CIs.
What it means
Hardware assets track financial and contractual data in alm_asset, while CIs track operational data in cmdb_ci. Model categories define which asset class pairs with which CI class.
What to understand
When a model category names both an asset class and a CI class, creating one record creates its counterpart. Asset-CI field mappings keep shared fields such as serial number, assigned to and location synchronised, and install status and asset state mappings keep lifecycle values aligned. Software and consumables may have assets without CIs.
How it shows up
A question asks why discovered servers have no matching asset records, expecting a model category without an asset class.
Common confusion
Treating the asset and the CI as the same record, or assuming every CI must have an asset.

High-yield concepts

The terms, rules and formulas this domain turns into questions most often.

  • Service Graph Connector

    A certified integration from the ServiceNow Store that loads third-party data into the CMDB through IntegrationHub ETL and IRE.

    Remember
    Use it whenever a connector exists for the source tool.
    Tested as
    The usual answer for SCCM, Intune, Jamf and cloud provider data.
    Trap
    Building a custom REST integration when a connector already exists.

  • IntegrationHub ETL

    A guided tool that maps a source's import data to CMDB classes and relationships and sends the result through IRE.

    Remember
    Service Graph Connectors are built with it, and you can use it for custom sources.
    Tested as
    Expected when no connector exists but the data is recurring.
    Trap
    Using a transform map script that writes directly to CI tables.

  • ServiceNow Discovery

    Agentless discovery of infrastructure through a MID Server using probes and patterns.

    Remember
    It creates CIs and infrastructure relationships horizontally.
    Tested as
    The default for on-premises servers and network devices.
    Trap
    Expecting it to discover owner or support group values.

  • MID Server

    A Java application inside the customer network that runs discovery and integration work for the instance.

    Remember
    Discovery cannot reach internal devices without it.
    Tested as
    Questions on why Discovery fails for an internal subnet may involve the MID Server.
    Trap
    Thinking the instance scans internal networks directly from the cloud.

  • Service Mapping

    Top-down discovery of the CIs and connections that support a specific service.

    Remember
    It builds application service maps from an entry point.
    Tested as
    Appears when service-level relationships are needed.
    Trap
    Confusing it with horizontal Discovery.

  • Import sets

    Staging tables loaded from files or sources and moved to targets by transform maps.

    Remember
    Suitable for simple loads; CMDB targets should still use IRE.
    Tested as
    Often the distractor in ingestion-method questions.
    Trap
    Picking them for recurring, high-volume third-party data.

  • Model category

    Defines the pairing between an asset class and a CI class for a type of model.

    Remember
    Missing either side stops automatic creation of the counterpart.
    Tested as
    The key setting in asset and CI alignment questions.
    Trap
    Looking for a business rule instead of checking the model category.

  • Asset-CI field mapping

    Configuration that synchronises shared fields between an asset and its CI.

    Remember
    Changes on either record flow to the other for mapped fields.
    Tested as
    Explains why location or assigned to stays in step.
    Trap
    Assuming all fields synchronise by default.

  • Non-discoverable attributes

    Fields such as owned by, support group and business criticality that no discovery tool can set.

    Remember
    Maintain them through authoritative imports, owners and attestation.
    Tested as
    Common stem in Ingest and Govern questions.
    Trap
    Expecting a discovery schedule to fill them.

  • Technical debt

    Customisation that raises upgrade and maintenance effort.

    Remember
    Out-of-box classes, connectors and IRE reduce it.
    Tested as
    Many Ingest answers are chosen by minimising it.
    Trap
    Assuming working code has no future cost.

Side by side

Concepts candidates mix up, compared on the point that decides the answer.

Ingestion methods

ConceptMeaningKey differenceCommon trap
Discovery Agentless scanning through a MID Server Best for reachable on-premises and cloud infrastructure Expecting organisational data from scans
Service Graph Connector Certified Store integration for a named tool Least effort and upgrade-safe for supported tools Ignoring it in favour of custom code
IntegrationHub ETL Guided mapping of custom sources through IRE For recurring sources without a connector Confusing it with plain transform maps
Import sets Staging and transform maps For simple, occasional loads Using it for daily third-party feeds

Asset vs CI

ConceptMeaningKey differenceCommon trap
Asset (alm_asset) Financial, contractual and ownership record Tracks cost, contracts and asset state Using it for operational dependencies
CI (cmdb_ci) Operational record used by service processes Tracks relationships, status and support Using it for depreciation or contracts
Model category Pairs asset and CI classes Drives automatic counterpart creation Overlooking it when records do not pair

Exam traps

Each trap is an answer that looks right. Learn the clue that should change your mind.

  1. Custom code for a common tool

    What candidates pick
    A scripted REST integration to load Jamf devices.
    Why it looks right
    It offers full control over mapping.
    Why it's wrong
    A Service Graph Connector for Jamf already exists, uses IRE and is maintained by ServiceNow across upgrades.

    Clue to watch for The tool named in the stem is a mainstream product.

  2. Discovery for organisational data

    What candidates pick
    Configure a Discovery pattern to populate support group.
    Why it looks right
    Discovery seems to fill every CI field.
    Why it's wrong
    Support group is an organisational fact; it comes from authoritative systems, owners or attestation, not from device scans.

    Clue to watch for Fields describing people, teams or business value.

  3. Import sets for everything

    What candidates pick
    Load cloud resources weekly with spreadsheet imports.
    Why it looks right
    Import sets are familiar and easy.
    Why it's wrong
    Manual, infrequent loads go stale quickly and may bypass IRE; connectors or Discovery keep cloud data current.

    Clue to watch for Frequent change or a cloud provider is mentioned.

  4. Modify the base pattern

    What candidates pick
    Edit the out-of-box Discovery pattern to collect one more field.
    Why it looks right
    It is the fastest change.
    Why it's wrong
    Editing the base pattern blocks future upgrades to it; copy and extend instead.

    Clue to watch for The stem mentions upgradeability or maintenance.

  5. Asset sync by script

    What candidates pick
    A business rule that copies location from asset to CI.
    Why it looks right
    Scripts seem the obvious way to synchronise fields.
    Why it's wrong
    Asset-CI field mappings already do this as configuration without custom code.

    Clue to watch for Two paired records with one field out of step.

  6. Compliance on servers only

    What candidates pick
    Tag every server that processes regulated data.
    Why it looks right
    Servers are where the data physically sits.
    Why it's wrong
    Server tags break when infrastructure changes; compliance context belongs on CSDM objects related to that infrastructure.

    Clue to watch for A regulation or data classification applies to an application.

Answer framework

Use these steps to choose the right approach for an Ingest scenario.

  1. Identify the source type: reachable infrastructure, a third-party tool, a custom recurring feed, a one-off file or human knowledge; the type usually narrows the answer to one ingestion path.
  2. If a certified Service Graph Connector exists for the tool, choose it; otherwise prefer Discovery for reachable infrastructure and IntegrationHub ETL for custom feeds.
  3. Reject options that write to CI tables without IRE or that modify base patterns and out-of-box tables directly.
  4. For non-discoverable data, choose authoritative imports, CI ownership and Data Manager attestation or certification rather than discovery settings.
  5. For assets and CIs that do not pair or synchronise, check model categories and asset-CI field mappings before considering scripts, because a missing model category stops the asset and CI pairing.

Practice question

A retailer stores manufacturing equipment details in an in-house database that no Store connector supports. The data changes weekly and must create and update CIs without duplicates. Which approach best meets the requirement while limiting technical debt?

Want more like this? The CIS-DF practice test has questions for every domain.

Checklist

0 of 12

Tick each one you can do without notes. Your ticks stay in this browser.

Study strategy

Start by drawing a simple map with sources on the left (data center, cloud accounts, endpoint tools, HR or finance systems, spreadsheets) and ingestion methods on the right. For each source, write the preferred method and one reason. That map answers most method-selection questions.

Next, read the Store listing and documentation for two Service Graph Connectors, such as Microsoft Intune and AWS, and note that both rely on IntegrationHub ETL and IRE. In a Personal Developer Instance, open IntegrationHub ETL and step through creating a mapping for a small CSV so you understand classes, relationships and the IRE preview.

Then spend a session on asset and CI alignment. Open several model categories, check their asset and CI classes, and create a hardware asset to see the CI appear. Review the asset-CI field mappings.

The objective candidates underrate is technical debt: list five common customisations and the upgrade-safe alternative for each. Finish with mixed scenario practice that combines Ingest with Govern items.

FAQ

How many Ingest questions should I expect?

Ingest is weighted at 19% of the 75-question exam, which is roughly fourteen questions as a planning estimate rather than an official count.

Is ServiceNow Discovery configuration tested in depth?

Not at the level of the Discovery specialist exam. You need to know when Discovery is the right method and what the MID Server does, not detailed probe or pattern troubleshooting.

When is an import set acceptable for CMDB data?

For simple, occasional loads where no connector or recurring integration is justified, and ideally with IRE handling the CMDB writes.

What is the difference between a Service Graph Connector and IntegrationHub ETL?

A connector is a ready-made, certified integration for a specific product. IntegrationHub ETL is the mapping tool connectors are built with, and you use it directly for sources that have no connector.

Does every hardware CI need an asset?

No. Model categories decide pairing, and some CI classes, such as virtual or logical items, are often tracked without a matching asset.

Why does Ingest mention compliance identifiers?

Security and risk processes use the CMDB, so the exam checks that you record regulatory context on CSDM objects where it can be related to supporting infrastructure.

Which ingestion topics overlap with other domains?

IRE links Ingest to Configuration, non-discoverable data links to Data Manager in Govern, and compliance identifiers link to CSDM Fundamentals.

Other CRM & Business Applications certifications

Where the same subject shows up in other exams. Cards marked "Same subject" open the matching domain.

Test your Domain 2 knowledge

You have worked through ingestion methods, relationships, manual data and asset alignment. Check that you can choose the right method quickly in timed scenarios.

Take the CIS-DF practice test Check your CIS-DF readiness

Keep going