On Friday, November 13, 2026, the FAA ordered a nationwide ground stop after disruptions to aircraft communications and radar feeds. Routine daytime work was underway at a remote FAA site as part of a move from copper telecommunications lines to fiber. The FAA identified that site and a second facility as possible sources of a surge of roughly a thousand connection attempts every four to five seconds. Older equipment on the network became unstable. The FAA said it would reconsider when it schedules this kind of work. What triggered the surge remains under investigation.
FAA ground-stop Q&A · Al Jazeera reporting on the copper-to-fiber work
I would start with the people who set the objective. The scope should say what service must continue through the cutover and what additional traffic the network is expected to carry afterward. If the ambition is a screen that shows every flight, everywhere, as it moves, work backward from that picture. How much data has to arrive, how quickly, and what happens when several sites try to reconnect at once? Decide what must work on the day of the switch and what can wait for a later phase.
Start by reading the router manuals. Pull the manuals for the exact models already in service and the ones being installed. Read the rated throughput, the published limits on new connections, and what the manufacturer says happens at those limits. Check the software versions and settings actually in use. Then follow the connections to the equipment that receives them. Where the manuals leave a question open, test it. After the disruption, the FAA says a broadcasting function was disabled on the new routers as a precaution. That is one example of why the settings belong in the review. FAA ground-stop Q&A
I’d go back fifty years if that is what it takes to understand the network we inherited. A line gets replaced. An application moves. A workaround becomes the way people do the job. Each team may know its own part well. Where is the record that shows how those parts work together, what each change assumed, and which assumptions still hold? Finding that out belongs in the scope and budget of the work.
The older equipment was part of the system the project team agreed to change. Its limits belonged in the requirements. Whoever approved the switch should be able to show what the network was tested against, what it handled, and what the team was prepared to do if the change disrupted service. Decide what would make you stop and how you would restore service before you begin.
Give me the objective, the manuals, the test results, and the decision that authorized the switch. I’d start with the people who commissioned the work and ask what they required the team to prove before touching a live circuit. That is the lesson I took from the sergeant major. If you re-engineer mush, all you have is re-engineered mush.
Philip Andreae · 16 November 2026
Swirl review · Philip set the intent · ChatGPT drafted and revised · Claude checked facts and sources · Gemini reviewed authenticity and challenged the argument
