When the FortiGate enters Conserve Mode due to high memory pressure (specifically reaching the Extreme Threshold at 95% memory usage, or the Red Threshold for proxy traffic), the system prioritizes stability and preventing a system crash (kernel panic).
D). FortiGate begins dropping all new sessions to protect resources:
In Extreme Conserve Mode (95%), the FortiGate kernel acts to preserve the remaining memory for system- critical tasks (like admin access and basic packet forwarding of existing sessions). To achieve this, it drops all new session initiation requests regardless of the inspection type.
In Red Conserve Mode (88%), it specifically drops new sessions that require proxy-based inspection (as these consume the most memory), while often still allowing flow-based traffic.
Among the provided choices, "dropping new sessions" is the only standard protective mechanism FortiOS employs to stop memory usage from climbing further.
Why other options are incorrect:
A: FortiGate does not automatically reboot in conserve mode; it attempts to recover by restricting traffic.
(Reboot is a last-resort crash, not a configured action).
B: Inspection modes (Proxy vs. Flow) are defined in firewall policies and cannot be dynamically switched by the system during runtime.
C: The system does not arbitrarily stop "non-essential processes" like logging or AV. Logging is critical for audit trails. While av-failopen can be configured to bypass scanning, the system typically defaults to "Fail- Close" (dropping traffic) rather than stopping the engines themselves.
Reference:
FortiGate Security 7.6 Study Guide (Diagnostics & Resource Usage): "When memory usage reaches the extreme threshold (95%), all new sessions are dropped to prevent memory exhaustion."

The key point is that the two VPNs are dynamic dialup IPsec tunnels on the same interface and both are using IKEv1 main mode. In this design, FortiGate cannot reliably distinguish which dialup phase1 to match before phase 1 completes.
The uploaded Network Security Support Engineer 7.6 Study Guide shows that XAuth happens only after phase 1 is already established:
"The IKE real-time debug shows, after phase 1, the exchange of extended authentication (XAuth) packets... You can also see the CFG_REPLY, showing the XAuth user and group name." That means the user group is learned too late to be used for selecting the correct phase1 definition. So the fix must be applied to the phase1 matching method itself, not to XAuth.
The FortiOS administration guide gives the exact rule for this scenario:
"When the remote VPN peer has a dynamic IP address and is authenticated by a pre-shared key you must select Aggressive mode if there is more than one dialup phase 1 configuration for the interface IP address."

The correct answer is C .
The exhibit shows the neighbor state as Connect in the State/PfxRcd column. The study guide explains the BGP states exactly as follows:
* "Connect: Waiting for a successful three-way TCP connection"
* "OpenSent: Waiting for an OPEN message from the peer"
* "Established: Peers have successfully exchanged OPEN and keepalive messages" Because the router is still in Connect state, the TCP three-way handshake has not completed yet . In practical terms, the local router has sent the TCP SYN but has not successfully received the SYN/ACK needed to complete the handshake . That is why C is correct.
Why the other options are wrong:
* A is wrong because the message counters alone do not prove that the neighbor is unreachable. The study guide says the State/PfxRcd field shows the BGP state when the session is not established, and here that state is specifically Connect
* B is wrong because waiting for an OPEN message happens in OpenSent , not Connect
* D is not the best answer for this output. The study guide ties the displayed state directly to the protocol phase: Connect means the device is still waiting for a successful TCP handshake So the verified answer is: C .

To determine the correct reasons for the adjacency failure, we must analyze the standard OSPF real-time debug output (diagnose ip router ospf all enable or diagnose sniffer packet) typically provided in this exam exhibit.
Analyze the Debug Output:
The debug output in this specific question scenario typically displays an incoming Hello packet line: OSPF:
RECV[Hello]: ... auth-type 0 ...
" RECV " : Indicates the packet is coming from the Remote peer.
" auth-type 0 " : Indicates the Remote peer is sending " Null " (No) authentication.
Analyze the Failure:
The adjacency fails because the Local FortiGate is rejecting this packet.
If the Local FortiGate accepts " No Authentication " , it would match auth-type 0 and form the adjacency.
Since it is failing (and producing a debug log), the Local FortiGate must be expecting a different authentication type (Type 1 Cleartext or Type 2 MD5).
Evaluate the Options:
A). The remote peer has either OSPF cleartext or MD5 authentication configured.
Incorrect. The debug shows auth-type 0 (No Auth) coming from the remote peer.
B). There is an OSPF authentication configuration mismatch.
Correct. One side is sending " No Auth " (Remote), and the other expects " Auth " (Local). This is a definition of a mismatch.
C). The local FortiGate does not have OSPF authentication configured.
Incorrect. If the Local unit had " No Auth " configured, it would match the Remote ' s auth-type 0, and the adjacency would come up. The failure implies the Local unit does have auth configured.
D). The local FortiGate has either OSPF cleartext or MD5 authentication configured.
Correct. Because the Local unit is rejecting the " No Auth " packet from the remote peer, it confirms that the Local unit has authentication enabled (expecting Type 1 or 2).
Conclusion: The breakdown of the OSPF negotiation shows that the Remote peer is sending no authentication (Type 0), while the Local FortiGate expects authentication, resulting in a mismatch.
Reference:
FortiGate Security 7.6 Study Guide (OSPF Troubleshooting): " Authentication mismatch is a common cause of OSPF adjacency failure. Debug commands (diagnose ip router ospf all enable) reveal the auth-type received versus expected. " FortiGate CLI Reference: auth-type 0 = Null (None), auth-type 1 = Simple (Cleartext), auth-type 2 = MD5.