T Techclick ← All lessons
Fortinet · FortiOS 7.4 · Lesson 9 of 12

SD-WAN: SLA steers after the member is alive

Two WANs. You want Office 365 on the good link. A static policy-route to WAN1 does not measure packet loss. SD-WAN members + SLA health-check + an SD-WAN rule does. If you skip the health-check, you built a pretty policy route.

Updated 2026-08-18·18 min read·L2 primary·Quiz at end

After this page you can

The ticket

WAN1 is 8% loss. Users still prefer it because the rule is “manual WAN1.” SLA was never attached.

Quick interview answer

FortiOS 7.4 SD-WAN: put WAN interfaces in the SD-WAN zone (virtual-wan-link). Create a Performance SLA (health-check) that pings/HTTP a target. Create an SD-WAN rule (service) that matches apps/dest and strategy (lowest cost, best quality, SLA target). Firewall policy destination interface becomes the SD-WAN zone, not a single WAN. Policy-route is static; SLA is measured.

Hero · two WANs
Two WAN links with an SLA probe choosing the healthy path
If you do not probe, you are not doing SD-WAN.
Lab data · dummy only

FortiGate fgt-hq mgmt 10.10.10.1 · WAN 203.0.113.10 · LAN 10.20.30.0/24 · FortiManager 10.10.10.5 · FortiAnalyzer 10.10.10.6 · Priya 10.20.30.80 · branch peer WAN 198.51.100.10 · Azure VPN GW public 203.0.113.50 · AWS VGW public 203.0.113.60. RFC 5737. Not a customer.

Mental model

  1. Members = the cables.
  2. SLA = the thermometer.
  3. Rule = who uses which thermometer.
  4. Firewall policy = still first match, dest-intf = SD-WAN zone.
https://fgt-hq.techclick-lab.in
Training mock · not live
DashboardPolicyVPNNetwork
Network → SD-WAN → Performance SLAs

SLA-Internet

Ping
1.1.1.1
WAN1, WAN2
latency 100ms loss 2%
Rule 1 — default / SaaS
FortiOS 7.4 SD-WAN Performance SLA. Training mock.

How to choose

NeedTool
Always this WAN, no measurePolicy route / manual strategy
Prefer cheapest if healthySLA + lowest cost
Voice on best qualitySLA + best quality / sla-target

Runbook

  1. Side A — members

    WAN1 203.0.113.10, WAN2 203.0.113.18 in SD-WAN zone.

  2. Side B — SLA + rule

    Ping 1.1.1.1 from both. Rule: dest all, strategy Lowest Cost (SLA), interface preference WAN2 then WAN1.

  3. Side C — policy

    LAN → SD-WAN zone ACCEPT + NAT. Not LAN→WAN1 only.

Is the SLA alive?
diagnose sys sdwan health-check
# SLA-Internet  1.1.1.1
#   WAN1: packet-loss=8%  latency=40  status: dead
#   WAN2: packet-loss=0%  latency=18  status: alive

diagnose sys sdwan service
# service 1  sla mode  members: WAN2

Four failures

1 · Policy still to WAN1

SD-WAN rules never see the packet.

2 · SLA with no probe reachability

Both members dead — traffic blackholes if you required SLA.

3 · Health-check via the wrong member

Probe must be sourced per member.

4 · Confusing SD-WAN rule with firewall policy

Both exist. Rule steers. Policy allows.

How to prove it

Close only when

1) Health-check shows WAN1 dead / WAN2 alive. 2) Session egress is WAN2. 3) Firewall policy dest-intf is the SD-WAN zone.

Traps

Looks like SD-WANIs actually
Two defaults, no SLAECMP / policy route
Green members, users on the bad linkRule strategy = manual

Knowledge check

Judgment items. One best answer. Reasons send you back to the matching section.

Q1

SD-WAN SLA is…

Correct: a. Concept.
Q2

Firewall policy for SD-WAN users should dest-intf…

Correct: b. Failure 1.
Q3

Both SLAs dead and strategy requires SLA. Risk?

Correct: a. Failure 2.
Q4

Policy route vs SD-WAN SLA?

Correct: a. Interview.
Q5

Proof WAN1 is unused because of loss?

Correct: a. Runbook.
Q6

SD-WAN rule vs firewall policy?

Correct: a. Failure 4.

FortiGate class series: FGT / FMG / FAZ · First day · Policy first match · SNAT vs VIP · Profiles + SSL · debug flow · IPsec S2S · SSL-VPN vs RA · SD-WAN SLA · VDOM · FGCP HA · Interview

Sources

Related: FortiGate session factory · VPN series.