Question 1
AdvancedSecurity Architecture · Chassis solutions 6000/7000 modules and architecture
A large enterprise is deploying a standalone FortiGate 7000 series chassis in a high-traffic data center, using Session-Aware Load Balancing (SLBC).
During a failure test, one of the processing modules (FPM) fails. The administrator notices that while new sessions are distributed to the remaining FPMs, a subset of long-lived TCP sessions was dropped and had to be re-established by the clients.
Based on the architecture of the 7000 series, what is the most likely cause of this behavior?
Answer and explanation
Correct answer: C
SLBC sends each session to one FPM, and that FPM holds the session state. FPMs in the same chassis do not synchronize sessions with each other, so when an FPM fails, the sessions it was processing are lost and the clients must reconnect. Only an FGCP HA cluster of two FortiGate 7000 chassis with session-pickup enabled keeps such sessions: the primary syncs its sessions to the secondary chassis, and an FPM failure makes the chassis with more working FPMs the primary.
