All Cisco exams

Free 300-740 practice test: 30 questions with answers

30 original questions for Designing and Implementing Secure Cloud Access for Users and Endpoints (SCAZT) v2.0, drawn from every topic area. Answer each one to see whether you got it right and why. No sign-up.

Free questions
30
Full bank
112 questions
Real exam
90 min

300-740 exam guide: topics, cost and format

All 30 300-740 questions with answers

The same questions as the test above, grouped by exam topic. Open any question to see the answer and explanation.

Policies (6 questions)

  1. Question 1

    A company stores customer records in cloud object storage. Its security standard requires that the data stay unreadable if the underlying storage media is physically stolen. Which control meets this requirement?

    1. AHTTPS redirects that send all browser requests to the storage endpoint over port 443
    2. BEncryption of stored objects with keys managed through a key management service
    3. CTLS encryption that protects records while they travel between client applications and the storage service
    4. DAn IP allow list that restricts which source addresses can connect to the storage service
    Show answer and explanation
    Correct answer: B

    Encryption at rest protects stored data, so keys managed by a key management service keep the objects unreadable even if the media leaves the provider's control. TLS protects data in transit, while HTTPS redirects and IP allow lists control transport and access but do not encrypt stored data.

  2. Question 2

    A security engineer needs a control placed inline in the traffic path that blocks packets matching known exploit signatures instead of only raising an alert. Which technology should be deployed?

    1. AAn intrusion detection system fed by a mirrored SPAN port
    2. BA packet capture appliance that stores traffic for later forensic review
    3. CA network access control server that authenticates devices at each switch port
    4. DAn intrusion prevention system deployed inline in the traffic path
    Show answer and explanation
    Correct answer: D

    An IPS sits inline and can drop matching packets, while an IDS on a mirrored port can only alert. NAC authenticates devices for network admission and a packet capture appliance records traffic, so neither blocks exploit traffic.

  3. Question 3

    A policy must stop employees from uploading files that contain payment card numbers to personal file-sharing websites. Which capability is most appropriate?

    1. AA web content filter that blocks entire categories of social media and file-sharing sites
    2. BData loss prevention that inspects uploads and blocks matches for card number patterns
    3. CTwo-factor authentication required for every login to any file-sharing website
    4. DCertificate pinning on the corporate browser to stop it from trusting unknown certificates
    Show answer and explanation
    Correct answer: B

    Data loss prevention inspects the content of transfers and blocks data that matches sensitive patterns such as card numbers. Category filtering checks destinations rather than content, MFA controls logins, and certificate pinning addresses trust in TLS certificates.

  4. Question 4

    A public web application is receiving SQL injection attempts against its login form. Which control inspects HTTP requests at the application layer and blocks such payloads?

    1. AA volumetric DDoS scrubbing service that filters high-rate traffic floods
    2. BA web application firewall with rules that detect SQL injection patterns in requests
    3. CA network firewall rule that allows only inbound TCP port 443
    4. DLoad balancer health checks that remove failing backend servers from the pool
    Show answer and explanation
    Correct answer: B

    A WAF inspects HTTP requests and applies rules that detect injection attacks such as SQL injection. DDoS scrubbing targets traffic volume, a port-based firewall rule does not examine request contents, and health checks only manage backend availability.

  5. Question 5

    A team wants segmentation policies that keep following an application's workloads when they are moved between hosts or when their IP addresses change. Which approach fits this goal?

    1. ADefine policies only by static IP address ranges tied to each physical subnet
    2. BAllow all east-west traffic inside the data center and rely on the perimeter firewall for segmentation
    3. CDefine policies by workload labels or tags that enforcement follows as workloads move
    4. DAttach a separate VPN profile to each user that carries the segmentation rules with it
    Show answer and explanation
    Correct answer: C

    Label- or tag-based policy describes workloads by their attributes, so the rules stay valid when IP addresses change or workloads migrate. IP-only rules break under movement, allowing all east-west traffic defeats segmentation, and per-user VPN profiles do not govern workload-to-workload flows.

  6. Question 6

    A company routes branch and remote user web traffic through a cloud security service edge rather than inspecting it at each branch device. What is the main advantage of this design?

    1. ABranch devices no longer need any network connectivity, so remote users can work offline
    2. BEncryption keys for all applications are generated and stored by the branch routers
    3. CThe same policy applies wherever users connect, without separate appliances at each site
    4. DInspection is performed only for traffic originating from the corporate data center
    Show answer and explanation
    Correct answer: C

    A cloud security edge applies one policy to users regardless of location, so branches and remote users do not each need their own inspection appliance. The other options describe connectivity or key storage claims that are incorrect, and they would restrict inspection rather than extend it.

Access (6 questions)

  1. Question 7

    A user clicks a link in an email that points to a domain registered only days ago and linked to phishing. The organization wants the connection stopped before any page content is loaded. Which control provides the earliest protection?

    1. ADNS-layer security that refuses to resolve domains known to be malicious
    2. BFull packet capture of the user's browsing sessions retained for later investigation
    3. CTwo-factor authentication on the user's email account to prevent credential theft
    4. DEndpoint antivirus that scans the web page content after the browser has loaded it
    Show answer and explanation
    Correct answer: A

    DNS-layer security blocks the name lookup before any connection to the malicious host is made, which stops the page before it loads. Antivirus scanning happens after content arrives, packet capture only records traffic, and MFA protects the email account rather than blocking the destination.

  2. Question 8

    An administrator's secure web gateway decrypts and inspects HTTPS over TCP, but some browsers move their traffic to QUIC. Why do many designs block UDP port 443 toward the internet in this situation?

    1. ABlocking UDP 443 forces browsers to fall back to HTTPS over TCP, where the gateway can decrypt and inspect the session
    2. BBlocking UDP 443 is required because QUIC cannot be encrypted and therefore exposes all page content
    3. CBlocking UDP 443 disables DNS resolution, which makes the gateway see more of the traffic
    4. DBlocking UDP 443 lets the gateway skip certificate validation for destinations that use HTTP/3
    Show answer and explanation
    Correct answer: A

    QUIC runs over UDP and encrypts end to end with TLS 1.3, so a proxy built to inspect TCP-based HTTPS cannot see inside it. Blocking UDP 443 makes browsers fall back to HTTPS over TCP, which the gateway can inspect. QUIC is encrypted, DNS normally uses port 53 or encrypted DNS, and blocking a port does not change certificate validation.

  3. Question 9

    Employees have begun uploading files to unsanctioned SaaS file-sharing apps. The security team needs visibility into which apps are used and control over sensitive data in them. Which technology is designed for this?

    1. AA hardware load balancer that distributes requests across application servers
    2. BAn intrusion prevention system deployed at the data center perimeter
    3. CA cloud access security broker that discovers SaaS usage and enforces data policies
    4. DA security information and event management platform that collects on-premises server logs
    Show answer and explanation
    Correct answer: C

    A CASB gives visibility into SaaS and cloud app usage, including unsanctioned apps, and applies data protection policy to those apps. A load balancer, a perimeter IPS, and an on-premises SIEM do not discover or control SaaS app use.

  4. Question 10

    A DLP policy must catch copies of confidential design documents even after employees rename the files. Which detection method is the best fit?

    1. AFile-extension filtering that blocks every upload of a .docx file type
    2. BFilename matching against a list of internal project code names
    3. CBlocking every upload that originates from the design department's network segment
    4. DDocument fingerprinting that matches the content of known confidential files
    Show answer and explanation
    Correct answer: D

    Fingerprinting compares content, so renamed copies still match. Extension-based blocking is too broad and is easily bypassed by renaming, filename matching fails after a rename, and blocking a whole network segment does not identify the sensitive content.

  5. Question 11

    A contractor needs to reach one internal web application hosted in a private data center. Security requires that the contractor get no network-level reachability to other internal subnets. Which model meets this requirement?

    1. AA site-to-site IPsec tunnel that opens the entire internal subnet to the contractor's firm
    2. BA shared VPN account that places every contractor on the same flat internal network
    3. CZero trust network access that publishes only the one authorized application to the user
    4. DA remote desktop gateway that gives each session access to the full corporate intranet range
    Show answer and explanation
    Correct answer: C

    Zero trust network access brokers connections to individual applications, so the user never gets routed reachability to the internal network. The site-to-site tunnel, shared VPN account, and full-intranet gateway all expose more than the single application the contractor needs.

  6. Question 12

    Which statement about QUIC is correct?

    1. AQUIC is a routing protocol that exchanges route prefixes between autonomous systems
    2. BQUIC is a UDP-based transport protocol that builds TLS 1.3 into its connection setup
    3. CQUIC is a VPN protocol that requires IPsec for its key exchange and encryption
    4. DQUIC is a DNS extension that encrypts queries using certificates issued by the registrar
    Show answer and explanation
    Correct answer: B

    QUIC runs over UDP and integrates TLS 1.3 for its handshake and encryption, which is why modern web and remote access clients use it. Routing protocols such as BGP exchange prefixes, QUIC does not need IPsec, and it is not a DNS extension.

Identity (6 questions)

  1. Question 13

    A company configures single sign-on to a SaaS application using SAML. When the application receives an authentication assertion, what does it use to confirm the assertion really came from the trusted identity provider?

    1. AThe user's password, which the application stores in hashed form and compares against the value entered during each login attempt
    2. BA static API key that the application generated for each user when the account was first created during onboarding
    3. CThe identity provider's digital signature on the assertion, checked against the trusted certificate
    4. DThe client's source IP address, which the application records at login and checks during each later request
    Show answer and explanation
    Correct answer: C

    SAML assertions are signed by the identity provider, and the service provider validates that signature with the IdP certificate exchanged during trust setup. In SAML SSO the password is not sent to the SaaS application, and static keys or IP address checks are not how SAML establishes trust.

  2. Question 14

    An administrator wants an account disabled in the corporate identity provider to be deactivated automatically in several cloud applications. Which standard is designed for automated user provisioning and deprovisioning between systems?

    1. ASCIM, the System for Cross-domain Identity Management
    2. BOAuth 2.0, a framework for delegated authorization that issues access tokens
    3. CKerberos, a ticket-based authentication protocol used in Windows domains
    4. DRADIUS, a protocol for centralized authentication of network access requests
    Show answer and explanation
    Correct answer: A

    SCIM is an open standard for creating, updating, and removing user accounts across identity-aware cloud services. OAuth issues tokens for delegated access, while Kerberos and RADIUS authenticate users to domains or network access and do not manage account lifecycle across SaaS apps.

  3. Question 15

    A security team wants MFA that still protects users when a phishing site relays their password and one-time codes in real time. Which authentication method best meets this goal?

    1. ASMS one-time codes sent to the user's registered mobile phone number
    2. BA FIDO2 or WebAuthn security key that binds the login to the legitimate site
    3. CPush notifications that ask the user to approve or deny each sign-in attempt
    4. DTime-based one-time passwords generated by an authenticator app on the phone
    Show answer and explanation
    Correct answer: B

    FIDO2 and WebAuthn credentials are bound to the origin of the legitimate site, so a phishing proxy cannot reuse them. SMS codes and TOTP codes can be captured and relayed in real time, and push approvals can be abused through approval fatigue.

  4. Question 16

    A company allows only corporate-owned laptops to reach internal applications. Each laptop holds a certificate issued by the company's internal certificate authority. What is the main role of that certificate in the access design?

    1. ATo prove the device's identity to the access policy during authentication
    2. BTo replace the user's password for logging in to the operating system
    3. CTo encrypt every file stored on the laptop's disk and any attached removable media as it is written
    4. DTo publish a public DNS record that maps the laptop's hostname to its address
    Show answer and explanation
    Correct answer: A

    A device certificate from an internal CA lets the access service verify that a request comes from a managed device, and that check is combined with user identity and posture. Disk encryption, OS login, and DNS records are separate functions that a device certificate does not perform.

  5. Question 17

    Which two checks are typical endpoint posture conditions an access policy might require before a corporate laptop reaches an internal application? (Choose two.)

    1. AFull-disk encryption is enabled on the laptop
    2. BThe laptop has a personal browser bookmark folder with more than ten entries
    3. CThe laptop's desktop theme is set to a dark colour scheme
    4. DThe host firewall is turned on and running
    5. EThe laptop was purchased in the current calendar quarter
    Show answer and explanation
    Correct answer: A, D

    Posture policies check security state such as disk encryption, host firewall status, OS patch level, and endpoint protection presence. Bookmark counts, desktop themes, and purchase dates say nothing about the device's security state.

  6. Question 18

    A security team sees one employee with unusual sign-in locations in one identity provider and repeated failed MFA prompts in a second identity provider. The team wants to connect these signals into one risk view. Which capability addresses this need?

    1. AStoring every identity provider's password hashes in one shared vault for backup
    2. BDisabling federation so that each identity provider operates in complete isolation
    3. CReplacing SAML and OIDC with basic authentication to simplify sign-in across providers
    4. DCorrelating sign-in risk signals from several identity providers into one risk view
    Show answer and explanation
    Correct answer: D

    Identity intelligence correlates sign-in and risk events across providers so that patterns become visible in one place. Pooling password hashes, disabling federation, or using basic authentication would weaken security or fragment visibility instead of improving it.

Concepts (6 questions)

  1. Question 19

    Which statement best reflects the core principle of zero trust as described in NIST SP 800-207?

    1. AUsers on the corporate network are granted broad access because the perimeter has already been secured against outside threats.
    2. BTrust is never implied by network location, so each access request is evaluated before access is granted.
    3. CAccess rights are established once at login and stay valid for the whole working day unless the password changes.
    4. DDevices listed on the managed asset inventory are trusted automatically, because their configuration was checked when they were added.
    Show answer and explanation
    Correct answer: B

    NIST SP 800-207 assumes no implicit trust based on location or ownership, and each request is authenticated and authorized against policy. The perimeter model and one-time login model describe the legacy approach zero trust replaces, and inventory membership alone does not establish trust.

  2. Question 20

    A security architect is measuring the organization's progress with the CISA Zero Trust Maturity Model. Which list names the model's five pillars?

    1. AIdentity, Devices, Networks, Applications and Workloads, Data
    2. BIdentify, Protect, Detect, Respond, Recover
    3. CUsers, Endpoints, Perimeter, Cloud Applications, Logging
    4. DAuthentication, Authorization, Accounting, Encryption, Monitoring
    Show answer and explanation
    Correct answer: A

    The CISA Zero Trust Maturity Model organizes progress across five pillars: Identity, Devices, Networks, Applications and Workloads, and Data. The other lists are the NIST Cybersecurity Framework functions, AAA terms, or pillar names that do not appear in the model.

  3. Question 21

    A SOC team wants a shared vocabulary for describing how a cloud intruder escalates privileges, so that detections can be mapped to known adversary behaviors. Which resource is designed for this purpose?

    1. AISO/IEC 27017 cloud security control catalog
    2. BCIS Benchmarks, which publish hardened configuration settings for operating systems and platforms
    3. CNIST SP 800-53 security and privacy control catalog
    4. DMITRE ATT&CK cloud matrix, a catalog of adversary tactics and techniques
    Show answer and explanation
    Correct answer: D

    MITRE ATT&CK catalogs adversary tactics and techniques, including cloud-specific matrices, which lets detections be mapped to observed behavior. ISO 27017 and NIST SP 800-53 are control catalogs describing safeguards to implement, and CIS Benchmarks give hardening settings rather than adversary behavior.

  4. Question 22

    A platform team wants runtime visibility into process execution and network connections on Kubernetes nodes without changing application code. Which technology is well suited to this?

    1. AWeekly vulnerability scans of container images stored in the registry before any deployment is approved
    2. BFull-disk encryption applied to each Kubernetes worker node
    3. CSensors built on eBPF that watch process and network activity in the Linux kernel
    4. DSNMP polling of hypervisor interface counters for each virtual machine
    Show answer and explanation
    Correct answer: C

    eBPF lets sandboxed programs run inside the Linux kernel and observe runtime events such as process execution and network connections, which is the basis of tools such as Cilium and Tetragon. Image scanning happens before deployment, disk encryption protects data at rest, and SNMP counters report interface totals without showing process activity.

  5. Question 23

    Which hypervisor design runs directly on the physical server hardware and does not depend on a general-purpose host operating system?

    1. AA Type 1 hypervisor installed directly on the server's bare-metal hardware
    2. BA Type 2 hypervisor installed as an application on a desktop operating system
    3. CA container runtime that shares the host kernel among isolated application processes
    4. DAn application virtualization layer that streams desktop applications to endpoints
    Show answer and explanation
    Correct answer: A

    A Type 1 (bare-metal) hypervisor boots on the hardware and manages virtual machines directly. A Type 2 hypervisor runs on top of a host OS, container runtimes share a kernel rather than virtualizing hardware, and application virtualization does not host full virtual machines.

  6. Question 24

    In an infrastructure-as-a-service public cloud deployment, which two responsibilities typically remain with the customer? (Choose two.)

    1. APhysical security of the provider's data center facilities
    2. BConfiguring identity, access, and network policies for the resources they create
    3. CPatching the hypervisor software that runs on the provider's physical hosts
    4. DClassifying and protecting the data stored inside their virtual machines
    5. EMaintaining the physical network switches that connect the provider's racks
    Show answer and explanation
    Correct answer: B, D

    In IaaS the customer controls the operating systems, applications, data, and the access and network configuration of the resources it deploys. The provider secures the physical facilities, the physical network, and the hypervisor layer.

Operations (6 questions)

  1. Question 25

    A breach in a cloud application tier spread to database servers that share the same subnet. Which control best limits east-west movement between these workloads?

    1. AA single flat subnet for all workloads, which simplifies routing and firewall rules
    2. BDisabled flow logging on the database servers so that east-west traffic generates less noise
    3. COpen ports between all tiers so that application teams can troubleshoot connectivity faster
    4. DMicrosegmentation policies that permit only required flows between workload groups
    Show answer and explanation
    Correct answer: D

    Microsegmentation enforces allow lists between workload groups, so a compromised workload cannot reach others it has no need to contact. A flat subnet, disabled logging, and open ports all increase lateral movement or reduce visibility.

  2. Question 26

    Which data source records metadata about network connections, such as source, destination, ports and byte counts, without capturing the payload content?

    1. AFlow records such as NetFlow or IPFIX exported by network devices
    2. BFull packet capture of every session crossing the network edge
    3. CApplication debug logs that include complete request bodies for troubleshooting
    4. DScreenshots of administrator consoles taken at regular intervals for auditing
    Show answer and explanation
    Correct answer: A

    Flow records summarize who communicated with whom, on which ports, and how much data moved, without storing the content. Full packet capture stores payloads, debug logs can contain request bodies, and console screenshots do not describe network connections.

  3. Question 27

    Outbound traffic from a database server changes sharply at 2 a.m., after months of stable behavior. Which analysis is designed to flag this kind of deviation?

    1. ASignature matching of packet payloads against a catalog of known exploits
    2. BConfiguration compliance checking against a hardened benchmark document
    3. CBaseline behavior analysis that compares current traffic with learned normal activity
    4. DLog retention testing to confirm that audit and security records are kept for the full required period
    Show answer and explanation
    Correct answer: C

    Baseline behavior analysis learns normal patterns and alerts on significant deviations, which fits a sudden change after a stable history. Signature matching finds known exploit content, configuration checks compare settings to a benchmark, and retention testing verifies storage duration rather than behavior.

  4. Question 28

    Which type of check compares cloud resource configurations against a defined policy or benchmark and reports any drift?

    1. ATraffic flow baselining that tracks changes in conversation volumes between workloads over time
    2. BCompliance posture assessment that evaluates resource configuration settings
    3. CIntrusion signature matching against captured packet payloads
    4. DLog retention testing that verifies how long audit records persist
    Show answer and explanation
    Correct answer: B

    A compliance posture assessment evaluates configuration settings, such as storage access or encryption options, against a policy or benchmark. Flow baselining and signature matching examine traffic rather than settings, and retention testing checks how long logs are stored.

  5. Question 29

    Before enforcing microsegmentation rules on an application tier, what should an administrator do first?

    1. AEnforce a deny-all rule on every workload so that nothing can connect by accident
    2. BDisable flow logging so that the dependency mapping does not generate large data volumes
    3. CObserve and map actual traffic dependencies in monitor mode, then build allow policies
    4. DRemove all existing firewall rules so that the new policies start from a clean slate
    Show answer and explanation
    Correct answer: C

    Monitor-mode discovery shows the real dependencies between workloads, so allow policies can be built without breaking legitimate flows. Enforcing deny-all first, disabling logs, or removing existing rules would cause outages or leave gaps in visibility and protection.

  6. Question 30

    A legitimate application release changes the normal traffic pattern of a server. Which action keeps baseline analysis useful after this change?

    1. AKeep the original baseline forever so that the release is always treated as suspicious
    2. BTurn off baseline analysis for the server until the next breach investigation
    3. CDelete the server's flow history so that the old pattern cannot be compared with the new one
    4. DRe-baseline the server after the approved change, so that new normal behavior is recognized
    Show answer and explanation
    Correct answer: D

    A baseline represents normal behavior, so after an approved change it must be updated or the new pattern will keep generating false alerts. Keeping the old baseline, disabling analysis, or deleting history either produces noise or removes the basis for comparison.