CISSP domain - 13% of the exam

Identity and Access Management (IAM)

Identity and Access Management (IAM) is 13% of the CISSP exam. These are the objectives it covers, each with practice questions, with every answer explained.

The domain in numbers

39
Practice questions
6
Objectives
9
Easy
12
Medium
18
Hard

Where people go wrong

  • Pairing a scenario with a protocol built for another job

    The wrong answers pair a scenario with a protocol built for another job. SAML suits browser single sign-on between organisations, not native mobile clients calling REST APIs, and the resource owner password credentials grant defeats consent and MFA. TACACS+ separates authentication, authorisation and accounting and supports per-command control on network devices, which RADIUS and Kerberos do not. OpenID Connect adds an ID token to OAuth 2.0, which is an authorisation framework.

    See it in a sample question
  • Resetting a password when the attacker holds a token

    The wrong answers treat a stolen session cookie or forged ticket as if it were a stolen password. Resetting user passwords, re-enrolling TOTP, requiring smartcards, blocking the operator's own IP address or shortening ticket lifetimes all leave the attacker's token valid. A session cookie is a bearer token that works until the server-side session is revoked, so revoking active sessions comes first. A forged ticket-granting ticket is signed with the krbtgt key, so the krbtgt password is reset twice, with replication between the resets.

    See it in a sample question
  • Choosing an access control model that does not fit the scenario

    The wrong answers pick an access control model that does not fit the scenario. DAC leaves grant decisions to data owners, so recipients can re-share and entitlements drift across branches. RBAC anchored to an HR job catalogue suits stable roles but explodes when entitlements depend on many attributes, and it cannot react to session risk. ABAC suits multi-attribute, multi-jurisdiction decisions, and risk-based control gates sensitive actions on a live risk score with step-up authentication.

    See it in a sample question
  • Revoking or rotating in bulk before fixing the process

    The wrong answers act in bulk before the process behind the problem is understood. Mass revocation of secondary roles, automated revocation of unused entitlements, and mass rotation of service account credentials can break patient care or production workloads. The correct answers start with governance: a recertification campaign attested by role owners, a repaired review procedure that gives managers usage data, and a named human owner and purpose for each service account.

    See it in a sample question
  • Securing a sensitive room with a credential that identifies nobody

    The wrong answers secure a sensitive room with a credential that identifies nobody: a shared key, a shared PIN, a shared card, or a building-wide master key. These give no individual accountability, cannot be revoked cleanly when one person leaves, and admit people whose duties do not require entry. The correct answers use individually issued smart cards, restricted to the named staff whose role needs the room, with the reader logging each entry.

    See it in a sample question

Objectives in this domain

What to study

The official documents this domain's practice questions cite most.

Sample questions from this domain

Free sampleIdentity and Access Management (IAM)hard

A regional bank is launching a public-facing mobile banking app that will call its own first-party REST APIs and a third-party budgeting partner's API on behalf of customers. The CISO wants the customer to authenticate once with strong assurance, the mobile app to never see the partner's tokens, and the partner to receive only the consents the customer approved. Which authentication and authorisation architecture should the security architect recommend as the BEST fit?

  • AImplement OpenID Connect for customer authentication at the bank's IdP and use OAuth 2.0 authorisation code with PKCE so the mobile app obtains scoped access tokens, with a separate token exchange for the partner API. Correct
  • BIssue customers a SAML 2.0 assertion from the bank's IdP and have the mobile app post it directly to the partner's SaaS endpoint as a bearer credential.
  • CUse the OAuth 2.0 resource owner password credentials grant so the mobile app collects the bank password once and stores a long-lived refresh token for both the bank and partner APIs.
  • DFront the APIs with a RADIUS server that authenticates each API call using the customer's banking PIN as a shared secret over TLS.
Select OpenID Connect with OAuth 2.0 authorisation code plus PKCE for mobile clients that need user authentication and delegated, scoped API access. OpenID Connect layers a signed id_token on top of OAuth 2.0 to give the relying party verifiable authentication, while OAuth 2.0 authorisation code with PKCE is the IETF-recommended flow for public clients such as native mobile apps because it binds the code exchange to a per-session secret the attacker cannot replay. Scoped access tokens plus a token exchange to the partner mean the mobile app and partner each receive only the rights the customer consented to, satisfying least privilege and credential isolation.

Why A is correct: OIDC handles the authentication assertion (id_token) while OAuth 2.0 authorisation code with PKCE is the recommended public-client flow for mobile apps, and scoped access tokens plus token exchange let the partner receive only the consents granted without seeing customer credentials.

Why B is wrong: SAML is optimised for browser-based SSO between enterprise web applications, not for native mobile clients calling REST APIs, and forwarding the assertion to the partner gives the partner full session rights rather than scoped consent.

Why C is wrong: ROPC requires the client to handle the user's primary credentials and is deprecated for public clients precisely because it defeats consent, MFA prompts, and credential isolation, all of which the CISO explicitly required.

Why D is wrong: RADIUS is a network access authentication protocol for devices such as VPN concentrators and Wi-Fi controllers, not a token-issuing authorisation framework for federated mobile API access, and per-call PIN replay is poor assurance.

Free sampleIdentity and Access Management (IAM)hard

During a tabletop exercise, a CISSP discovers that a Kerberos realm at a manufacturing firm uses a single service account for the KDC's krbtgt principal whose password has not been rotated in nine years. The blue team also finds a domain compromise four months ago that was remediated by resetting user passwords only. Which action should the security manager prioritise FIRST to restore trust in the Kerberos infrastructure?

  • AForce a password reset on every domain user and require smartcard logon for administrators across the realm.
  • BReset the krbtgt account password twice in succession, allowing replication between the resets, then rotate service account keys identified as exposed. Correct
  • CShorten the maximum TGT lifetime from ten hours to one hour and enable AES-256 only across all domain controllers.
  • DMigrate domain authentication from Kerberos to RADIUS with EAP-TLS so that ticket forgery is no longer possible.
After a domain compromise, rotate the krbtgt account twice with replication between resets to invalidate any golden tickets minted from a stolen krbtgt hash. Kerberos issues a TGT signed and encrypted with the krbtgt account's long-term key. An attacker who has obtained that key can mint arbitrary TGTs (golden tickets) for any principal at any privilege for the lifetime of the key. The KDC keeps the current and previous krbtgt key to allow in-flight tickets to validate during rotation, so a single reset only invalidates tickets signed with the older-previous key. Two sequential resets, with replication between them, are required to retire both the compromised key and its remembered predecessor.

Why A is wrong: User password resets and smartcards do not invalidate forged ticket-granting tickets, because the TGT is signed with the krbtgt key, not user keys, so any golden ticket from the breach remains valid.

Why B is correct: The krbtgt key signs every TGT; resetting it twice (with replication between) invalidates any golden tickets minted from the stolen hash, which is the only way to actually evict an attacker holding krbtgt material from a prior compromise.

Why C is wrong: Shorter lifetimes and stronger ciphers raise the bar for new attacks but do not invalidate forged tickets already signed with the long-lived krbtgt hash that the attacker likely captured.

Why D is wrong: RADIUS is not a replacement for the realm's SSO ticketing service for desktop, file, and application access, and a rushed protocol swap would break dependent services without addressing the immediate forged-ticket exposure.

Free sampleIdentity and Access Management (IAM)hard

A pharmaceutical company runs clinical trial documents on a shared platform where each document is owned by a principal investigator. Investigators frequently grant collaborators read or comment access, and the platform's access control lists are managed by the document owners themselves. A regulator asks how the company prevents authorized collaborators from passing trial data to unauthorized third parties through onward sharing. Which BEST describes the inherent limitation the security manager must explain to the regulator?

  • ADiscretionary access control delegates grant authority to data owners, which means an authorized recipient can re-share the data and the owner cannot reliably constrain that downstream flow Correct
  • BRule-based access control cannot express conditional sharing rules, so onward sharing is blocked only when an explicit deny rule is added to each document
  • CAttribute-based access control treats collaborator attributes as transitive, which means any onward share preserves the original confidentiality envelope of the document
  • DMandatory access control labels are stripped when a document is shared outside the originating compartment, exposing it to onward sharing risk
Recognise that DAC's owner-delegated grants cannot enforce constraints on onward sharing by authorized recipients. Under DAC, the owner of an object grants permissions to subjects, but once a subject has read access, the model cannot prevent them from copying, exporting, or re-sharing the data. This is a well-known intrinsic limitation of DAC, addressed in stricter regulated settings by overlaying MAC, information rights management, or data loss prevention controls.

Why A is correct: DAC's defining limitation is exactly this: once an owner grants access, the recipient has discretion to copy or re-share, and the model provides no enforced constraint on the downstream propagation of data.

Why B is wrong: The platform is not described as rule-based, and rule-based controls are not the model in play; this answer misnames the issue and points the regulator at the wrong control gap.

Why C is wrong: ABAC is not the model described, and onward sharing is not automatically envelope-preserving; this answer invents a property that ABAC does not provide and would not satisfy a regulator.

Why D is wrong: MAC is not in use on this owner-managed platform, and well-implemented MAC does not strip labels on sharing; this answer misattributes the weakness to the wrong model.

Free sampleIdentity and Access Management (IAM)hard

A hospital network's RBAC model has grown to over four hundred roles after years of ad hoc additions, and an internal audit reports that nearly a third of clinical staff hold at least one role they no longer need. The CISO has been asked for the FIRST corrective step that will most reduce standing entitlement risk without disrupting patient care. Which action should the security manager take first?

  • AImplement attribute-based access control across all clinical systems to replace the existing role catalogue within the next quarter
  • BDisable all secondary role assignments for clinical staff and require them to request reinstatement through the service desk
  • CEngage a consultancy to perform role mining on entitlement logs and design a consolidated role catalogue before any recertification
  • DCommission a structured access recertification campaign with managers and role owners attesting that each member's current role assignments remain justified Correct
Prioritise access recertification as the first response to documented over-entitlement in a mature RBAC environment. CISSP guidance treats periodic access reviews and recertification as the primary control against entitlement drift and role bloat. With auditors highlighting a third of clinical staff holding unneeded roles, a manager-attested recertification campaign removes standing risk fastest and feeds cleaner data into any subsequent role redesign or ABAC migration.

Why A is wrong: A wholesale move to ABAC is a multi-year programme in a hospital and does not address the immediate over-entitlement problem; it also introduces patient-care risk if rushed, so it is not the right first step.

Why B is wrong: Mass-revoking secondary roles is operationally reckless in a hospital, will disrupt patient care, and short-circuits the governance process that should decide which assignments are actually unjustified.

Why C is wrong: Role mining and consolidation are valuable medium-term work, but they do not by themselves remove the unjustified assignments highlighted by audit; recertification should run first to address the standing risk.

Why D is correct: Access recertification driven by managers and role owners directly removes unjustified assignments, reduces standing entitlement risk quickly, and is the manager-led first move CISSP expects before deeper redesign work.

Free sampleIdentity and Access Management (IAM)easy

A regional bank is fitting out a new branch and the security manager must decide how staff will enter the cash-handling room behind the teller line. The room currently has a single keyed lock shared by all branch staff. What is the BEST control to recommend for access to this room?

  • AKeep the keyed lock but issue a separate key to every member of branch staff so that nobody has to share.
  • BInstall a four-digit numeric keypad on the door and rotate the code each quarter through an all-staff email.
  • CReplace the keyed lock with a smart-card reader tied to each employee's identity, restricted to staff whose role requires cash handling. Correct
  • DLeave the existing keyed lock in place and add a closed-circuit camera pointed at the door to deter misuse.
Recognise that role-based individual-identity credentials are the primary physical access control for sensitive areas, supported by other layers. Physical access to a sensitive area should be controlled by binding entry to an individually issued credential and restricting it to roles that require the access. This implements least privilege, supports accountability through an entry log, and allows access to be revoked when a person changes role or leaves. Shared keys, shared PINs and cameras alone fail one or more of these properties and so sit beneath identity-based entry in the control hierarchy.

Why A is wrong: Tempting because it removes the shared-key problem, but physical keys cannot be revoked cleanly when staff leave or change role, produce no log of entries, and still grant access far more broadly than the cash-handling role requires.

Why B is wrong: A shared PIN looks cheaper than card readers and is easy to deploy, but it provides no individual accountability, is trivially shoulder-surfed or shared, and quarterly rotation by mass email defeats any confidentiality the code might otherwise have.

Why C is correct: Binding entry to an individual identity with role-based restriction gives accountability, supports the least-privilege principle, and produces an audit trail of who entered the cash room and when, which is the manager-led, risk-based choice for a sensitive area.

Why D is wrong: Cameras are a useful detective layer, but on their own they do not prevent unauthorised entry, do not enforce role-based restriction, and assume someone will review the footage promptly, which makes this a supporting control rather than the primary access control.

Other domains in this exam

See also the CISSP cert hub, the study guide, and the cheat sheet.

Examworthy is not affiliated with or endorsed by ISC2. Original, blueprint-aligned practice material only.