Re: Single point of failure?
"I thought the whole point of TCP/IP and routing traffic was to automatically re-route traffic via an alternate path in the event of a link failure."
Unfortunately, budgets often don't allow enough capacity for full redundancy at most service providers. And even when their is redundancy, often the redundant links are "folded", meaning they follow the same route. And the other repair that was in process was probably one of their other major links.