RED 50 Manually Trigger Failover to WAN 2

HI.

We have a RED 50 on a remote site, used primarily for remote monitoring and data upload.

The primary uplink is ADSL (low speed, 4Mb d/l, 380 Kb u/l), low latency and no data cap.

The back up uplink is satellite (speeds of c 20 Mb d/l and u/l, high latency  and limited data cap (3.5 Gb a month u/l).

Data wise we primarily upload from the site, so upload consistency is important.

For the last few nights, our local software has been losing connection with the remote site. As the primary broadband connection has not dropped, the RED box never failed over to the backup uplink.  We resolved the issue by disconnecting the primary uplink, resulting in the RED box failing over to WAN 2, and everything was fine. It would appear that the primary uplink performance (particularly its already low upload) is been degraded, but since it never disconnects, is not triggering a failover.

Later in the day we reconnected the primary uplink, the RED box reverted to WAN 1 and everything worked fine. Until the issue repeated in the night.

We tried changing the backup uplink from failover to balancing, but that had no noticeable effect. (The difference in latency and performance between primary and backup may make balance loading unworkable). Finally an engineer had to travel onsite at 01:00 to manually disconnect the primary uplink. When this occurred, the box failed over to WAN 2 and everything settled down.

All evidence points to an issue with the primary uplink, and we've engaged with the ISP to investigate and fix that. 

What we would like to be able to do if this issue reoccurs (here or elsewhere) is trigger a manual failover remotely. Examining the RED management screens in the UTM, we have not been able to find an obvious way to do this.

One suggested solution would be to change the (primary) uplink from DHCP to Static Address (with a non-function IP address) - this should cause the RED box to lose connection with the primary uplink and failover to the backup.

Would this approach work as a work around to manually trigger a failover?

Thanks,

Noel