Fortigate SD-WAN
Hello Guys,
I hope you're all doing well.
I'm fairly new to Fortinet's SD-WAN and recently tried to do a deep dive into it.
So I set up a lab with fortimanger and 3 fortigate, one as a hub and the other two as spoke.
The aim is to understand how it really works and to set up all the configuration using a CLI model without using “orchestration overlay” in order to understand all the steps.
I've read a lot of different articles on Fortinet documentation and other websites and I've seen a lot of different configurations/design depending on the Forti OS version.
I'm really struggling to understand the different designs and the best mode to use depending on the situation, and also how certain concepts work. So I hope to find my answers here...
Since the 7.0.x version :
=> BGP on loopback + ADVPN 2.0 (RR-less) :

=> BGP on loopback + ADVPN 2.0 (RR) :

Before the 7.0.x version :
=> BGP on overlay interface (VTI) + ADVPN ?

So here are my questions :
BGP on loopback + ADVPN 2.0 :
-Do you validate the both design ? One without BGP RR and the other with BGP RR.
BGP on overlay interface (VTI) + ADVPN :
-Is the route reflector necessary?
-How are shortcuts created between Spoke when using ADVPN? How are VTIs communicated between the 2 spokes? How are LAN prefixes notified between the 2 spokes?
Use cases between the 2 modes (BGP on VTI or BGP on loopback):
-If I've understood correctly (or not) it would be more interesting to use the “BGP on loopback” mode if spoke to spoke communications are necessary? Also, this may prevent the HUB from having too many BGP neighbors. Are there any other cases where it would be preferable to use one mode rather than another?
Thanks a lot!
Regards
