I am wondering if there is a way within the Balance 20/30 series to direct specific machines to use specific WAN’s. I have a situation where my client has two internet connections. We are directing FTP traffic to a specific WAN and we want our server to ONLY utilize that WAN so FTP connections can work properly without the server round-robining between the two WAN connections. I see there are INBOUND rules but I’m not sure if this the correct area and I’m not sure the Balance 20/30 can even do this. It would be great because we could then route other clients machines to utilize a specific WAN.
We can use outbound policy to control which server or what type of traffic should go out on which link. Return traffic will then come back on the link it goes out.
If the FTP session is established on WAN1 Peplink will always send this FTP traffic to WAN1. I am not 100% clear of what you are asking but does this help?
Appreciate the response. What I’m trying to do is to make sure a particular work station ONLY uses a particular WAN. What I’m finding if I run a public IP check is that it keeps changing between one WAN and the other. This means if an outside FTP request is made and the server hosting the service happens to be getting it’s connection from the wrong WAN there will be no response. the setup is:
WAN1: Cable Internet
Wan2: DSL
I want said server tonly be fed by WAN2.
This is no problem. A simple outbound policy rule is all that is needed and will take around 10 seconds to setup. 
And that simple rule would be?
Under the Network settings, click on Outbound Policy and make sure it is set to “Managed by Custom Rules”. Make sure to leave the two default rules in place, “HTTPS Persistence” and “Default”. You can then simply add a new rule to your liking. Here is a screenshot example of a rule to enforce traffic from a LAN client to a specific WAN:
You can add as many rules as you want to and after clicking Save always remember to click “Apply Changes” at the top.
Hope this helps…
Much appreciated. I wonder what the rule is for when a client station moves between the connection without a rule. Anyway, thank you for your insight.
I’m not sure I understand your question, if you can be more specific I can try to help answer.
I’ve noticed that without a rule any given client station will access one WAN connection or an another when there is more than one and I was curious how that is determined. When I’ve run a test such as going to whatismyip.com on a client one moment I will get the public IP of one Wan and then the IP of the second Wan.
The “Default” outbound policy rule of “Auto” uses a simple weighted balance algorithm that is based on the downstream bandwidth setting of each individual WAN:
This should be set to the actual speed of the line on the WAN setup page:
In my case, WAN1 is 50Mbps and WAN2 is 100Mbps. The “Auto” rule will send 2/3 of the traffic out WAN2 and the other 1/3 out WAN1. If I go to whatsmyip three times, two of them will show that I am coming from WAN2 and one will show WAN1.
Of course if you don’t want to use the “Auto” rule, you can simply change it to “Custom” and make your own default rule.
Hope this helps…