Best practice

FAA ground stop: OT change management lessons from a router upgrade

US officials say the nationwide FAA ground stop followed upgrade work on air traffic control communications, with no sign of a cyberattack. OTVista researchers explain the lessons for OT change management, segmentation and resilience.

OTVista article cover: FAA ground stop: OT change management lessons from a router upgrade. Photo: Fibre optic cables in a network switch

The nationwide US flight ground stop on 13 November 2026 followed upgrade work on the communications network behind air traffic control, not a cyberattack, according to US officials. For OT operators, the lesson is that a routine network change on legacy infrastructure can take down a critical operation on its own, so change management, dependency mapping and network resilience deserve the same attention as threat defence.

What happened

On Friday 13 November 2026, the Federal Aviation Administration (FAA) halted departing flights across the United States after a communications problem hit its air route traffic control centres. According to SecurityWeek, the ground stop was lifted at 8:15 pm ET, more than four hours later. CNBC described it as the longest pause in US departures in 25 years.

At a press conference on 14 November, Transportation Secretary Sean Duffy said the problem came after updates to a telecommunications system managed by contractor L3Harris, as reported by CNBC. He said work at "at least" one remote location was part of an ongoing switch from copper wiring to fibre. A router reportedly flooded other routers in the air traffic control network with more than 1,000 connection attempts roughly every five seconds, disrupting aircraft communications and radar feeds. Duffy said investigators were working out whether the cause was installer error or a faulty router.

FAA Administrator Bryan Bedford said no FAA facilities went dark, but information feeds to en route centres became unstable, and officials were unsure of the cause for about 45 minutes, according to CNBC. The FAA grounded flights because it was unwilling to risk a centre losing contact with aircraft. Before departures resumed, the FAA said it had tested its systems to confirm they were stable.

Officials said there was no evidence of a cyberattack. SecurityWeek reported that Homeland Security Secretary Markwayne Mullin said there was "no chatter" and no "footprints", while not fully ruling it out, and that the NSA, the Department of Defense and CISA were involved in the investigation. Bloomberg, cited by SecurityWeek, reported that the FBI was examining whether a software update was responsible. L3Harris said it was supporting the FAA investigation and that more than 70% of the copper-to-fibre migration was complete. At the time of writing, no official root cause had been published.

The disruption was large. CNBC, citing FlightAware, reported more than 9,000 US flight cancellations on Friday, about a third of the day's schedule (some weather-related), and nearly 6,500 more on Saturday. SecurityWeek noted that the previous nationwide ground stop, in January 2023, also followed contractor work on a critical system rather than an attack.

Why it matters for OT

Air traffic control is not a factory or a substation, but the failure pattern will look familiar to anyone who runs industrial networks. Based on what officials have said so far, three points stand out.

Change on legacy infrastructure is a hazard in itself. The incident reportedly started during a planned migration at a remote site. In OT, similar work happens all the time: replacing serial or copper links, swapping switches in a substation, adding a new firewall at the IT/OT boundary, or upgrading a remote terminal unit. Every one of these changes touches systems that were designed for stable, predictable traffic.

One misbehaving device can propagate across a flat or tightly coupled network. A single router generating excessive connection attempts was enough to destabilise feeds across multiple centres. Industrial networks with limited segmentation, shared broadcast domains or no rate limiting between sites can fail the same way, whether the trigger is a faulty device, a misconfiguration, a broadcast storm or malware.

Diagnosis time is operational time. Officials reportedly needed around 45 minutes to understand the cause. In OT, the same uncertainty forces operators into conservative safety decisions, such as tripping a process or moving to manual operation, which is the right call but expensive. Good network visibility and documented architecture shorten that window.

It is also worth stressing what this incident is not. There is currently no evidence of malicious activity, and we should not treat it as a cyber incident. But "no attacker" does not mean "no security lesson". Availability is the first concern in most OT environments, and the controls that contain an attack (segmentation, monitoring, tested recovery) also contain accidental failures.

Who should act

Any organisation running distributed operational networks should take note, especially:

  • Electricity transmission and distribution, where substation and control-centre communications are being migrated to new technologies.
  • Rail and transport operators, whose signalling and operational communications span many remote sites.
  • Water and wastewater utilities with telemetry links to pumping stations and treatment sites.
  • Oil, gas and pipeline operators relying on wide-area SCADA communications.

For EU operators in scope of NIS2 and UK operators working to the NCSC Cyber Assessment Framework, it is worth checking that accidental failures during planned changes are reflected in risk assessments and resilience planning, not only deliberate attacks.

What to do now

  1. Treat network migrations as OT changes. Route copper-to-fibre, switch, router and firewall changes through formal OT change management, with operations sign-off, a defined maintenance window and a written rollback plan. Our notes on patching when outages are limited cover how to plan changes around constrained windows.
  2. Map dependencies before the window. Confirm which control systems, feeds and sites depend on the links and devices being changed. An up-to-date OT asset inventory that records network equipment, firmware versions and owners makes this much faster.
  3. Stage the change and test first. Where practical, test new devices and configurations in a representative lab or on a single low-impact site, then roll out in stages with clear go or no-go checkpoints.
  4. Contain failure between zones and sites. Use segmentation, rate limiting, storm control and connection limits so that one misbehaving device cannot saturate the wider network. Review this as part of an OT security architecture review.
  5. Monitor the change as it happens. Watch traffic levels, connection counts and the health of critical feeds during and after the work, and agree in advance which signals trigger a rollback.
  6. Supervise contractor and supplier work. Make sure third parties follow your change procedures, use approved access paths and hand over updated documentation when they finish.
  7. Exercise manual and degraded operation. Confirm that operators can safely run, or safely stop, the process if communications become unreliable, and that this mode has been practised, not just written down.

How OTVista helps

OTVista researchers study how industrial networks fail, by accident as well as by attack. Our OT security architecture reviews look at segmentation, communication paths and resilience between sites, and our policies and procedures work covers practical OT change and supplier management. If you want a quick view of where your own environment stands, try the free OT security self-check.

Sources: SecurityWeek, No Signs of Hacking in FAA Ground Stop as FBI Reportedly Probes Software Update; CNBC, Nationwide ground stop followed contractor upgrade to L3Harris telecommunication system

Cover photo: Kirill Sh on Unsplash, used under the Unsplash License.

Topics: FAA ground stop, OT change management, OT network resilience, legacy infrastructure upgrade, OT segmentation, critical infrastructure outage

Frequently asked questions

Was the FAA ground stop on 13 November 2026 a cyberattack?

US officials said they found no evidence of a cyberattack, although investigators had not fully ruled it out. The FBI was reportedly examining whether a software update was responsible.

What caused the FAA ground stop?

Transportation Secretary Sean Duffy said it followed updates to a telecommunications system managed by L3Harris, during a copper-to-fibre migration. He said a router flooded other routers with more than 1,000 connection attempts roughly every five seconds. No official root cause had been published at the time of writing.

Why does this matter for OT operators?

Industrial networks also depend on legacy communications that are being upgraded. A single faulty device or misconfiguration during a change can disrupt critical operations without any attacker involved.

How can OT teams reduce the risk of change-related outages?

Use formal OT change management with rollback plans, map dependencies before the work, stage and test changes, and use segmentation and rate limiting so one device cannot saturate the wider network.

See OTVista in action

Join the research preview or talk to us about your OT security priorities.

Join the research