BGP Process During HA (Active-Standby) Failover

Hi,

I operate two FortiGate 120G devices in an Active-Standby HA configuration with a BGP connection.
When an Active-Standby failover occurs, there is a  BGP interruption of approximately 30 seconds.

The following KB article states that the BGP process only runs on the daemon process of the active unit, and that a new process is launched upon failover.
Could you please provide detailed documentation or evidence regarding this point?

KB
https://community.fortinet.com/t5/FortiGate/Technical-Tip-Configuring-FortiGate-HA-and-BGP-graceful-restart/ta-p/196150
On a FortiGate HA cluster, the BGP router daemon process is only running on the Primary unit. When there is an HA failover a new BGP process will be launched on the newly elected primary.

Additionally, I observed a outage of 25-30 seconds during failover.
I understand this is not related to the BGP keepalive timer (30 seconds) but is simply the time required for the BGP process to launch and establish a connection on the new primary unit.
I would appreciate it if you could point out any misunderstandings regarding this point.

Hardware: FortiGate 120G
OS: v7.4.8 build2795

Thank you
Kenji