NSE6_SDW_AD-7.6 Fortinet NSE 6 - SD-WAN 7.6 Enterprise Administrator Questions and Answers

FortiManager enforces a strict distinction:

"Firewall policies must reference SD-WAN zones, not individual SD-WAN members, when used in conjunction with SD-WAN templates. Attempting to install a policy that references a specific member (interface) will result in a deployment error, as member-level targeting is not supported in SD-WAN policy abstraction. This enforces centralized policy consistency and proper SD-WAN operation."

Ensuring policies target zones allows FortiGate to dynamically select the optimal member.

When the member priority of a port is increased (e.g., port2 to 20), FortiGate evaluates existing sessions and applies “dirty” flags where applicable. The SD-WAN session management mechanism is described in detail: “Upon a change in SD-WAN member priority, all existing sessions using that member are marked as dirty. For SNAT sessions, the gateway information is updated to ensure future packets are routed through the newly preferred member, in this case, port1. This automatic re-evaluation allows SD-WAN to dynamically respond to topology or priority changes, maintaining optimal routing.” This is fundamental to seamless failover and session persistence in Fortinet SD-WAN, ensuring active flows are redirected based on updated priorities or health status.

To allow spokes in different hub-and-spoke groups to establish ADVPN shortcuts, the hubs must be configured to forward and send ADVPN shortcut offers. The key required settings on the hub are auto-discovery-forwarder (for VPNs to hubs) and auto-discovery-sender (for VPNs to spokes). This ensures the hub can facilitate and advertise ADVPN shortcut offers between spokes.

With the Best Quality strategy, FortiGate first checks which links meet the SLA targets defined in the selected performance SLA (in this case, Custom-profile-1). Among those qualified links, FortiGate then uses the member configuration order to decide preference. If multiple links still qualify, it finally considers the member local cost to select the best path.

The exhibit shows the runtime details of two SD-WAN services (rules):

Service(2)

Mode(manual hash-mode=inbandwidth)

Members(2): port2 (WAN2), port1 (WAN1)

Application matching: Facebook, LinkedIn, Game

Source: 10.0.1.0–10.0.1.255This rule is clearly intended for internet/DIA application steering and does not show a corporate destination range.

Service(3)

Mode(sla hash-mode=round-robin)

Members(6): HUB1-VPN1/2/3 and HUB2-VPN1/2/3

Source: 10.0.1.0–10.0.1.255

Destination: 10.0.0.0–10.255.255.255

Traffic from 10.0.1.124 to 10.0.0.254 matches Service(3) because the destination IP 10.0.0.254 falls within the destination range 10.0.0.0–10.255.255.255.

Within Service(3), the member list shows SLA results per interface:

HUB1-VPN2 has sla(0x1) and num of pass(1)

HUB2-VPN2 has sla(0x2) and num of pass(1)

The remaining members (HUB1-VPN1, HUB2-VPN1, HUB1-VPN3, HUB2-VPN3) show sla(0x0) and num of pass(0)

This indicates that, for Service(3), only HUB1-VPN2 and HUB2-VPN2 are currently meeting the SLA requirements (passing), and because the rule uses hash-mode=round-robin, FortiGate load-balances sessions across the passing members.

Therefore, FortiGate will steer the traffic using HUB1-VPN2 or HUB2-VPN2, which corresponds to Option B.

In the SD-WAN GUI, the absence of members in a zone is visually represented, and the Fortinet guide confirms:

"If a zone such as overlay-factories contains no members, it will be displayed as empty in the SD-WAN GUI. This may occur when the zone is reserved for future expansion, or if members have been temporarily removed for maintenance or reconfiguration. Traffic cannot be steered via an empty zone until at least one SD-WAN member is added."

Such visual cues help operators quickly assess configuration status and readiness.

For SD-WAN deployments that span multiple regions—where most traffic is intra-region and there is no requirement for inter-region ADVPN—the best practice is to use IBGP with BGP on loopback interfaces for routing within each region and EBGP between the regions. This approach ensures robust and scalable routing, isolates regional routing domains, and enables policy control at region boundaries. BGP on loopback is preferred for its reliability and flexibility, as it enables peering that is not tied to specific physical interfaces. EBGP between regions allows each region to maintain independent routing policies and summarization, optimizing performance and manageability. By separating IBGP (intra-region) and EBGP (inter-region), you create a modular architecture that scales easily and simplifies fault isolation and troubleshooting.

The use of SLA targets is specific to certain SD-WAN strategies. The "Lowest Cost (SLA)" and "Maximize Bandwidth (SLA)" strategies are explicitly designed to use the configured SLA targets to make routing decisions. The "Best Quality" strategy uses performance metrics but does not necessarily require or reference SLA targets in the same way, while "Manual" does not use metrics at all for path selection.

This is a core function of SD-WAN rules with SLA targets. The purpose of configuring an SLA target with specific thresholds for latency, jitter, and packet loss is to define what is considered "acceptable" performance for an application. SD-WAN rules then use these targets to check if the members (interfaces) meet these requirements before a flow is steered over them, ensuring that a preferred path still offers a good user experience.

FortiGate allows for a single SD-WAN rule to reference multiple, different performance SLAs. This is crucial for complex deployments where a single SD-WAN rule needs to handle traffic for multiple applications that have distinct performance requirements. For example, a single rule might direct VoIP traffic based on one performance SLA with strict latency/jitter targets, while simultaneously handling general web traffic using another performance SLA with more lenient requirements.

When SD-WAN load balancing is required with link quality awareness, FortiOS relies on SLA-based strategies. These strategies evaluate link performance using performance SLAs (latency, jitter, packet loss, MOS) and then make forwarding decisions accordingly.

Option A is correct.

In FortiOS 7.6, when an SLA-based SD-WAN rule has load balancing enabled, FortiGate distributes traffic only across the members that meet the SLA targets. Any member that is out of SLA is excluded from load balancing. This behavior ensures that traffic is not forwarded over degraded links while still allowing load distribution across healthy paths.

Option C is correct.

The lowest cost (SLA) strategy is an SLA-based strategy that considers link quality while also allowing SD-WAN load balancing. When multiple members meet the SLA requirements and have equal cost, FortiGate can load balance traffic across them using the configured hash mode. This makes the lowest cost SLA strategy suitable when both link quality and load balancing are required.

Why the other options are incorrect:

Option B is incorrect because the best quality strategy is designed to select the single best-performing link based on SLA metrics. It does not support SD-WAN load balancing across multiple links.

Option D is incorrect because the best quality strategy does not support load balancing at all, so the statement about round-robin hash mode is invalid.

Therefore, the two correct facts to consider are A and C.

In an SD-WAN rule, you can steer application traffic by using Internet Service Database (ISDB) entries. Facebook and LinkedIn are predefined ISDB objects in FortiGate, so the correct way is to select them in the Internet service field under Destination. This ensures that all traffic to these applications is matched and routed through the chosen (less costly) link.

When referencing a zone in a static route, FortiGate’s behavior is described as:

"Referencing a zone in a static route causes FortiGate to install a static route for each member interface of the zone. This enables ECMP (Equal-Cost Multi-Path) and load balancing where supported and ensures that traffic can be steered over any valid zone member according to SD-WAN rules or standard routing."

This mechanism is fundamental to Fortinet’s implementation of SD-WAN and simplifies large, multi-interface deployments.

The hub must use a performance SLA with the same criteria as the spoke's health check. The spoke's health check is using ping (protocol ping) and measuring latency (link-cost-factor latency). For the hub to use the data sent by the spoke, its performance SLA must be configured to measure the same metrics. If the hub is looking for jitter or packet loss, it will not use the latency data sent by the spoke.

When a spoke sends embedded health data, the hub FortiGate must be configured to receive and use it. This is done by setting set embedded-measure accept within the performance SLA configuration on the hub. This setting explicitly tells the hub to trust and use the performance metrics received from the remote FortiGate (the spoke). Without this setting, the hub will likely ignore the embedded health data and rely on its own health checks, which could lead to incorrect traffic prioritization.

Fortinet outlines key SD-WAN routing principles:

"Policy routes are always evaluated before SD-WAN rules, meaning if a policy route matches, SD-WAN steering is bypassed. If the best route for a destination is not via an SD-WAN member, SD-WAN rules do not apply, and members are ignored if they lack a valid route. This hierarchy ensures traffic always follows the most deterministic and valid path according to configuration."

Understanding these principles is critical for correct SD-WAN and routing integration.

In FortiOS 7.6, static routes can reference either:

an SD-WAN zone (for example, virtual-wan-link or a user-defined SD-WAN zone), or

a specific SD-WAN member interface that belongs to that zone.

However, FortiOS enforces a routing constraint to avoid ambiguity during route resolution. Two static routes cannot have the same destination prefix if one points to an SD-WAN zone and the other points to an SD-WAN member within that zone. This would create an overlapping and conflicting forwarding decision.

Therefore, if you configure:

one static route that references an SD-WAN zone, and

another static route that references an SD-WAN member belonging to that same zone,

the destination subnets of the two static routes must be different.

Why the other options are incorrect:

Option A is incorrect because FortiOS does allow static routes that reference individual SD-WAN members.

Option B is incorrect because static routes can reference SD-WAN zones.

Option D is incorrect because static routing decisions in FortiOS are based on destination prefixes, not source prefixes.

Thus, the correct answer is C.

When asymmetric routing is observed (such as reply traffic not following the optimal path), the solution is:

"The auxiliary-session feature, enabled under config system settings, allows FortiGate to consider multiple egress interfaces for reply traffic, not just the original ingress interface. This is crucial for SD-WAN environments where the best path may differ between forward and return directions, especially when performance or policy rules are dynamically evaluated."

Activating this ensures reply traffic is always sent on the member with the best real-time metrics.

Application-based SD-WAN rules enable intelligent traffic steering. The guide specifies:

"If a flow is identified as belonging to a defined application category (such as social media), FortiGate will match it to the corresponding service rule (rule 2) and route it through the specified interface, such as port2. However, if the application is not recognized during the session setup, the system defaults to load balancing the traffic using the available tunnels according to the policy for unclassified traffic, ensuring continuous connectivity while waiting for application classification."

This guarantees both performance and resilience.

The implicit SD-WAN rule serves as the final catch-all. Per Fortinet:

"Sessions matching the implicit SD-WAN rule do not have an SD-WAN service id, as they are not associated with any specific user-defined SD-WAN rule. Additionally, this occurs only when traffic fails to match any entry in the policy route table. This default handling guarantees connectivity while minimizing the risk of blackholed traffic."

Administrators can observe this in diagnostic outputs for troubleshooting.

The hub must send additional paths to spokes (set additional-path send).

The hub must treat each spoke as a route-reflector client so spoke routes are reflected to other spokes.

The hub must specify how many additional paths to advertise (set adv-additional-path ).