November 2026 Outages: FAA, Banking and Web Services
LodiEye — November 2026
Summary
No published evidence ties the November 12–17 outages to one another or to an attack. Each operator has given its own explanation, and nearly all trace to a change or maintenance on the operator's own systems.
- Largest two. Microsoft 365 was impaired for nearly 24 hours (Microsoft report). The FAA halted departures nationwide from about 3:30 to 8:15 p.m. Eastern on November 13 (FAA).
- Banking was one outage. A software problem at JPMorgan Chase delayed transactions, and customers of other banks were affected where a payment involved Chase.
- Official position. The Director of National Intelligence, the Homeland Security Secretary and the FBI's deputy director told a House committee on November 18 that the incidents are separate and show no sign of malicious activity (CBS News).
- Still open. The FAA has not determined why its router flooded the network (FOX 4 / Reuters). Chase and Apple have not published technical accounts.
- The shared pattern is how they failed, not why. Most were triggered by a planned change, and several spread further than the design should have allowed.
- The patching explanation is only partly supported. Officials and Senator Mark Warner point to a wave of security patching. That fits the NSA disruption but not the Microsoft or FAA failures.
Information is current to about 10 a.m. Pacific on November 18, 2026. Times are Eastern unless noted.
Timeline
Eleven incidents were reported between November 12 and 17. Ten were outages and one was a disclosed security incident.
Incidents reported November 12–17, 2026
| Date | Incident | Cause as stated by the operator or officials | Duration or status | Source |
|---|---|---|---|---|
| Nov 17 | Apple: iCloud sign-in, App Store, Apple Music, AppleCare and others | An internal update, not a cyberattack | Morning until about 6 p.m. | TechCrunch |
| Nov 17 | NSA classified internal network | Scheduled maintenance; the DNI cites patching | Temporary | EconoTimes |
| Nov 17 | National Weather Service | Routine network maintenance | 40 minutes | Fox News |
| Nov 17 | Marriott sign-in on website and Bonvoy app | Not stated; no cybersecurity cause found | Resolved | Fox News |
| Nov 17 | Optimizely (security incident, disclosed) | An attacker exploited a flaw in its web experimentation software | Fixed, per the company | Click2Houston |
| Nov 16 | JPMorgan Chase, with delays at other banks | A software bug in on-premise hardware systems | About 10 a.m. until evening | Banking Dive |
| Nov 16 | United Airlines departures | None given; pilot applications and final aircraft weights unavailable | Resolved the same day | Airways / Reuters |
| Nov 15 | AWS US-EAST-1 sign-ins | None given | About 3 hours (not verified against an AWS post) | IsDown tracker |
| Nov 13 | FAA nationwide ground stop | A data flood on the legacy telecom network tied to new router work; root cause open | 3:54 to 8:15 p.m. | FAA |
| Nov 13 | Workday | A problem found while testing backup systems, per the San Francisco Chronicle | Began shortly after 5 p.m. | LiveNOW from FOX |
| Nov 12–13 | Microsoft 365: Exchange Online, Outlook, Teams, Copilot | A latent gateway bug, a Teams failover and reduced capacity | 16:20 UTC Nov 12 to 16:10 UTC Nov 13 | Microsoft report |
FAA and aviation
A flood of traffic on the FAA's legacy telecommunications network forced the November 13 ground stop. The trigger has been traced to new router work, but the reason for it is not yet known.
What happened
- Instability began about 3:30 p.m. The nationwide stop was ordered at 3:54 p.m. and lifted at 8:15 p.m. (GovInfoSecurity, FAA).
- It began at 17 of the 22 en route centers and spread to several approach control facilities (FAA).
- About 3,500 aircraft were airborne when the stop was issued. All landed safely.
- Operators canceled about 12,462 flights on November 13, including 9,082 by major airlines. The FAA says not all were caused by the stop.
- Bernstein estimates the stop and recovery cost the four largest U.S. airlines $345 million combined (AP).
What is known about the cause
- The affected network is FTI, which L3Harris owns and operates for the FAA.
- Data flooded it at about 1,000 connection attempts every 4 to 5 seconds (FAA).
- Two sites are named as potential sources. One is a radio site in El Dorado, Arkansas, where L3Harris deployed a new router that day. The other is a radar site in Amarillo, Texas, where a router had been installed about three weeks earlier (FOX 4 / Reuters).
- L3Harris blocked both sites and the network stabilized.
- Finding which contractor was working where took about 45 minutes. FAA Administrator Bryan Bedford attributed part of that to maintenance being logged on paper (GovInfoSecurity).
- The new routers are Juniper units replacing older Cisco equipment. Transportation Secretary Sean Duffy says they were tested for a year and that 68 earlier installations went without problems (ABC News).
- HPE, which owns Juniper, says its equipment was working as configured (Terminair).
- Duffy said on November 17 that investigators have recreated what happened in El Dorado but do not yet have the root cause.
- Bedford lists defective hardware, misconfiguration and improper installation as the possibilities.
Assessment
These points are inferences from the published facts, not findings.
- The Amarillo router ran for about three weeks before it misbehaved. Together with HPE's statement, that points toward configuration or an interaction between sites more than a single defective unit.
- The FAA turned off a "broadcasting function" on the new routers and has not defined the term. The flood rate and that step are consistent with a broadcast or control-plane storm on a network with weak containment.
- Which component failed matters less than why one remote site could degrade 17 of 22 centers. Senate Commerce Democrats asked the same question: why the system lacked the redundancy to recover immediately (Senate Commerce Committee).
Precedent and context
- On November 19, 2009, a Harris engineer misconfigured a temporary router during an earlier move to fiber. Service was lost at 129 facilities (GovInfoSecurity). Harris later became L3Harris.
- A November 2023 safety review described FTI as obsolete and said carriers no longer support its underlying technology.
- The replacement network, FENS, was restructured around AT&T in August 2026 and targets December 2028.
- L3Harris figures cited by senators show 3,648 routers to be converted, about 400 installed and 69 operational.
- While the investigation continues, the FAA is relying on Cisco routers that have been in service for 25 years (AP).
United Airlines, November 16
United's outage was a separate failure in the airline's own systems. Several pilot applications went down, including crew communications, and United could not supply final aircraft weights for departures (Airways / Reuters). FlightAware showed 418 United delays and 21 cancellations at 5:30 p.m., though not all are attributable to the outage. United said there was no indication of a cybersecurity issue and gave no technical cause. Reuters reported no immediate link to the FAA outage.
Banking
The November 16 banking disruption was a single outage at JPMorgan Chase. It reached customers of other banks through transactions that involved Chase.
What happened
- Downdetector reports for Chase spiked around 10 a.m. and grew around noon (Banking Dive).
- Chase posted its explanation at roughly 3 p.m. A spokesperson said transactions were processing normally by Monday evening.
- The problem spread from the website and app to some branches and to outside apps. Plaid reported connection problems with Chase accounts (Fortune).
- About 53% of Chase complaints involved balances and transactions, 20% web logins, and 17% transfers and wires.
Report counts differ by source. Banking Dive gives a peak of nearly 4,700 at 2 p.m. Fortune says the count passed 20,000 by 2:22 p.m. Either way, Downdetector counts user complaints, not customers affected.
Cause
- Chase described "a temporary issue related to a software bug in on-premise hardware systems" and said there was no evidence it was cyber-related.
- An FBI official traced it to a software release the bank rolled out earlier Monday, which affected a subset of its systems (Fox Business).
- Chase has not published a technical account.
Spillover to other banks
- Chase said other banks were affected only where they were party to a transaction with it, and that the issue was not systemic.
- Bank of America said its systems operated normally. Capital One and PNC said they were not affected. U.S. Bank said it was not directly affected but that Zelle payments with other institutions could be delayed. Wells Fargo declined to comment.
- Zelle said its network was operating as it should and that participating banks hold the funds.
- The seven largest banks with complaint spikes are all co-owners of Zelle (Banking Dive).
- The Federal Reserve's Fedwire, FedACH and FedNow systems reported normal operations (Ynet).
Assessment
- The complaint spikes at other banks are best read as delayed payments involving Chase, plus customer confusion. No other bank reported a failure of its own systems.
- Posts on X attribute to CEO Jamie Dimon a statement confirming a state-sponsored attack (example). The bank's own statements say the opposite. This review found no news outlet reporting such a statement.
- This review found no announced regulatory review of the outage.
Web and cloud services
Microsoft's outage was the longest of the week and the only one with a detailed public account. The others were shorter and are thinly explained.
Microsoft 365, November 12–13
All details below come from Microsoft's preliminary post-incident report dated November 15 (full text at Office Watch; summary at BleepingComputer).
- Impact. Most users could not reach Exchange Online mailboxes, and mail delivery was delayed. The incident ran 23 hours 50 minutes.
- Defect. Lock contention inside an internal Windows process exhausted CPU and memory on gateway servers. Microsoft has not named the process.
- Origin. The defect came from a fix that began deploying in July 2026. It ran for more than three months with no observed impact.
- Why testing missed it. The contention is worse on machines with more cores. Lab load tests used smaller hardware and a different traffic mix.
- Trigger. At 1:07 a.m. UTC on November 12, an automated failover moved Teams traffic onto the gateway that also serves Exchange Online. Nearly 14% of gateway capacity was already out of service from an unrelated fault.
- Cascade. Traffic management rerouted users to healthy servers, which concentrated load on them. The defect spread across the fleet in under two hours, and more than 80% of servers were unresponsive at the low point.
- Containment. Partitions between regions should have limited the failure. A separate issue, not yet identified, prevented that.
- Why mail was hit hardest. Teams, OneDrive and SharePoint could move off the faulty gateway. Exchange Online could not, because of scale constraints.
- Why recovery took a day. Servers coming back were flooded by queued traffic and failed again. Normal rollback was not possible for a change that had been live for months. Four early mitigations failed before the working fix began rolling out at 7:20 a.m. UTC on November 13.
- Repairs so far. Microsoft restored the earlier code path, added capacity, added monitoring and added limits on traffic redistribution. The final report is due by November 20.
Workday, November 13
Reports began shortly after 5 p.m., with nearly 3,000 on Downdetector. More than 11,500 organizations use Workday. The company's statement gave no cause. The San Francisco Chronicle reported that Workday found a problem while testing its backup systems (LiveNOW from FOX).
Amazon Web Services, November 15
Outage trackers relayed AWS notices of connectivity problems in the US-EAST-1 region. Sign-ins through IAM Identity Center and new account sign-ups were affected for about three hours (IsDown tracker). AWS gave no cause. This review could not open a primary AWS post, so these details are approximate.
Apple, November 17
Apple's status page listed iCloud Account and Sign In as affected for all users, along with the App Store, AppleCare on Device and Apple Music. Apple School Manager, iCloud Mail and others were added later. Apple said an internal update caused the outages and that it was not a cyberattack. The status page showed everything resolved by 6 p.m. (TechCrunch). Apple has given no further detail.
Marriott and Optimizely, November 17
- Marriott resolved a sign-in problem on its website and Bonvoy app. Its review found no sign of a cybersecurity cause (Fox News).
- Optimizely is the one confirmed malicious event of the week, and it is different in kind. An attacker exploited a flaw in its web experimentation software and inserted JavaScript into some customers' projects. The company says the flaw is fixed (Click2Houston).
Government systems
Two federal disruptions on November 17 were attributed to maintenance, with no indication of outside activity.
- National Security Agency. The agency's classified internal network had a temporary disruption during scheduled maintenance. The NSA said mission-critical capabilities stayed operational (Click2Houston).
- Cause as described by the DNI. Director of National Intelligence Jay Clayton linked the disruption to patches applied after a PHP vulnerability was identified. The NSA's own statement did not name PHP (EconoTimes).
- At the November 18 hearing. Asked whether the cause was internal, Clayton said there was no indication of external threat actors. He did not give a more specific cause (CBS News).
- National Weather Service. Routine network maintenance caused problems from 11:05 to 11:45 a.m. NOAA Weather Radio and NOAAPort were not affected (Fox News).
The PHP flaws are CVE-2026-109618 and CVE-2026-109619, both rated 9.2 of 10. PHP released fixes on October 21, 2026, and federal civilian agencies had an October 29 deadline to apply them (Rapid7).
Are the outages connected?
No shared cause has been shown across the three categories. What the incidents share is a pattern in how they failed.
Across categories
- Secretary of State Marco Rubio said officials have "zero evidence" linking the incidents to each other or to malicious actors (Fox News live coverage).
- At the November 18 House hearing, the DNI described separate outages associated with maintenance and technical changes. The FBI's deputy director reported no indication of malicious cyber activity (CBS News).
- These findings are self-reported and mostly preliminary. "No evidence of a link" is the accurate wording, not "ruled out."
How unusual was the week?
The worldwide count was not unusual, but the U.S. count rose. ThousandEyes logged 419 outage events worldwide for November 9–15, down 7% from 450 the week before. U.S. events rose 21%, from 234 to 282, and public cloud network outages rose from 137 to 169 (Network World).
Outage events logged by ThousandEyes, week over week
Source: ThousandEyes weekly data, reported by Network World.
Worldwide: 450 then 419 (down 7%). United States: 234 then 282 (up 21%). Public cloud network: 137 then 169.
What stands out is the prominence of the operators involved. Coverage also widened as the week went on: a 40-minute maintenance problem at the National Weather Service became national news.
The shared pattern
- Triggered by a planned change. A router installation, a software release, an internal update, a backup test, scheduled maintenance, and a months-old fix exposed by a failover.
- Weak containment. One FAA site degraded 17 of 22 centers. Microsoft's regional partitions did not hold. Chase's problem reached its counterparties.
- Testing that did not match production. Microsoft documents this directly. Senators question whether the FAA Technical Center's sample tests reflected real operating conditions (Senate Commerce Committee).
Within each category
- Aviation. The FAA and United failures have unrelated causes. The FAA's has a direct precedent in the 2009 router misconfiguration.
- Banking. There was one outage. The shared element is other banks' dependence on Chase as a counterparty.
- Web and cloud. No common vendor or component has been identified.
The patching explanation, tested
Senator Mark Warner argues that AI is finding vulnerabilities at a pace that pushes organizations to patch everything at once. He says this strains systems not built for that rate of change, and that there is no indication of a cyberattack (Fox News live coverage). The DNI made a similar point: institutions were installing patches and upgrading old technology at the same time (Yahoo News).
Checked against each operator's own account, the explanation fits unevenly.
Does the patching explanation fit each incident?
| Incident | Trigger, per the operator or officials | Fits a rushed security patch? |
|---|---|---|
| NSA | Patches after a PHP vulnerability, per the DNI | Yes |
| Apple | An internal update, not described | Unknown |
| JPMorgan Chase | A software release, not described | Unknown |
| United | None given | Unknown |
| AWS | None given | Unknown |
| National Weather Service | Routine network maintenance | Unknown |
| Microsoft 365 | A fix deployed since July, exposed by a failover | No |
| FAA | A router replacement in a long-running program, tested for a year | No |
| Workday | A test of backup systems | No |
"Triggered by a change" holds for nearly all of them. "Rushed security patch" is shown for one. Neither Warner nor the DNI has tied a specific patch to the Microsoft, FAA, Chase or Apple incidents, so the explanation is a hypothesis for those until the operators say more.
A rush to finish changes before year-end freezes could also explain the clustering. No source establishes that.
Investigation and remediation
The FAA's response is the most active. Microsoft has committed to a final report. The other operators have announced little.
FAA
- The agency disabled the suspect function on the originating routers and on more than 60 other Juniper routers, and paused further installations (FAA).
- It will reevaluate doing this kind of maintenance at midday.
- The routers from both sites are at an L3Harris lab in Florida. Duffy says FAA staff and FBI agents are there for the analysis (FedScoop).
- Senate Commerce Democrats wrote to Duffy and Bedford on November 16. They asked for a safety assessment of the operational Juniper devices and an explanation of why the system lacked redundancy, both by November 23. They also asked for a joint technical briefing with the FAA, L3Harris and Juniper, and a briefing before deployment resumes (Senate Commerce Committee).
- The administration has asked Congress for $50.1 billion, about $10 billion of it for air traffic control modernization. Senators Ted Cruz and John Kennedy both say the full amount will not be funded (FOX 4 / Reuters).
Microsoft
- Completed: the earlier code path restored, capacity added, monitoring added, and limits placed on traffic redistribution.
- Pending: a final report by November 20 with the full repair list and the longer-term root cause, including what defeated the regional partitions (Microsoft report).
JPMorgan Chase
Service is restored. The bank has not published a technical account or said that one is coming.
Apple, Workday, AWS and United
Apple and United report service restored. None of the four has published a detailed cause or a remediation plan.
Federal government
- Homeland Security contacted private industry while reviewing the incidents.
- The FBI's deputy director said its cybersecurity specialists have examined the incidents from the start and will keep working with private-sector partners (CBS News).
Open questions and dates to watch
Seven questions remain unanswered, and two dated events should narrow them.
Open questions
- Why did the El Dorado router flood the network: defective hardware, misconfiguration or improper installation?
- Why could traffic from one remote site reach 17 of 22 en route centers?
- What defeated Microsoft's regional partitions?
- What did JPMorgan Chase's Monday software release change?
- What was Apple's internal update?
- Was the NSA disruption caused by the PHP patch, as the DNI indicated?
- What caused the AWS and United outages?
Dates to watch
- Nov 20, 2026: Microsoft's final post-incident report is due.
- Nov 23, 2026: the deadline Senate Commerce Democrats set for the FAA's initial answers.
- No date set: the FAA's root-cause findings from the router analysis.
LodiEye is the original civic-reporting and analysis arm of Lodi411.com, a citizen-run civic data and transparency platform serving Lodi, California and San Joaquin County. LodiEye gathers information of public interest, applies editorial judgment to public records, meetings, and data, and publishes original explanatory reporting for its readers — the work of a newsroom, and a representative of the news media as that term is defined under federal law. Our reporting emphasizes primary sources, public data, and full source transparency so readers can check every claim. LodiEye complements, and does not replace, the other outlets covering this region; for additional reporting on Lodi, San Joaquin County, and the broader region, we also encourage readers to consult the Lodi News-Sentinel, Stocktonia, The Sacramento Bee, CalMatters, and other established news organizations. Our full editorial standards and news-media-status statement is published at lodi411.com/editorial-standards.
This LodiEye analysis was produced using artificial intelligence tools under the direction and review of the founder. Lodi411 uses multiple AI platforms in its research and publication workflow, including Anthropic's Claude (primarily Opus and Sonnet models) and Perplexity AI across a variety of large language models offered by each. These tools were used in the following capacities:
Source Discovery: AI-assisted search and retrieval identified 27 sources covering eleven incidents between November 12 and 17, 2026: FAA and Senate Commerce Committee releases, Microsoft's preliminary post-incident report, operator statements from JPMorgan Chase, Apple, United, Workday, Marriott, Optimizely, the NSA and the National Weather Service, testimony from the November 18 House hearing, ThousandEyes outage data and national news coverage. Perplexity AI was used for initial source discovery and real-time data retrieval; Claude was used for deeper analysis of identified sources.
Credibility Validation: AI cross-referenced claims across multiple independent sources, prioritizing operator post-incident reports and agency statements, then congressional records and testimony, then security-research advisories, then news reporting. Multiple AI models were used to independently verify key data points and flag inconsistencies, including the differing Downdetector peaks for Chase, the unverified social-media claim attributed to Jamie Dimon, and an AWS tracker page that could not be opened.
Analysis and Synthesis: Claude Opus and Sonnet assisted in comparing triggers, containment failures and test conditions across the incidents, and in testing the security-patching explanation offered by Senator Mark Warner and the Director of National Intelligence against each operator's own account.
Presentation: Claude assisted in drafting, structuring, and formatting the report for clarity and readability, including the incident timeline table, the patching-fit table, the ThousandEyes week-over-week chart and the open-questions and dates-to-watch lists.
Final Review: Multiple AI models reviewed the completed draft for factual consistency, source attribution accuracy, logical coherence, and balanced presentation. Throughout the process, the editor sets the report's goals, scope, and tone; creates and shapes draft content; reviews and edits the report; integrates independent fact checks; and reviews the AI cross-checks and validations. Multi-tool cross-checking across independent models and sources is the primary error-reduction mechanism.
Lodi411/LodiEye believes that transparency about how our research is produced — including our use of AI under human direction — strengthens trust with readers and the broader information ecosystem. Readers who spot an error are encouraged to write editor@lodi411.com so we can correct it.
References
All sources were accessed on November 18, 2026. One could not be opened and is marked.
FAA and aviation
- Federal Aviation Administration, Nationwide Ground Stop: Questions and Answers, November 15, 2026, since updated.
- U.S. Senate Committee on Commerce, Science, and Transportation (Democrats), Commerce Committee Democrats Demand Clear and Fact-Based Answers From DOT, FAA on Root Cause of Nov. 13th Nationwide Ground Stop, November 17, 2026.
- FOX 4 News with Reuters, 'We don't have clarity' on what caused FAA groundstop, Duffy says, November 18, 2026.
- Associated Press via KTVN 2 News, FAA temporarily stops router installations connected to air travel stoppage, November 2026.
- ABC News, Mullin faces congressional hearing in the wake of historic nationwide ground stop, November 18, 2026.
- GovInfoSecurity, Anatomy of the FAA Ground Stop: One Backbone, One Bad Change, November 2026.
- FedScoop, FAA, FBI investigating router responsible for nationwide ground stop, November 16, 2026.
- Terminair, The FAA ground stop and the paused router rollout: what we know, November 2026.
- Airways Magazine, citing Reuters and the Associated Press, United resolves technology outage affecting departures, November 2026.
Banking
- Banking Dive, Banking disruption resolved after thousands report access issues, November 17, 2026.
- Fox Business, JPMorgan Chase systems recovering following hourslong nationwide technical outage, November 16, 2026.
- Fortune, Days after Microsoft and the FAA went dark, Chase and Wells Fargo customers can't get to their money, November 16, 2026.
- Ynet News, Major US banks hit by online outages, leaving thousands unable to access accounts, November 2026.
- Post on X attributing an unverified statement to Jamie Dimon, cited only as an example of the claim.
Web and cloud services
- Office Watch, Microsoft 365 outage: what went wrong, in plain English, with the full text of Microsoft's preliminary post-incident report dated November 15, 2026.
- BleepingComputer, Microsoft says months-old bug caused global Microsoft 365 outage, November 2026.
- LiveNOW from FOX, Workday outage means employees across thousands of businesses can't access accounts, November 13, 2026.
- IsDown, AWS: increased connectivity issues in US-EAST-1, November 15, 2026. Not opened; details taken from a search excerpt.
- TechCrunch, Apple services, including iCloud, App Store, Apple Music, and more, hit by outage, November 17, 2026.
- Network World, 2026 network outage report and internet health check, ThousandEyes weekly data, November 2026.
Government statements and cross-cutting coverage
- CBS News, "No indication" recent outages are related to coordinated cyberattacks, officials say, November 18, 2026.
- Fox News, TIMELINE: America hits a tech wall as outages cripple email, ground flights and disrupt major banks, November 2026.
- Fox News live coverage, Top Dem senator says AI is contributing to outages as Rubio says incidents not connected, November 18, 2026.
- Click2Houston (KPRC), Recent tech outages: Are Microsoft, FAA, Chase and United disruptions connected?, November 18, 2026.
- EconoTimes, Clayton Says U.S. Outages Show No Signs of a Coordinated Cyberattack, November 2026.
- Yahoo News, DNI Dismisses Cyberattack Fears Amid Widespread Outages, November 2026.
- Rapid7, CVE-2026-109618 and CVE-2026-109619: Critical Unauthenticated Remote Code Execution in PHP, October 2026.