Trust to Untrust Policy Question

Hi, and welcome to the Forums! Coming from Netscreen/Juniper, eh? You' re right, a policy that wide open covers all traffic from internal to WAN (this is FortiSpeak). The destination ALL or ' 0.0.0.0' stands for all networks and is most often used on WAN interfaces. You' ll find yourself working with interfaces a lot more than you' re used to. But, there is a zone construct in FortiOS as well. It combines several interfaces so that it can replace multiple policies with just one. You can allow or deny intra-zone traffic. If you want to create a zone, you have to do that before referencing the member interfaces elsewhere. Common scenario for a zone: combine multiple VPN tunnel ends into a VPN zone for a hub-and-spoke VPN.

When I see trust and untrust I too think " netscreen" since I' m a netscreen kinda of guy and been working with them years before fortigates. If that' s all of the policies that you need, then you are correct that you don' t need anything else. A lot of organizations are restricting what they are allowing in out, so keep that in mind. If you need this kinda of policing, then create a service list of the service that you want to trust outbound. You can create this quite easily thru the webUI by selection of services.