The problem first appeared at the Indianapolis Center, where staff had trouble communicating with flight crews. The FAA traced it to two sources: a radio site in El Dorado, Arkansas and a radar site in Amarillo, Texas, where new routers had begun sending a high volume of data across the network. FAA technology chief Rebecca Guy said they were generating about a thousand connections every four to five seconds.
A modern network would handle that load easily, but the FAA still relies on a decades-old copper network that is being replaced with fiber. A specific router feature was eventually found to cause the overload, and FAA staff switched it off on all Juniper routers across the network. It took about four hours to bring the network back to normal.
FAA Administrator Bryan Bedford said the agency is examining whether it was a faulty installation, a faulty configuration or bad equipment. An HPE spokesperson said the routers "did not fail" and worked as intended. On-board safety systems such as TCAS were not affected. It is only the third nationwide ground stop the FAA has ordered, after 11 September 2001 and the NOTAM system outage in January 2023.