Lately the Scheduler appears to be having major problems whenever there is heavier than usual network traffic. The usual response to reporting or requesting work is "Scheduler request failed: Timeout was reached".
Routing the Scheduler traffic to the campus network (the one used by the web site & forums) may help with this issue.
At it's heaviest, inbound traffic is around 20Mb/s. During the outages when the Scheduler is down, but uploads are still possible the inbound traffic is generally around 8Mb/s.
So that would shift roughly 12MB/s of traffic from the present saturated link, it wouldn't overwhelm the campus network, & it would allow people to report & at least get new work allocated without difficulty.