Skip to content

Part of my AI collaborative lab series for Microsoft Identity and Access Administration skill development and expertise pathway.

Notifications You must be signed in to change notification settings

Compcode1/lab-12-365-apps-id-objects

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 

Repository files navigation

Lab 12 — Microsoft 365 Applications as Entra ID Identity Objects

Understanding Resource-Based Conditional Access and Sign-In Logs


Lab Purpose

This lab establishes a foundational and exam-critical understanding of how Microsoft 365 applications actually interact with Microsoft Entra ID.

The goal is not to configure new controls, but to observe and correctly interpret:

  • How Microsoft 365 workloads appear as enterprise applications
  • How a single user sign-in produces multiple Entra ID sign-in events
  • How Conditional Access (CA) is evaluated per resource, not per “login”
  • Why sign-in logs look confusing unless this model is understood

This lab exists to eliminate one of the most common conceptual failures on the SC-300 exam.


What We Did (High-Level)

  1. Identified Microsoft 365 workloads as Enterprise Applications (Service Principals)
  2. Signed in as a user and accessed Microsoft 365 resources (Office portal / SharePoint)
  3. Examined Entra ID sign-in logs generated by that single user action
  4. Traced how multiple resources were evaluated independently by Entra ID
  5. Interpreted log results correctly without mislabeling success/failure

Core Technical Findings

1. Microsoft 365 Is Not a Single Application

“Microsoft 365” is not one app from Entra ID’s perspective.

It is a collection of independent enterprise applications, including (but not limited to):

  • Microsoft Graph
  • Office Home
  • My Apps (App Catalog)
  • SharePoint Online
  • Exchange-related services
  • Teams-related services

Each of these exists as a distinct service principal in the tenant.


2. One User Action Creates Multiple Sign-In Events

A single human action such as:

  • Signing into office.com
  • Opening SharePoint
  • Launching apps from My Apps

results in multiple Entra ID sign-in events, because:

  • Each downstream resource requests its own token
  • Each token request triggers its own Conditional Access evaluation

There is no such thing as a single “Microsoft 365 login” internally.


3. Conditional Access Is Evaluated Per Resource

Conditional Access policies are evaluated:

  • Per sign-in event
  • Per resource (enterprise application)

Not per user session. Not per browser window. Not per “login attempt.”

This explains why one user sign-in can show:

  • One event marked Interrupted
  • Followed by multiple Success events
  • Each tied to a different resource

This is expected behavior.


4. Why Sign-In Logs Look “Strange” at First

During the lab, we observed a sequence similar to:

  • Resource: Microsoft Graph → Status: Interrupted
  • Resource: Office Home → Status: Success
  • Resource: My Apps → Status: Success

This does not indicate an error.

It reflects the authentication flow:

  • Microsoft Graph is used early for identity and directory access
  • Office Home represents the Microsoft 365 landing workload
  • My Apps represents the application catalog experience

Each step is logged separately because each is a separate authorization decision.


5. “Success” and “Failure” Mean User Outcome, Not Policy Quality

In Entra ID sign-in logs:

  • Success = the user obtained access
  • Failure = the user did not obtain access

These labels do not mean:

  • “The policy worked”
  • “The policy failed”
  • “The configuration is wrong”

A Conditional Access policy can work perfectly and still produce a Failure if the policy is designed to block access.

This distinction is critical for both labs and the SC-300 exam.


Exam-Critical Takeaways (SC-300)

  • “All cloud apps” ≠ “Microsoft 365”
  • Microsoft 365 is a bundle, not a single enforcement target
  • Conditional Access is always resource-scoped
  • Sign-in logs reflect authorization outcomes, not intent
  • Multiple log entries for one user action are normal

About

Part of my AI collaborative lab series for Microsoft Identity and Access Administration skill development and expertise pathway.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published