Fortinet NSE4_FGT_AD-7.6 Practice Test 26
Question 250
FortiGate accepts a client connection on wan1 but return traffic arrives through wan2 and is dropped as out of state. Which information is missing?
- Add a WAN2-to-LAN Accept for the observed return traffic with the translated source and destination, retaining the original WAN1 session state unexamined.
- Inspect client TCP-window negotiation and extend retransmission timing for flows whose replies arrive over WAN2 after the original NATed session exited WAN1.
- Compare both WAN captures against the NATed session tuple and routing state to explain the out-of-state reply.
- Change outbound source NAT to a different public pool on wan2 and assume the old reverse packets will then match the existing wan1-created session without checking either original tuple or active session state.
Explanation
Correct Answer
C:An accepted outbound connection can still fail when replies follow an asymmetric path and encounter a firewall lacking the expected session state. Inspect the original five-tuple, NAT mapping, selected outbound route, reverse next hop and drop reason for return packets. Repair route or translation symmetry as required, then verify a complete TCP handshake and application response across the intended interfaces.
Incorrect Answers
A:A new permit may allow separate initiation but does not make unmatched return packets align with the existing session and NAT tuple. Capture both directions and determine state consistency first.
B:TCP receive windows and retransmission timers influence end-host behavior after packets are delivered; they cannot make FortiGate accept a reverse tuple that its state or NAT mapping does not recognize. The trace explicitly shows return packets on a different interface being dropped out of state. Compare both WAN packet captures with the translated session tuple and selected return route before touching endpoint timers.
D:Changing NAT on a different outbound path does not retroactively rewrite the recorded original session tuple or repair an out-of-state return. Capture the actual inbound and return packets, effective translations and FortiOS session lookup first; only adjust routing or NAT after identifying the specific mismatch.
Question 251
In profile-based FortiOS mode, a VIP forwards public TCP/8443 to a private server listening on TCP/443. The WAN-to-DMZ firewall policy references the VIP, but its custom Service object includes only TCP/8443; new sessions reach implicit policy 0 after DNAT. Which service/translation configuration should be verified?
- Change the VIP internal mapped port from TCP/443 to TCP/8443 so its external service number equals the existing inbound policy service.
- Enable source NAT on the inbound security policy.
- Use post-DNAT TCP/443 as the matching policy service while retaining the VIP public TCP/8443 mapping.
- Use a profile-based inbound Accept with service ALL to avoid the external-to-internal port mismatch and rely on the VIP to restrict access.
Explanation
Correct Answer
C:FortiOS performs VIP destination NAT before forwarding-policy selection in the described profile-based configuration. The policy's service test must accommodate the post-DNAT mapped TCP/443 port; the external VIP still listens on TCP/8443. Verify with targeted debug flow showing the original public destination, translated private TCP/443 endpoint and selected firewall policy. Avoid using an unrestricted ALL service if a scoped TCP/443 service suffices.
Incorrect Answers
A:The backend is explicitly listening on TCP/443, so altering the destination translation would send traffic to an unserved port. The inbound policy should match the supported translated destination service.
B:Source NAT changes client source addressing, not the destination service used by the policy after VIP port forwarding. Altering the client's translated source cannot make a TCP/8443-only service object match traffic evaluated after destination port translation to TCP/443.
D:ALL services unnecessarily broadens allowed traffic and obscures the actual mapped-port check. Preserve the narrow VIP port-forward and authorize its translated backend TCP/443 service.
Question 252
A branch can send traffic through its tunnel to headquarters, but return packets from headquarters do not arrive. The two FortiGates have separate tunnel firewall policies. What should be checked?
- Add a branch-side IPsec-to-LAN permit for all headquarters subnets but leave the headquarters route back to the branch network unchanged.
- Rotate branch IKE identities and PSKs while keeping the same headquarters return route for the still-established child-SA.
- Validate HQ return routes and VPN firewall directions for the branch prefix, including selector and session counters.
- Check the headquarters protected-prefix NAT policy and enable interface-address translation on all cross-site replies before evaluating the existing return route.
Explanation
Correct Answer
C:FortiGate route-based IPsec access depends on appropriate forward and return traffic policy along with tunnel selectors and installed subnet routes. Check branch-to-headquarters and headquarters-to-branch policy matches at both peers, including exact source/destination objects and service. Capture the two traffic directions and SA counters to locate which peer denies or misroutes the returning protected packet.
Incorrect Answers
A:One direction of access may need a policy, but the failed response leaves headquarters and must route through its tunnel to the branch. Test HQ return prefix and directional rule before expanding branch permits.
B:A working forward protected connection shows successful tunnel establishment. Credentials do not install the missing HQ return path or correct a directional firewall policy.
D:Source translation is sometimes used for overlapping networks or deliberate tunnel topology, and a policy may legitimately require it. But the case provides no overlap or source-routing limitation; it reports branch-to-HQ traffic working while the HQ return traffic is missing. Broad SNAT may break agreed traffic selectors or conceal source identity without correcting the actual HQ route or directional VPN policy. Diagnose those first.
Question 253
Immediately after a power-loss reboot, a FortiGate clock reads a date years in the past. Administrators see certificate-not-yet-valid messages when connecting to trusted management services, even though the issuing CAs did not change. What should be corrected first?
- Create a temporary global SSL inspection exception for all invalid certificate dates until the appliance has been running long enough after reboot.
- Replace the entire HTTPS security profile without checking device clock or NTP status.
- Reimport the same certificate and restart the SSL inspection service, leaving the FortiGate system time several years behind.
- Restore trusted NTP time, then recheck TLS certificate validity dates and the issuer chain.
Explanation
Correct Answer
D:Certificate validation relies on correct current time. After a power-loss reboot, the FortiGate clock reading years in the past can make legitimately issued certificates appear not yet valid. Diagnose the configured time source, NTP reachability and system time first, then retry the affected service with normal certificate verification rather than replacing trust roots or disabling checks.
Incorrect Answers
A:Certificate validity is evaluated against wall clock, not uptime. Bypassing validation across all destinations conceals a wrong device date and weakens trust; restore NTP first.
B:A different profile will not correct the recorded time used in certificate-date evaluation. Compare current FortiGate time, configured NTP and certificate notBefore/notAfter fields first. Verify NTP synchronization, timezone display, certificate notBefore/notAfter dates and the active SSL inspection profile before bypassing checks.
C:Reimporting does not change the certificate’s signed Not Before/After values, and a clock in the distant past will continue to produce invalid-date outcomes. Synchronize the clock and recheck.
Question 254
An AWS inspection design routes subnet traffic through a FortiGate-VM and then to an internet gateway. After enabling the appliance path, some requests arrive but return through a different network interface; the VM still has source/destination checks enabled. Which cloud and firewall forwarding settings should be validated?
- Increase the FortiGate-VM security-policy coverage of the returning subnet while retaining EC2 source/destination checks on the forwarding interface.
- Associate the destination workload subnet with an internet-gateway route instead of the FortiGate return-path route.
- Open the application's return TCP ports in both VPC security groups while keeping the VM source/destination check enabled.
- Disable required AWS source/destination checks on forwarding FortiGate-VM NICs and validate both directions of cloud route steering.
Explanation
Correct Answer
D:AWS virtual appliances forwarding third-party packets typically require the EC2 network-interface source/destination check to be disabled. That change alone does not steer workload traffic; inspect both subnet route tables, gateways, appropriate FortiGate policy/NAT and any high-availability return route. Compare forward and reverse packet captures and route tables to ensure the same stateful appliance processes the complete session. Apply changes only on the authorized firewall forwarding interfaces.
Incorrect Answers
A:A FortiGate security rule may be needed for forwarded traffic, but an AWS forwarding appliance generally requires the applicable instance NIC source/destination check to be disabled. Extra FortiOS permits cannot override a cloud-fabric forwarding block or repair the observed asymmetric return route.
B:A direct internet-gateway path may improve raw reachability for public workloads but deliberately bypasses the intended stateful inspection service. The question needs both request and return traffic to traverse the forwarding FortiGate-VM with source/destination check and cloud routes configured correctly.
C:Security-group rules determine whether traffic is permitted at a cloud NIC, but they do not remove AWS's source/destination forwarding check or correct a subnet route selecting the wrong return interface. Validate VPC routes and appliance forwarding behavior before widening security permissions.
Question 255
FortiGate's user authentication portal does not appear on the expected LAN interface, although clients can reach the firewall by ping. What should be verified?
- Check the actual captive-portal listener and local-in access on the expected LAN interface and its matching identity policy.
- Enable FortiGate administrative HTTPS on the LAN and raise the administrator session idle timeout.
- Create a URL-filter Allow entry for the expected captive portal hostname and retest web login.
- Inspect the identity-based captive-authentication policy and replace its user-group condition with an unrestricted address-based ACCEPT for the affected LAN.
Explanation
Correct Answer
A:ICMP reachability to a FortiGate interface does not prove that the captive portal or firewall authentication service listens there. Verify the portal host/IP and scheme, administrative or authentication-service access as applicable, matching user/group policy, client DNS and the source interface/route. Capture the browser redirect and FortiGate local response, then correct only the mismatched listener or authentication-policy configuration.
Incorrect Answers
B:Administrative HTTPS and captive authentication are related to different local services. Opening the management GUI may not establish the required user portal redirect or identity policy, and raising an admin idle timer cannot bring up a missing captive listener. Verify the actual portal settings and relevant interface/policy.
C:A URL allow can be helpful if a web filter is blocking access to an external portal, but the question reports that the expected FortiGate interface fails to present its own portal. Listener/security mode, authentication policy and local-in reachability must be established before attributing failure to website categorization.
D:Replacing identity conditions with an unrestricted forwarding rule may bypass a broken portal symptom for some web traffic, but it defeats the required user authentication and does not repair the FortiGate portal listener. The administrator must verify the actual LAN interface, local-in authentication service, redirect settings and the appropriate user-policy match before changing access scope.
Question 256
A fresh HTTPS session matches an ACCEPT firewall rule, then receives a web-filter block page. An engineer wants to remove the ACCEPT rule. Which sequence should be examined?
- Review the HTTPS ACCEPT rule’s order, move it above nearby permits and repeat the blocked website request.
- Inspect the browser certificate chain and reinstall the managed FortiGate SSL-inspection CA if its TLS validation is failing.
- Correlate the accepted firewall session with the web-filter profile’s actual URL/category Block verdict before changing security rules.
- Add a second broader ACCEPT policy above the first, keeping the same filtering profile, to force the site open.
Explanation
Correct Answer
C:The forwarding rule can ACCEPT an HTTPS flow while its attached web-filter profile subsequently blocks a URL or FortiGuard category. Match the same session and policy ID to its category, static URL entry and security-profile action. Only change the demonstrated filtering cause, and then retest the allowed site and known prohibited requests on a fresh session.
Incorrect Answers
A:Ordered forwarding rules determine the initial policy match, but the observed flow already matched an ACCEPT rule and then received a web-filter Block page. Reordering that same permit need not change the attached URL/category action. Identify the security-profile event and exact reason before deleting or moving the rule.
B:Client trust repair is relevant when a browser cannot establish the inspected TLS connection. Here the client receives a FortiGate-generated web-filter Block page after the forwarded session was accepted, indicating a content-policy decision rather than proving a certificate-chain error. Review the web-filter event first.
D:A different forwarding permit does not necessarily change the later web-filter decision. Determine which matched profile denied the URL and apply only an approved exception. Test a fresh request and change only the justified URL/category control rather than removing the legitimate forwarding permission.
Question 257
An LTE link becomes the third SD-WAN member as a metered backup. It must not carry regular downloads when both broadband members meet SLA. How should it be constrained?
- Give LTE the highest SD-WAN interface preference whenever it passes the shared broadband health check, reducing latency while also leaving downloads eligible.
- Add LTE to the default load-balancing pool with the same weight as broadband and rely on usage billing alerts.
- Disable the LTE health-check probe so the modem remains physically connected but is selected automatically after broadband performance degrades.
- Exclude metered LTE from normal balancing and configure it only for explicit eligible failover with cost safeguards.
Explanation
Correct Answer
D:A metered backup should remain monitored and reachable for genuine broadband failure without sharing routine sessions while the cheaper links meet their SLAs. Check SD-WAN member priority, the service-rule strategy and candidate members; make broadband the eligible preferred set and LTE a bounded fallback for required traffic. Verify the selected link for ordinary downloads in healthy conditions and after both broadband members fail.
Incorrect Answers
A:LTE might meet technical SLA but still be a metered standby; prioritizing it while broadband is healthy violates cost policy. Member eligibility and service-rule priority must explicitly reserve LTE for failover.
B:Equal participation can send ordinary sessions across the costly link before alerts fire. Define explicit steering policy and measured failover behavior. Check each member's configured cost, interface preference, health target, SLA condition and rule position relative to general internet traffic.
C:A physically linked modem may have no usable upstream path. Without valid SLA evidence, failover can select an unhealthy or unmetered-by-rule member; test LTE readiness in a separate fallback policy.
Question 258
An IPsec peer still proposes an obsolete weak encryption algorithm in IKE and ESP, while the organization requires supported modern authenticated encryption. The other FortiGate supports stronger suites. What should be negotiated and tested to avoid a weaker fallback or repeated proposal mismatch?
- Retain the existing obsolete IKE and ESP algorithms until all spokes migrate, using longer SA lifetime to reduce renegotiation failures.
- Configure only the headquarters appliance for the strongest AEAD encryption, removing every shared transform before checking the older peer’s support.
- Turn off IKE authentication to compensate for rejecting the old encryption suite.
- Agree on approved IKE and ESP suites supported by both peers, preserving PFS and authentication before retiring weak proposals.
Explanation
Correct Answer
D:IPsec peers must agree on supported cryptographic proposals, authentication and any PFS settings required for the child security association. Compare phase-one/phase-two negotiation errors, algorithm capabilities, selector requirements and relevant peer software versions rather than removing encryption. Select an approved interoperable proposal for that peer and verify authenticated IKE, child-SA establishment and protected traffic afterwards.
Incorrect Answers
A:Longer lifetime does not make the weak algorithms compliant or prove a secure negotiated overlap. Identify mutually supported approved IKE/ESP/PFS proposals before retiring the obsolete suite under a controlled transition.
B:A cryptographically stronger local proposal cannot establish an SA if the remote endpoint supports no compatible transform. Use an approved secure common suite and verify the actual negotiation results on both peers.
C:Peer authentication and encryption strength protect different tunnel properties. Removing authentication cannot fix a proposal mismatch and creates a separate impersonation risk. Select an approved interoperable proposal for that peer and verify authenticated IKE, child-SA establishment and protected traffic afterwards.
Question 259
A SOC analyst cannot find FortiSASE security events in the configured reporting console, although the tenant policy is enforcing correctly. What should be verified?
- Increase FortiSASE threat-policy severity and disable existing benign-rule logs, expecting the missing web-security events to appear in the reporting console.
- Search an on-premises FortiGate’s traffic log for each cloud FortiSASE web event without configuring a tenant-to-collector event export.
- Review the remote endpoint’s FortiSASE traffic-steering policy and route additional client browsing through the secure service before checking reporting-console ingestion.
- Validate FortiSASE event generation and the configured export destination, ingestion mapping, retention and query time window.
Explanation
Correct Answer
D:Enforcement at the cloud security service does not guarantee that every event was forwarded to the particular SOC dashboard. Compare a known action with FortiSASE tenant log generation and retention, the enabled event category, external forwarding/integration and collector receipt. Validate the time zone, user/device identity and original source IP fields before declaring an event absent.
Incorrect Answers
A:Changing enforcement severity does not create an absent export integration or a correctly indexed event. Check generation at the tenant, destination permissions, retention and actual search fields first.
B:FortiSASE cloud events are not guaranteed to appear in unrelated local firewall traffic logs. Confirm supported tenant logging and any export connector before treating the central reporting gap as loss of enforcement.
C:Traffic steering can determine whether web requests receive cloud enforcement and associated logs, so it is a credible troubleshooting dependency. The scenario already states the tenant policy is enforcing correctly but events are absent from the designated reporting console. Changing steering risks modifying working protection without resolving event export, collector authorization, field mapping, retention or the search interval.
Question 260
A security engineer changes an application-control signature from Monitor to Block but an already established session still passes. What should be tested?
- Inspect the updated signature setting in the Application Control GUI and reload the profile on the existing firewall policy.
- Review the FortiGuard application-signature package version and schedule an immediate database update for the same sensor.
- Retest a new session under the updated application-control Block action and compare its verdict with the existing cached session.
- Disable session tracking globally so all active applications are instantly rescanned against the new signature.
Explanation
Correct Answer
C:An existing application session may retain classification or policy state after a sensor edit, so a new connection is needed to evaluate updated enforcement. Confirm the selected application-control sensor, signature ID, old session creation time and new action in the current policy. Establish a fresh test session, compare its actual block/monitor action and logging, and document whether clearing old state is necessary.
Incorrect Answers
A:Confirming the saved profile is useful, but a GUI reload does not establish whether a previously classified session was re-evaluated after the Monitor-to-Block change. Start a fresh matching flow and compare its signature action and logs with the older session state before resetting unrelated active connections.
B:An outdated signature definition can affect recognition, but the observed issue is an administrator changing an already identified signature action. Updating definitions does not by itself guarantee retroactive blocking of an established session. Test a newly created flow against the saved Block action, then decide whether targeted state cleanup is necessary.
D:Removing stateful controls introduces major security and stability risk. Use a controlled session reset if justified, along with a new-session validation test. Establish a fresh test session, compare its actual block/monitor action and logging, and document whether clearing old state is necessary.