Skip to content
Home » Articles » Entra ID Conditional Access Lockout Playbook

Entra ID Conditional Access Lockout Playbook

Entra ID Conditional Access policies are the bedrock of a Zero Trust identity posture, but misconfigurations can inadvertently lock out administrators and users alike. A poorly designed policy, especially one impacting all users or all administrators, can lead to a full tenant lockout, bringing business operations to a standstill. This playbook details how to prevent, diagnose, and remediate Entra ID Conditional Access lockouts, ensuring your administrative access remains resilient.

Understanding Conditional Access Policy Interaction

Conditional Access policies evaluate user, device, application, and location signals to enforce access controls. Their strength lies in their granular control, but complexity can introduce unexpected interactions. Each policy rule is processed, and if multiple policies apply, the most restrictive outcome is enforced. This “AND” logic can quickly lead to an unintended lockout if, for example, a policy requires MFA from a trusted location, but another policy blocks all access from that same trusted location for a specific application.

The core components of a Conditional Access policy include:

  • Assignments: Who (users, groups), what (cloud apps), where (locations), and how (client apps, device platforms).
  • Access Controls: Grant (block access, require MFA, require compliant device) or Session (sign-in frequency, persistent browser session).

A critical step in preventing lockouts is to always exclude emergency access accounts from all Conditional Access policies.

⚡ Executive Operational Threat & Outage Index

Active Lockout Conditions:

  • High: Misconfigured Conditional Access policies (e.g., “All Users” + “All Cloud Apps” + “Block Access”) without exclusions.
  • Medium: Conflicting policies requiring different MFA methods or device states for the same user/app combination.
  • Low: Single policy affecting a small user group, easily remediated with a targeted exclusion or policy disable.

Primary MITRE ATT&CK Technique:

  • Defense Evasion (TA0005): Attackers often target misconfigured Conditional Access policies to bypass security controls.
  • Initial Access (TA0001): Lack of robust Conditional Access can facilitate initial breach attempts.

CISA KEV Catalog Additions (Past 7 Days):

  • No direct Conditional Access misconfiguration CVEs were added to the CISA KEV catalog in the past 7 days. However, vulnerabilities in identity providers or MFA bypass techniques (e.g., CVE-2024-38077 for a theoretical MFA bypass) would necessitate strong Conditional Access to mitigate.

Emergency Access Accounts: Your Lifeline

Before deploying any Conditional Access policies, establish and secure at least two “emergency access” or “break-glass” accounts. These accounts must be cloud-only, highly privileged (e.g., Global Administrator), and permanently excluded from all Conditional Access policies.

Creating Emergency Access Accounts:

  1. Create New User Accounts:
    • Navigate to the Entra ID admin center.
    • Go to Users > All users > New user > Create new user.
    • Name them clearly (e.g., breakglassadmin01@yourdomain.com).
    • Set a strong, complex password and store it securely (e.g., in a physical safe, split knowledge across executives).
    • Do not assign MFA to these accounts initially, or ensure it is the most resilient form (e.g., FIDO2 keys).
  2. Assign Global Administrator Role:
    • For each emergency account, go to Assigned roles > Add assignments.
    • Select Global Administrator.
  3. Exclude from Conditional Access:
    • In every Conditional Access policy you create, navigate to Users > Exclude and add these emergency accounts. This is non-negotiable.

Diagnosing a Conditional Access Lockout

When a lockout occurs, the first step is to confirm it is indeed a Conditional Access policy causing the issue.

1. Check Entra ID Sign-in Logs:

This is the most crucial diagnostic tool.
* Log in with an unaffected administrator account (ideally an emergency access account).
* Navigate to Entra ID > Monitoring & health > Sign-in logs.
* Filter by the affected user, application, or status (“Failure”).
* Look for entries with Conditional Access status indicating “Failure” or “Not applied (Policy not enabled)”.
* Click on a failed sign-in attempt and examine the Conditional Access tab. It will list the policies that were applied and why access was blocked.

2. Identify the Conflicting Policy:

The sign-in logs will pinpoint the exact policy (or policies) that blocked access. Common culprits include:
* Location-based policies: Requiring access from trusted IPs when the user is outside that range.
* Device compliance policies: Requiring a compliant device when the device is not enrolled in Intune or fails compliance checks.
* MFA policies: Requiring MFA when the user cannot complete it (e.g., lost phone, misconfigured MFA method).
* Client app policies: Blocking older authentication protocols or specific client apps.

Remediation Playbook: Resolving Lockouts

Once the offending policy is identified, immediate action is required.

Step 1: Disable the Problematic Policy (Emergency):

If you are locked out and have access to an emergency account:

  1. Navigate to Entra ID > Protection > Conditional Access > Policies.
  2. Locate the policy identified in the sign-in logs.
  3. Change its state from “On” to “Off” or “Report-only” (if you need to observe impact before full disablement).
  4. Save the change.

This should immediately restore access for affected users.

Step 2: Audit and Refine the Policy:

After restoring access, do not simply re-enable the policy without a thorough audit.

  1. Review Assignments:
    • Are “All users” or “All guest and external users” included when only a subset is intended?
    • Are your emergency access accounts explicitly excluded?
    • Are “All cloud apps” targeted when only specific apps need protection?
  2. Review Conditions:
    • Are “Locations” correctly defined? Have your trusted IPs changed?
    • Are “Device platforms” or “Client apps” too restrictive for your user base?
    • Are “Device state” requirements realistic for your managed devices?
  3. Review Access Controls:
    • Is “Block access” truly the desired outcome for the given conditions?
    • If “Require MFA” is set, are users properly enrolled and configured for MFA?

Step 3: Implement Report-Only Mode for Safe Deployment

The “Report-only” mode is a game-changer for safe Conditional Access deployment. It allows you to simulate the impact of a policy without enforcing it.

  1. When creating or modifying a policy, set its state to “Report-only”.
  2. Deploy the policy.
  3. Monitor the Entra ID Sign-in logs and specifically the Report-only tab within the Conditional Access details. This shows what would have happened if the policy were enforced.
  4. Adjust the policy based on Report-only findings until it behaves as expected.
  5. Only then, switch the policy state to “On”.

IT Admin & SecOps Action Playbook (Before 12:00 PM)

This playbook assumes a lockout scenario where an emergency access account is available.

Action 1: Verify Emergency Access Accounts

Ensure your emergency access accounts are fully functional and excluded from all Conditional Access policies.

# Verify emergency access account exclusion from all Conditional Access policies
# Requires Azure AD PowerShell module (Install-Module AzureAD)

Connect-AzureAD

# Replace with your emergency access account UPN
$emergencyAccountUPN = "breakglassadmin01@yourdomain.com"
$emergencyUser = Get-AzureADUser -Filter "userPrincipalName eq '$emergencyAccountUPN'"

if ($emergencyUser) {
Write-Host "Checking Conditional Access policies for exclusions of $($emergencyUser.DisplayName)..."
$policies = Get-AzureADMSConditionalAccessPolicy
foreach ($policy in $policies) {
if ($policy.Conditions.Users.ExcludedUsers -contains $emergencyUser.ObjectId) {
    Write-Host "Policy $($policy.DisplayName) properly excludes $($emergencyUser.DisplayName)."
} else {
    Write-Warning "WARNING: Policy $($policy.DisplayName) DOES NOT exclude $($emergencyUser.DisplayName). IMMEDIATE ACTION REQUIRED!"
}
}
} else {
Write-Error "Emergency account '$emergencyAccountUPN' not found."
}

Action 2: Diagnose Lockout Using Sign-in Logs

Use the Entra ID sign-in logs to pinpoint the problematic policy.

  1. Navigate to Entra ID admin center > Monitoring & health > Sign-in logs.
  2. Filter by User (the locked-out user) and Status: Failure.
  3. Click on the most recent failed sign-in attempt.
  4. Go to the Conditional Access tab to identify the policy that caused the block. Note the policy name.

Action 3: Disable Identified Policy (Temporary Fix)

Temporarily disable the problematic Conditional Access policy.

  1. Navigate to Entra ID admin center > Protection > Conditional Access > Policies.
  2. Find the policy identified in Action 2.
  3. Click on the policy name.
  4. Under Enable policy, select Off.
  5. Click Save.

Action 4: Create a New Report-Only Policy for Testing (Optional, but Recommended)

If you disabled a critical policy, create a duplicate in Report-only mode to test changes before re-enabling.

  1. Navigate to Entra ID admin center > Protection > Conditional Access > Policies.
  2. Click New policy.
  3. Configure the policy with the desired assignments and controls, ensuring emergency accounts are excluded.
  4. Set Enable policy to Report-only.
  5. Click Create.
  6. Monitor sign-in logs for a few hours/days, checking the Report-only tab.

The 60-Second Client & Executive Brief

“Today, we’re addressing a critical operational risk: Entra ID Conditional Access lockouts. These security policies are essential for protecting our cloud data, but if misconfigured, they can accidentally block access for our entire team, including IT administrators. Our immediate priority is to ensure we have ‘break-glass’ accounts that can always access our systems, even if standard policies fail. We are implementing a ‘Report-only’ testing phase for all new or modified access policies. This allows us to predict policy impacts before they go live, preventing service disruptions and maintaining our operational continuity while strengthening our security posture against identity-based attacks.”


Need Urgent IT Assistance or Perimeter Defense?

Is your organization vulnerable to active edge exploits, credential harvesting, or compliance audit penalties? Helpdesk.ca provides 24/7/365 managed IT support, emergency incident triage, and co-managed IT escalation for Canadian enterprises.


CANADIAN DATA SOVEREIGNTY & REGULATORY COMPLIANCE

Ensure Your IT Support Infrastructure Complies with Canadian Law

From mandatory Quebec Law 25 audit logging to PIPEDA data residency and Bill C-26 supply chain standards, Helpdesk.ca guarantees all ticket telemetry and cloud backups remain securely hosted in Canada.

Sources & References

Helpdesk.ca Assistant

Please enter your details to start chatting.

Hello! How can I help you today?