Entrada de Pandipedia
Actualitzat el 21 Nov 202621 fontsExplora Pandipedia
Informe de recerca

What caused the FAA ground stop on 13 November 2026?

No root cause has been published. The FAA says data flooded its telecommunications network “at a rate of 1,000 connection attempts every 4 to 5 seconds”. It names two possible sources: a radar site in Amarillo, Texas, and a radio site in El Dorado, Arkansas, where its contractor L3Harris was deploying a new router.[1] The paper trail is longer. The same network, under the same contractor, failed nationwide in 2009 after a router change,[2] and auditors warned two months ago that, without a single schedule for its modernisation, the FAA would stay reliant on failure-prone systems, with more outages and ground stops.[3]

Last updated 20 November 2026. This page covers the FAA's paper trail only: what was written down before and after the ground stop, linked to where it was read.

The afternoon in the FAA's own log

Time (ET)FAA Command Center advisoryIts wording
3:10 pmGround stop for Chicago O'Hare begins (advisory 107, sent 3:28 pm)[4]“EQUIPMENT / OUTAGE”, “COMMS ISSUES”
3:54 pmNational ground stop (advisory 111)[5]“ALL AIRPORTS ARE GROUND STOPPED”
8:19 pmNationwide ground stop cancelled (advisory 005)[6]“HAS BEEN CANCELLED”

Times are converted from the log's UTC stamps. The national order lasted 4 h 25 min.

What we know

  • The trigger was contractor work on the old network. The FAA calls it “a routine replacement at a remote facility”, done in the daytime.[7] Duffy says the router “started to broadcast into our network” and overloaded it.[8]
  • No official blames an attack. The FAA has “no indication that this was a cyberattack”.[9] Duffy asked the FBI to join the examination of the router as a precaution.[10]
  • This network has failed this way before. In 2009 a router change by the contractor, then Harris Corporation, blocked most of its routes for five hours.[11]
  • The network was due to be replaced. The FAA chose a new provider in March 2023.[12] Duffy said on 14 November: “I'm stuck with L3Harris, who owns this network.”[13]

The paper trail before 13 November

One row per document. “As reported” means it was read in a news report, not at source.

WhenDocumentWhat it says
27 Apr 2006DOT Inspector General audit AV-2006-047[14]The network (FTI) is run by Harris under a July 2002 contract. In October 2005 FTI equipment failed and cut radar data to O'Hare's controllers; no backup was available because nobody had tested that one existed.
17 Jun 2010DOT Inspector General letter CC-2010-012[15]On 19 November 2009 a Harris engineer put a wrongly configured router into service during a network transition. 129 facilities were hit and about 75% of routes were blocked for 5 hours “while Harris tried to find the problem”. The FAA “was unaware” of the errors. Both said a risk of critical outages remained.
Jan 2023FAA statement on the NOTAM outage[16]“Contract personnel unintentionally deleted files.” It led to the first nationwide ground stop since 11 September 2001.[17]
2023 to 2026Replacement contract, as reported[18][19]The FTI contract was due to expire in April 2023. The FAA picked Verizon to replace it in March 2023, then made AT&T the lead in August 2026. Target: end of 2028.
23 Sep 2024GAO-24-107001[20]51 of the FAA's 138 air traffic control systems were unsustainable and 54 potentially unsustainable.
Sep 2025FAA statement on a Dallas ground stop, as reported[21]A carrier's data service failed several times, and an oversight by L3Harris meant the backups did not work properly.
14 Sep 2026GAO-26-107992[22]The telecom network's technology “is no longer supported by the communications providers”. The FAA has 11,389 project schedules and no integrated one, risking “more system outages, ground stops, flight delays”. The department said it has such a schedule; GAO disagreed.
Sep 2026FAA Administrator, as reported[23]A primary circuit at the Philadelphia radar facility failed and the backup fibre turned out to be cut.

The accounts since 13 November

WhenWhoWhat they say about the cause
13 NovFAA Command Center log[24][25]Equipment outage and “comms issues”. No cause.
14 NovSecretary Duffy and Administrator Bedford[26]Duffy: either an installer error by L3Harris or a faulty router, not yet known. Bedford: a “defective piece of equipment”.
14 NovL3Harris[27]No cause. It is “supporting the FAA in its investigation” and has converted more than 70% of the copper lines.
15 NovPASS, the FAA technicians' union[28]Two contractor-owned routers were being updated mid-day. Its technicians “were not involved in the upgrades”.
15 Nov, updated 17 Nov[29]FAA questions and answers[30]A data flood from two possible sites. The function was disabled on the “originating Juniper routers” and on more than 60 others. Installation is paused.
16 NovHewlett Packard Enterprise, owner of Juniper[31]Says “our equipment was working as configured”.
17 Nov13 Senate Democrats, letter as reported[32]Citing L3Harris: 3,648 routers are to be converted, 400 are installed and 69 are operating. They asked for a pause.
17 NovDuffy at the Senate, DOT release[33]It was the 69th installation of that router; 68 worked. The fault has been recreated in an L3Harris lab, but “the root cause of what happened, we don't have clarity yet”.

Where the official accounts conflict

  • When it began. The FAA's page says the problem began at “approximately 3:30 p.m. ET”.[34] Its log shows a ground stop for O'Hare from 3:10 pm, citing “COMMS ISSUES”,[35] and the national order at 3:54 pm.[36] Fast Company reported 3:36 pm.[37] Bedford said officials watched erratic systems for about 45 minutes before halting departures;[38] the log's gap is 44 minutes. The FAA has not said whether O'Hare was the first sign of the same fault.
  • Whether this was modernisation. The FAA answers “No” to whether its upgrades caused the outage and separates “routine equipment upgrades” from its “broader modernization effort”.[39] GAO counts the copper-to-fibre work as part of that effort's first phase.[40] The department's own release lists replacing 70% of copper lines among the results of the $12.5 billion modernisation fund.[41]
  • Equipment or installation. Bedford spoke of a defective piece of equipment.[42] The router's maker says it worked as configured.[43] Duffy says the fault is undetermined.[44]
  • One site or two. The FAA names Amarillo and El Dorado.[45] Duffy said on 15 November that the Amarillo unit, replaced about three weeks earlier, began flooding the system too.[46] His 17 November account describes one router.[47]

What is still unknown

  • The root cause: a configuration error, a defect, or something in the old network's design.
  • What the disabled “broadcasting” function is, and why it caused trouble on the 69th installation and not the first 68.[48]
  • What part the Amarillo site played.
  • Whether FAA staff assessed the risk of the El Dorado work beforehand. The technicians' union says that was the historic practice.[49]
  • Whether an inspector general review, like the one in 2010, will follow. None was found announced.

What would settle it

  • A written root cause report from the FAA and L3Harris. No date has been given.[50]
  • The FAA's reply to the senators' letter.[51]

How this page was made

Every source was opened and read on 20 November 2026. The times come from the FAA Command Center's advisory database, saved that day and converted by script. The senators' letter was not read at source, only as quoted in news reports.[52] The Fast Company article was read in part.[53] Statements by Duffy and Bedford at press conferences were read in news reports, except the 17 November remarks, which the department published.[54] A 2008 inspector general report on the same network was not read and is left out.

Continua explorant
Mostra-ho tot