FortiGate HA is compatible with DHCP and PPPoE but care should be taken when configuring a cluster that includes a FortiGate interface configured to get its IP address with DHCP or PPPoE. Fortinet recommends that you turn on DHCP or PPPoE addressing for an interface after the cluster has been configured. If an interface is configured for DHCP or PPPoE, turning on high availability may result in the interface receiving an incorrect address or not being able to connect to the DHCP or PPPoE server correctly.
|You cannot switch to operate in HA mode if one or more FortiGate interfaces is configured as a PPTP or L2TP client.|
You can configure a cluster to act as a DHCP server or a DHCP relay agent. In both active-passive and active-active clusters DHCP relay sessions are always handled by the primary unit. It is possible that a DHCP relay session could be interrupted by a failover. If this occurs the DHCP relay session is not resumed after the failover and the DHCP client may have to repeat the DHCP request.
When a cluster is operating as a DHCP server the primary unit responds to all DHCP requests and maintains the DHCP server address lease database. The cluster also dynamically synchronizes the DHCP server address lease database to the subordinate units. If a failover occurs, the new primary unit will have an up‑to-date DHCP server address lease database. Synchronizing the DHCP address lease database prevents the new primary unit from responding incorrectly to new DHCP requests after a failover.
Also, it is possible that when FortiGates first negotiate to form a cluster that a unit that ends up as a subordinate unit in the cluster will have information in its DHCP address lease database that the cluster unit operating as the primary unit does note have. This can happen if a FortiGate responds to DHCP requests while operating as a standalone unit and then when the cluster is formed this unit becomes a subordinate unit. Because of this possibility, after a cluster is formed the DHCP address lease databases of all of the cluster units are merged into one database which is then synchronized to all cluster units.