Fortinet NSE4_FGT_AD-7.6 Practice Test 35
Question 340
An approved HR SaaS page loads, but its embedded login widget comes from a second domain that the web filter blocks. What evidence should guide an exception?
- Set the entire blocked category to Allow for HR employees because the embedded widget belongs to an approved SaaS page.
- Disable deep inspection for all HR staff and assume that removes every secondary-domain web-filter decision.
- Identify the widget’s blocked secondary FQDN in matched web-filter logs and narrowly permit its approved dependency.
- Add another URL Allow for the main HR page but leave the secondary login widget hostname on the matched web-filter Block list.
Explanation
Correct Answer
C:A permitted parent page can load scripts, media and API endpoints hosted on different domains with separate FortiGuard ratings. Inspect the full browser request chain and FortiGate web-filter logs to identify the blocked secondary hostname and its matching profile. Correct a justified category or URL override for that endpoint only, then retest the app. Whitelisting the entire parent category may leave unrelated prohibited sites reachable.
Incorrect Answers
A:Category-wide permission can expose unrelated destinations that were intentionally blocked. Identify the specific secondary login widget FQDN and narrowly authorize only the verified application dependency.
B:Transport inspection and URL filtering are different controls. A blocked embedded hostname may still be rated and denied without decryption; diagnose the exact event before changing TLS policy. Switching off inspection for an entire department may weaken security while the secondary hostname still matches a blocked category.
D:The main page already loads, so another Allow for it cannot resolve a distinct embedded hostname denied by the security profile.
Question 341
A branch receives 198.51.100.10/30 for wan2 and cannot open HTTPS management on that interface even from the approved nearby console network. What should be verified?
- Add a generic operations-subnet forwarding firewall permit to WAN2 without enabling HTTPS administrative access or trusted-host authorization on the FortiGate interface.
- Confirm wan2 addressing, administrative HTTPS allowaccess, administrator trusted-host restrictions and the approved management source’s return route before testing a new login.
- Replace the interface's subnet mask with /8 to make a distant management workstation appear directly connected.
- Inspect the interface ping result from the nearby management host and use successful ICMP as the access test for HTTPS administrator login.
Explanation
Correct Answer
B:An assigned interface address is not sufficient for successful traffic or management. Check the actual interface mask and link status, interface role and permitted administrative services, then determine whether the upstream gateway has a usable return path. If the problem is management access, test the intended protocol from an authorized source; if it is transit traffic, inspect the route and firewall policy. Avoid broadening allowaccess or rules without establishing which path failed.
Incorrect Answers
A:Forwarded security policy does not create the device’s local HTTPS management listener. Validate allowaccess, trusted hosts, cloud/provider routing and the assigned interface.
C:An unapproved mask can create address overlap and routing instability. Confirm the assigned provider prefix and appropriate gateway/return path instead. An incorrect /8 mask invents local reachability and can disrupt addressing; retain the ISP-provided /30 and check remote path.
D:Ping can confirm basic IP connectivity and a return route between the approved management station and FortiGate. It cannot prove that the wan2 interface enables the administrative HTTPS listener, that local-in restrictions permit the source, or that the administrator account trusted-host setting matches. Validate the actual management service and authentication path independently.
Question 342
A new security operations policy should observe and log traffic from a pilot VLAN without denying it. The policy matches, but a security profile applied to it still uses a Block action for selected applications. How should the profile and policy be configured to provide monitoring without unintended disruption?
- Use an Accept pilot policy with Monitor/log application-control actions and verify no unapproved blocking occurs.
- Set the pilot VLAN forwarding policy to DENY with extensive logging, expecting application-control Monitor actions to let the traffic continue.
- Attach a default IPS sensor set to Block to the pilot Accept policy while calling its application-control Monitor profile a nonblocking evaluation.
- Use a policy that does not match the test interface, then rely on its monitoring logs.
Explanation
Correct Answer
A:An Accept firewall policy can still interrupt traffic if a security profile attached to it uses Block for the matched application. For a non-disruptive pilot, select supported Monitor actions for the scoped signature rules while retaining logs and any mandatory separate protections. Test representative sessions and verify log generation without an unexpected deny; restore enforcement deliberately after evaluation.
Incorrect Answers
B:A forwarding deny prevents eligible traffic before a Monitor-profile experiment. For nonblocking observation, the matched policy must permit traffic and security actions must be Monitor.
C:The applied IPS Block can still interrupt legitimate pilot flows. Verify all effective profile actions and test matched nonblocking traffic.
D:A profile on an unmatched rule cannot observe the intended session. The correct policy must match the actual source, destination, services and direction while accepting traffic.
Question 343
A requested static route points to gateway 203.0.113.254 on a WAN interface whose provider subnet cannot reach it. What is a valid prechange test?
- Check the static route’s installed FIB, gateway reachability, VDOM and competing priority before assuming it forwards.
- Accept the static route solely because its gateway IPv4 address is syntactically valid, leaving selected WAN subnet and ARP/next-hop reachability unverified.
- Review the gateway static route distance and lower it to zero before testing whether 203.0.113.254 is on link to the configured WAN interface.
- Inspect the WAN-to-destination firewall ACCEPT service and move the security permit above adjacent rules to reach the configured static gateway.
Explanation
Correct Answer
A:A configured static route with an unreachable next hop may not become a useful installed forwarding entry. Verify the gateway can be resolved through the specified interface, inspect the active routing table and check competing prefixes or policy routes before sending test traffic. If the route is rejected or inactive, correct reachability or the route parameters; do not assume the GUI configuration proves forwarding. Validate the packet's selected egress after the change.
Incorrect Answers
B:A configured route may be ineligible or unresolved even with valid syntax. Check actual on-link next-hop and installed FIB entry in the correct VDOM.
C:Lower distance improves preference among viable routes but cannot create physical or layer-two reachability to an ISP gateway outside the supplied link subnet. A more-preferred broken route may worsen the outage. Confirm the correct provider mask, next hop and active FIB before changing priority.
D:Firewall policy order can allow or deny a flow, but it cannot make an off-link gateway reachable by ARP on the wrong provider subnet. First confirm address and mask, interface state, installed route eligibility and next-hop adjacency; only then debug forwarding-policy matching if packets can leave.
Question 344
A remote branch should reach only one headquarters reporting server over a site-to-site IPsec tunnel. The current tunnel policy permits the entire server LAN. What should change?
- Keep the entire headquarters LAN network as the policy destination but require login at the reporting application to provide all branch segmentation.
- Increase the IPsec encryption level instead of reviewing the source/destination objects and allowed ports.
- Add the branch subnet to a shared VPN address group permitted to all headquarters private networks to simplify intersite access policy management.
- Restrict the route-based VPN policy to the reporting server /32, authorized branch sources and required service ports, without permitting the remainder of the headquarters LAN.
Explanation
Correct Answer
D:A functioning site-to-site tunnel may carry more traffic than the business requires if its firewall policy accepts the entire HQ server subnet. Keep the protected selectors and routes appropriate to the network design, but narrow WAN/VPN-to-HQ and return-direction policy address and service matches to the approved reporting server. Test both permitted reporting sessions and attempts to reach a neighboring HQ host to prove segmentation.
Incorrect Answers
A:An application login does not restrict which internal hosts are reachable at the firewall. The security requirement needs a /32 destination and narrowly defined service.
B:Strong encryption protects data in transit but does not decide which internal services a peer can contact. Correct the branch-specific firewall permission first. Stronger encryption protects confidentiality, not lateral movement between internal subnets; reduce permitted destinations and services.
C:A broad shared group grants access far beyond the one authorized reporting server and can create unwanted branch-to-HQ lateral access. Use precise protected source/destination objects and service matches.
Question 345
The security team must block anonymizing proxy applications while allowing normal HTTPS browsing. Which control avoids a port-wide outage?
- Deny TCP/443 for all application clients so the anonymizing proxy cannot establish encrypted sessions through the usual HTTPS port.
- Allow all HTTPS under a web-filter category profile and assume it can always distinguish anonymizing applications that reuse shared cloud/CDN hosts.
- Block supported anonymizer application signatures on the matched policy while testing that legitimate HTTPS remains allowed.
- Replace the application sensor with a generic destination IP allowlist containing only popular websites.
Explanation
Correct Answer
C:FortiOS application control can identify supported anonymizing tools independently of their common web transport port when signature and inspection visibility permit. Attach the correct sensor to the effective firewall rule, choose explicit actions for relevant anonymizer applications and test both prohibited tools and allowed ordinary HTTPS browsing. Review encryption, HTTP/3 and unknown traffic paths as additional bypass risks.
Incorrect Answers
A:This would also block permitted ordinary HTTPS, violating the business requirement. Application signatures and supported inspection provide a narrower boundary.
B:Domain categories and transport ports do not necessarily identify a specific anonymizer. Use supported application-control signatures and verified TLS/QUIC visibility.
D:IP-based lists may be brittle with CDNs and do not reliably identify anonymizing software. Use tested application signatures and exceptions for approved services. A destination allowlist cannot reliably distinguish proxy protocols hosted on permitted IPs; inspect supported application behavior.
Question 346
A newly joined FGCP member starts forwarding production traffic actively even though the design expects only a passive standby. Which setting should be checked?
- Increase HA heartbeat timeout to prevent the subordinate from becoming active after joining, leaving the configured active-active operating mode unchecked.
- Verify the configured FGCP active-passive mode and actual member/election status before changing security or heartbeat settings; determine whether the new member is actually carrying sessions or just synchronized state.
- Review the subordinate FGCP priority and set it to the minimum so the cluster behaves as active-passive during the next production test.
- Disable HA session synchronization so the secondary lacks state and is assumed unable to process new traffic when the cluster runs active-active.
Explanation
Correct Answer
B:Before treating a secondary unit carrying sessions as an HA failure, inspect the actual FGCP mode and the cluster's configured active-passive or active-active behavior. Confirm the member roles, load-distribution settings and health state from diagnostics, then compare them with the intended design. Unexpected processing might be a legitimate consequence of active-active configuration rather than a priority-election error. Changing device priority alone does not switch the cluster's operating mode.
Incorrect Answers
A:Heartbeat timing influences failure detection but does not define active/passive versus active/active load-sharing behavior. Inspect the actual FGCP mode and roles.
C:Member priority affects election preference and may help determine which device becomes primary in a supported healthy cluster. It is not the same as the configured HA mode: lowering standby priority cannot turn an active-active cluster into active-passive processing. Check the actual FGCP mode and observed traffic distribution, then correct the intended active-passive configuration rather than masking a mode mismatch.
D:Session-state replication and configured HA operating mode are different. Removing synchronization can cause connection loss without changing the intended traffic-sharing design.
Question 347
Two inbound VIP objects both use public 203.0.113.20 on the same WAN interface and overlap on TCP 8443, but each points to a different internal server. A test request reaches the wrong backend. Which VIP and service-match fields should the administrator reconcile before changing firewall permit scope?
- Resolve conflicting VIP matches for the same incoming WAN IP, protocol and TCP/8443 port by using nonoverlapping external port or interface bindings and matching policies.
- Choose one VIP by external public IP alone without comparing its port-forward protocol, external port and ingress interface tuple to the other object.
- Publish a single full-address VIP with no port restriction to replace all overlapping service objects.
- Increase the outbound source-NAT overload pool and test if the inbound TCP/8443 connection selects a different backend VIP afterward.
Explanation
Correct Answer
A:Overlapping VIPs can map the same external address to different servers or services, so the effective match must be inspected using the incoming interface, external IP, protocol and service port. Review the VIP definition and referencing firewall rules, including their order and any match-vip behavior. Adjust a conflicting port or interface binding with the smallest exposure change. Color labels and object names do not determine which translated destination receives a packet.
Incorrect Answers
B:Two VIPs can intentionally share a public IP with nonoverlapping ports. The problem is the overlapping complete external service tuple, which must be distinguished.
C:A broad mapping can redirect unrelated listeners and expose services that should remain separate. Examine external IP, translated port, interface and policy usage before consolidating anything.
D:Outbound SNAT pools govern source translation and do not remove conflicting destination VIP mappings. Correct the external port tuple collision.
Question 348
An HTTP/3 browser flow leaves over UDP/443 and is not decrypted by the FortiGate policy that deep-inspects TCP/443 HTTPS downloads. An application-control event identifies QUIC, but the security team requires full visibility into downloaded web content. Which verification is specific to the missing inspection capability?
- Inspect TCP TLS 1.3 cipher overlap on the firewall and adjust the browser-compatible SSL inspection profile for HTTPS.
- Add the website to the category Allow list and assume that makes HTTP/3 file payloads visible to the antivirus engine.
- Test supported FortiOS QUIC/HTTP3 visibility and, where needed, block UDP/443 for inspected TCP HTTPS fallback.
- Review the existing TCP/443 deep-inspection log and match its hostname against the current browser’s QUIC application-control event.
Explanation
Correct Answer
C:Application recognition of QUIC is not proof that the FortiGate has decrypted the HTTP/3 application payload. Check FortiOS 7.6.0 support, policy/SSL profile applicability, and file-scan event evidence. When the supported path cannot provide the required visibility, restrict QUIC narrowly so compatible clients retry using inspected TCP HTTPS, and verify both protection and legitimate function.
Incorrect Answers
A:Cipher selection is relevant to TCP TLS decryption, but the session in question is UDP/443 HTTP/3 QUIC. Tuning TCP TLS cipher suites will not turn its separate UDP payload into an inspected HTTPS file. Confirm supported QUIC visibility or intentionally test scoped fallback to TCP on the affected path.
B:A web category Allow action controls destination access and does not decrypt HTTP/3 file bytes. Antivirus can only scan content that its matched inspection path actually exposes; admitting the hostname is not equivalent to payload visibility.
D:A matching hostname in another TCP session does not prove that this UDP/443 download was decrypted. The original HTTP/3 flow uses a different transport and processing path. Correlate the actual QUIC session and its effective policy, then use a scoped supported action to establish the required content visibility.
Question 349
A probe target for ISP-B is reachable only through ISP-A. ISP-B appears healthy even when its own last mile is down. How should monitoring change?
- Keep the ISP-B SLA target reachable only over ISP-A and treat any returned probe as confirmation that ISP-B upstream routing is healthy.
- Bind ISP-B health probes to ISP-B itself and appropriate test servers; confirm failed ISP-B traffic cannot be judged healthy through ISP-A.
- Mark ISP-B healthy whenever its physical port is up, even if no packet reaches the provider gateway.
- Set ISP-B loss threshold to 100% to suppress false-negative alarms while leaving the upstream probe path routed through ISP-A.
Explanation
Correct Answer
B:A health-check target that can be reached only through ISP-A cannot validate ISP-B when the probe is inadvertently forwarded over A. ISP-B can look healthy while its last-mile path is down because the reply came over the surviving provider. Inspect the SD-WAN member bound to the probe, source-interface selection, installed route and next hop to the target. Correct member-specific monitoring and prove that disabling ISP-B changes its own health state rather than merely changing the shared target's reachability.
Incorrect Answers
A:A probe following the wrong member measures ISP-A, not the intended ISP-B path. Pin source/member and select reachable representative targets.
C:Carrier status does not prove end-to-end reachability or application SLA. Use meaningful monitored destinations and actual packet evidence per member. Physical link state cannot prove its upstream gateway or SLA target is reachable on ISP-B.
D:Loosening thresholds conceals genuine loss and does not repair the member-to-probe path. Verify ISP-B probe routing and independently test failure.
Question 350
An office switches to a different ISP. The FortiGate WAN interface has the new address but cannot reach its upstream gateway or internet sites. What should be verified?
- Retain the previous ISP default-gateway address but lower the route distance to make it the preferred path.
- Inspect egress firewall policy TCP/443 service matching and change its allowed services to ALL for the new provider path.
- Verify the new WAN mask, provider gateway ARP reachability and installed default route after ISP migration; retest provider return routing to both published services and outbound SNAT pools.
- Review Central SNAT public pools and disable central NAT before testing the new WAN subnet and provider default gateway.
Explanation
Correct Answer
C:A newly assigned WAN IP does not prove the interface can reach the ISP next hop. Compare the interface address and mask to the gateway supplied by the new provider, inspect link/VLAN and ARP, and verify the default route points to an actually reachable next hop. Only after upstream reachability is established should the operator review NAT pool ownership and security policies.
Incorrect Answers
A:The old gateway is not shown reachable on the newly addressed WAN interface. A lower distance can make an unusable next hop more preferred while leaving ARP and connectivity broken. Establish the actual new ISP gateway and mask, then verify the default route and permitted NAT return paths.
B:A restrictive service can prevent some client applications after basic routing works, but the symptom here starts with failure to reach the new WAN gateway. Widening policy services cannot repair an incorrect subnet mask or absent ARP adjacency. Verify on-link ISP gateway and installed default route first.
D:IP-pool translation may need updating during a carrier migration to preserve return routing, but neither removing source NAT nor changing its mode makes an off-link upstream gateway answer ARP. Check the new mask, gateway and physical path first, then reconcile allowed SNAT addresses with ISP routing.