Minimum Setup for Hot Failover

I’ve got a BR1 Max router here that I’m sending a video stream though. It is important to keep the stream alive without interruption so I’m looking into a way to do hot failover to 4G in case my WAN goes out.

At the moment, I do not have SpeedFusion enabled. When the WAN goes out, the 4G enables, but my stream is down for around 30 seconds until my computer can restore an internet connection.

I’ve been looking at this video: https://www.peplink.com/technology/speedfusion-bonding-technology/

In the video, they have what appears to be a fairly simple setup: one computer connected to a Peplink router, one WAN, and a 4G connection. They claim to be have SpeedFusion enabled to get the hot failover that they are demonstrating.

Now my question… Whenever I’ve looked into using SpeedFusion, it appears that I need to set up a VPN that includes either a second router in a remote location or I need to use FusionHub software on a cloud server. Is this true? The video sure didn’t show any of that business.

Is it possible to get hot failover using just one router, one WAN, and a 4G connection? If not, what is the simplest way to get hot failover and what are they not showing in that video?

Thanks,

Ron

Just want to say that I spent last night setting up a cloud server running FusionHub and the results are pretty amazing - much like in that video demonstration. I’ve been a bit miffed at having to set all of this up since that video seemed to imply a much simpler setup, but given the results that I’m getting, I’m coming around to it.

I was worried that all of my network traffic would have to pass through the cloud server, making my cloud server bandwidth fees very high. So far, I see very little traffic going in and out of the FusionHub cloud server even though I’m sending considerable data to the internet from one of the connected computers. Its a bit strange to me and kind of makes me wonder what this FusionHub bit is actually doing, but it is enabling the hot failover to happen, so I guess it has value somehow.

Ron

Are you sending all traffic from the BR1 through the VPN? This can be done using the “Send All Traffic To” feature under: Advanced > SpeedFusion. Alternatively, an outbound policy rule can be configured to specify certain or all traffic to send through the FusionHub.

1 Like

Hello Ron,

No, I don’t have “Send All Traffic To” the VPN set up. Should I? It kind of seems like I don’t want that, so longs as hot failover still happens. If I send all traffic through the VPN, then I would expect huge bandwidth fees from the cloud server host.

Thanks,

Ron

Yes, that video was misleading to me too, and it took some extra investigation to realize I needed a Fusionhub solo or additional device for Speedfusion technology to work. I wish they would fix the video to make that clear. What cloud provider are you using for the hub?

I managed to get things working with DigitalOcean. I was uncertain about how well this would be supported going into it, but it seems fine. Last year, DigitalOcean added a feature to upload disk images so all I needed to do was to upload the .vmdk file and create a server from that image. It went very smoothly.

One thing that I’m a bit unclear of though… My intent is to use FusionHub Solo rather than the full blown Fusion Hub. The Peplink webpages do not make it very clear if FusionHub Solo uses the same install binary as FusionHub. The “How to Download FushionHub Solo” section in the FAQ merely tell you how to get a license key but do not speak of downloading the binary at all. So I installed FusionHub using my FusionHub Solo license key and I got no complaints about anything. But… here’s the thing… In my inControl2 panel, the hub shows up as FusionHub (not FusionHub Solo) and I have a license expiration date one year from now. My understanding is that FusionHub Solo is a perpetual license. Not sure what’s going on there or what may happen next year when my license is supposedly expired.

1 Like