EN / 中文

FAA Nationwide Ground Stop: Lessons From the L3Harris Telecom Outage and Aging U.S. Air Traffic Infrastructure

by xianzaoketangxiaozaojun·November 19, 2026

A few days ago, the nationwide flight grounding incident in the United States caused quite a stir.

A routine communication line upgrade directly led to a suspension of US flights for over four hours. More than ten thousand flights were canceled or delayed, and a large number of passengers were stranded at airports. The impact was indeed severe.

As a veteran in the communications industry, Xiaozao Jun believes that although this incident has concluded, there are many aspects worth reviewing behind the event.

Beyond just watching the spectacle, we need to seriously consider what lessons and insights this event brings to the entire communications industry.

█ Event Review

Let us briefly review the course of events.

At 15:36 local time on November 13, unstable radar and communication data began appearing across all regional control centers and terminal radar approach control facilities in the United States.

At 15:53, which is 17 minutes later, the FAA (Federal Aviation Administration) Air Traffic Control System Command Center issued a notice ordering a ground stop at all airports nationwide due to "special circumstances and safety reasons." All domestic flights that had not yet taken off remained stationary, with departure times to be announced separately.

At 20:15, the FAA announced the lifting of the ground stop, and airspace was reopened.

This nationwide grounding lasted approximately 4 hours and 23 minutes, making it the longest national flight suspension in the US since the September 11 attacks.

█ Impact of the Incident

The impact of this nationwide grounding can be understood through several sets of data:

According to data from the flight tracking platform FlightAware on the 13th, approximately 9,200 flights in the US were canceled, and 7,500 were delayed.

By 18:45 on the 14th, airlines had canceled more than 6,150 additional flights and delayed over 7,300. Furthermore, more than 530 flights originally scheduled for the 15th were canceled in advance.

Combined over two days, more than 15,000 flights were canceled, and over 14,000 were delayed.

The impact extended beyond the US. On the evening of the 13th, Eurocontrol (the European Organization for the Safety of Air Navigation), at the request of the FAA, issued a notice temporarily suspending all flights departing from member countries bound for US airports. Air Canada also temporarily suspended round-trip flights between the US and Canada.

Lifting the ground stop does not mean everything returns to normal immediately. The schedules for aircraft, crews, and passengers were all disrupted, leaving the US aviation transport industry in chaos. Normalcy was only largely restored by the16th.

This grounding not only caused significant inconvenience to passengers but also resulted in massive economic losses.

According to estimates by Airlines for America, direct economic losses for major US airlines exceeded USD 250 million on the13th alone. This figure does not include indirect losses such as subsequent flight cancellations, passenger rebooking, refunds, and accommodation compensation. If the cumulative impact over two days is included, the total loss scale may exceed USD 500 million.

█ Cause of the Failure

So, how exactly did this failure occur?

The system involved was the FAA's Telecommunications Infrastructure (FTI).

This is a dedicated network carrying regional interconnectivity, radar data, and air traffic control dedicated voice lines. Radio communications between control centers, between controllers and aircraft, radar data, and aircraft position information all run on this network. It serves as the underlying foundation of the US air traffic control system, highlighting its critical importance.

The FTI is not operated by the FAA itself but is outsourced to a private contractor—L3Harris Technologies.

This company was formed in 2019 through the merger of L3 Technologies and Harris Corporation, two established defense giants. Its business covers private network communications, radar, avionics, electronic warfare, satellite and missile systems, among others.

On the day of the incident, L3Harris conducted routine maintenance work, upgrading old copper cables to fiber optics at a remote site's air traffic control system and replacing the corresponding routers.

During the upgrade, one router (some reports suggest more than one) newly installed experienced an anomaly, beginning to send a massive number of connection requests to other routers, exceeding 1,000 requests every 4 to 5 seconds.

Readers familiar with networking will recognize this as essentially an unintentional DDoS attack. A large volume of abnormal requests overloaded the communication network, affecting 17 out of the 22 Regional Air Traffic Control Centers under the FAA.

Whether the fault was due to installation or configuration errors by construction personnel, or defects in the router itself, remains under investigation.

Network congestion meant that different control centers received inconsistent aircraft position data, preventing controllers from determining if airspace was safe. Under these circumstances, new flight departures could only be suspended.

Additionally, systems such as NOTAMs (Notices to Airmen) and ADS-B (Automatic Dependent Surveillance–Broadcast) also rely on the FTI network. NOTAMs inform pilots about runway closures and navigation equipment failures; ADS-B allows aircraft to actively broadcast their positions. Once the FTI network fails to operate normally, the entire safety chain is affected.

When the failure first occurred, officials initially suspected a cyberattack or terrorist act. However, investigators quickly ruled out external intrusion, confirming it was an issue within the FTI itself. Later, technicians identified the source of the fault. After isolation measures were taken, the network gradually recovered.

█ Frequent Failures in Recent Years

This grounding incident is not the first time the US aviation system has experienced problems. In fact, in recent years, they have frequently encountered failures related to information and communication technologies:

In January 2023, the FAA ordered all domestic flights to suspend takeoffs for about two hours due to a NOTAM system failure, resulting in over 1,300 cancellations and nearly 10,000 delays. Post-incident investigations revealed that contractor staff accidentally deleted some files while correcting synchronization issues between primary and backup databases.

In April and May 2025, Newark Airport experienced two interruptions due to failures in aging communication lines, leaving controllers without radar information for approximately 90 seconds.

In September 2026, a communication line failure occurred at an air traffic control facility in Philadelphia. When the system switched to backup fiber optics, it was discovered that the backup line had been accidentally severed by construction workers, causing delays or cancellations for approximately 7,000 flights.

Going further back, on November 19, 2009, the FTI network also went down once: a long-dormant configuration error was triggered by the restart of a router in Los Angeles, blocking about 75% of network traffic. Recovery took nearly five hours. Although no nationwide ground stop order was issued then, the impact was equally severe.

█ Issues Exposed by This Incident

The above outlines the full course of the nationwide grounding incident. Next, let us examine the issues exposed by the event.

Firstly, the most profound feeling after the incident is that US public infrastructure is indeed very fragile.

As a national-level critical infrastructure covering US airspace, the overall technical level of the FTI still lags behind standards from the 1990s.

A 2024 report by the GAO (Government Accountability Office) stated that out of 138 air traffic control systems managed by the FAA, 51 were deemed unsustainable for operation, and another 54 posed similar risks. Together, these 105 systems account for approximately 76%.

Among these 105 systems, 73 have been in use for over 20 years, 40 for over 30 years, and 6 for even over 60 years. Spare parts are difficult to obtain for 61 systems, and 62 lack specialized maintenance personnel. Some systems' communication lines still use copper cables laid in the 1960s.

FAA Administrator Bedford stated at a press conference that maintaining this legacy network costs approximately USD 1 billion annually. Reuters noted that obsolete equipment often requires sourcing parts from online auction platforms. Over 400 air traffic control facilities' outdated floppy disks are not expected to be fully phased out until the end of 2028.

Floppy disks are something young people in China have rarely seen. Previous news indicated that the US nuclear weapons system and cockpits of some Boeing passenger jets still use floppy disks.

At the press conference, US Secretary of Transportation Duffy himself admitted that the air traffic control system still relies on analog communication technology (TDM, Time Division Multiplexing). Switches and routers overload due to excessive data loads, outputting inconsistent data. A communication foundation based on analog technology simply cannot withstand high loads, which is the primary reason for shutting down the system.

Americans are aware that their systems are outdated and wish to upgrade. The US does not lack the funds for such initiatives.

However, upgrades to US public infrastructure often stall due to political maneuvering, budget approvals, departmental coordination, and vested interests, making progress slow.

In 2025, the Biden administration proposed a five-year plan totaling approximately USD 40 billion for modernizing the air traffic control system. However, the plan faced resistance during congressional deliberation, with some legislators questioning the match between funding scale and actual effectiveness, leaving the plan stuck in the discussion stage to this day.

As early as 2018, the FAA launched the "NextGen" upgrade project, originally planning to complete core architecture replacement by 2025. However, constrained by budget allocation rhythms, contractor delivery delays, and risks associated with switching between old and new systems, the project timeline was repeatedly postponed.

The primary cause of this grounding—the replacement of copper wires with fiber optics—has been proposed by the FAA for decades. In 2025, the FAA suddenly compressed the original 20-year project planned until 2038 into 2.5 to 3 years, investing USD 12.5 billion to accelerate progress. As a result, when the progress was just past halfway, this accident occurred.

Interestingly, the day before the grounding, Duffy requested USD 30.1 billion from Congress. After the grounding, this figure jumped to USD 50 billion. Thinking primarily of asking for huge sums of money after an incident is truly remarkable.

Secondly, apart from the equipment itself, this accident definitely involved human factors.

Readers working in the communications industry know that upgrading existing communication systems, whether public or private networks, follows strict operating procedures, including cutover during nighttime low-traffic periods, service switchover verification, and rollback drill preparations.

The failure occurred at 15:36, during peak flight hours. Why at this time? Were routers and fiber optics replaced during the day? Why wasn't it scheduled for the night?

Did the upgrade replacement plan have flaws? Was business verification and testing conducted after replacement? Was there a checklist? Why wasn't the problem detected and rolled back immediately?

The accident lasted over four hours. Why did the troubleshooting process take so long? Troubleshooting anomalies in individual routers should not be particularly difficult.

Why wasn't the problem contained when localized network issues arose? Instead, the fault spread throughout the entire system. Are there design flaws in routing strategies and isolation mechanisms?

These questions point not just to technical oversights or mere bad luck, but to laxity in process execution and emergency management.

The professional skill levels of relevant personnel also warrant scrutiny. Did the engineers performing the operations lack technical competence? Or, under pressure from significantly compressed timelines, were processes that should have been strictly executed simplified or skipped?

We may need to wait for the final investigation report to fully understand the answers to these questions.

This accident also exposed issues with the operational model of US public infrastructure.

The operational rights of the FTI were wholly outsourced by the FAA to a single private contractor. While this model reduces direct government operating costs, it puts all eggs in one basket. Once the contractor makes a mistake at any link, the entire system faces cascading paralysis.

This outsourcing model is widely adopted in key areas such as the US power grid, water supply, telecommunications, and traffic signals, all harboring similar hidden dangers.

When the core lifelines of public safety are held by a few commercial companies, their operational transparency, timeliness of maintenance, and response capabilities to emergencies inevitably raise questions.

In summary, if the US continues to maintain its current attitude towards public infrastructure construction and maintenance, similar accidents will likely recur.

As a superpower, US infrastructure was once a global benchmark.

However, over time, this former benchmark is fading. From bridge collapses to grid blackouts, from railway accidents to aviation communication disruptions, a series of incidents repeatedly sting the nerves of the American public. If they do not fundamentally reflect upon and adjust their infrastructure operation and regulatory systems, they will pay an even heavier price.

Public trust in government departments will gradually collapse. The credibility of the US government will also suffer irreversible damage.

█ What Can We Learn from This Incident

After the grounding incident, domestic media provided detailed coverage. Domestic sentiment generally leaned towards spectating, even with some schadenfreude.

In reality, mockery holds little practical significance. Instead, calmly drawing lessons to avoid similar issues occurring here is what we should truly be doing.

Although our country's public infrastructure now ranks among the world's best overall, this does not mean we can rest easy.

In recent years, our country has achieved remarkable success in new infrastructure development. However, rapid construction and upgrades often introduce new technical risks and operational challenges.

To be honest, while we mock Americans for using technology from decades ago, copper cables, and analog tech, some of our own private networks also contain aging equipment. For example, SDH optical transmission devices and program-controlled switches are still in service at some critical nodes.

The requirement for private networks has never been technological advancement, but stability and reliability. Although these devices belong to older generations, they function normally with mature operations, leading to weak willingness among relevant units to replace them.

However, as equipment reaches the end of its lifecycle, issues such as discontinued spare parts, withdrawal of manufacturer support, and gaps in maintenance personnel are gradually emerging, potentially bringing security risks. Replacing and upgrading these devices also involves risks and challenges.

Everyone knows a characteristic of information and communication equipment—once running, try not to touch it; touching it may cause problems.

The upgrade and transformation process is where issues are most likely to arise. The phase where old and new systems coexist carries the highest risk.

No system is absolutely safe and reliable. Problems always appear in inexplicable ways and may also be resolved in inexplicable ways.

In recent years, many equipment vendors have touted how advanced their O&M technologies are and how secure their systems are. But when critical moments arrive, various issues still emerge. Either the equipment fails, or humans make mistakes.

Xiaozao Jun still believes we must have reverence for technology, reverence for systems, and reverence for risks. Do not become blindly confident because of technological advancement, nor complacent because of system stability.

We launch too many new systems, and the importance of these systems is increasing. Once a failure occurs, the scope of impact often exceeds expectations. In recent years, we have actually experienced many similar incidents, though none as severe as those in the US.

Finally, I want to mention AI.

AI is currently developing rapidly, with industries across the board pursuing intelligentization. Our ICT industry is no exception. Using AI to enhance network management, monitoring, and automated processing, and using AI for fault prevention and handling, is an important trend.

AI is actually a double-edged sword.

Its benefits lie in higher efficiency—alerts, logs, and performance data that humans cannot handle comprehensively, AI can; repetitive O&M actions, AI can perform on behalf of humans; fault localization, AI is faster than humans. With AI, system security and reliability can indeed be significantly improved, effectively reducing risks from human error, providing early warnings before faults occur, and eliminating problems at the bud.

But AI is not absolutely safe. The more AI is used, the easier it is to create an illusion that the system seems safer. In reality, it simultaneously raises complexity by a notch. Models hallucinate, automation misoperates, and black-box decisions are hard to trace when problems arise.

In this US grounding accident, the AI scheduling system recently launched by the FAA was initially suspected. Although eventually cleared, this indicates that whenever new technology enters the scene, everyone's first reaction is, "Could it be this?"

AI can serve as a co-pilot, but we must not let it drive blindfolded. As AI usage deepens, it itself should be listed on the risk register: Is every action monitorable, auditable, and rollbackable? If AI makes a mistake, can humans catch the fall?

All these require thorough consideration, along with establishing corresponding mechanisms and contingency plans. We cannot completely hand over the lifeline to machines.

█ Conclusion

Well, that concludes Xiaozao Jun's views on this nationwide grounding incident.

Turning others' lessons into our own experience is the correct approach. Maintaining reverence for equipment, caution regarding changes, and clarity regarding new technologies are what we should constantly remind ourselves of.

Looking toward the future, humanity's fate increasingly depends on the stability and reliability of ICT systems. Xiaozao Jun personally believes that faults will never cease or disappear; the impact and harm of faults will only grow. What we can pray for is that apocalyptic-scale system disasters never descend upon humanity.