The November Tech Outages, One Failure at a Time
First it was Outlook. Then it was air traffic control. By Monday, people were waiting on bank transfers and flights all over again. By Tuesday, Apple and the NSA were in the headlines. I had the same question a lot of people did. Were these November tech outages connected?
I went through public incident reports, technical discussions, and security reporting. Here’s where the evidence stands as of November 18, 2026. We can see some of what happened from the outside. We don’t have the internal logs or investigation records, and the public evidence leaves the question of a cyberattack open wherever the facts are incomplete. I haven’t seen evidence that connects these incidents to a single attack.
The FAA’s Friday afternoon cutover
On November 13, the FAA’s telecommunications network became unstable during work to move a remote radio site in El Dorado, Arkansas, from copper to fiber. Older equipment struggled with excessive network traffic. The disruption affected 17 of 22 en route control centers. The FAA stopped departures nationwide and began resuming operations at 8:15 p.m. Eastern. Some radio and radar feeds stayed available, and the aircraft already in the air landed safely. (FAA incident Q&A · DOT’s November 14 briefing)

Think of a highway ramp feeding too many cars into one stretch of road. The backup stops being a problem for that ramp alone. Here, the traffic was network messages, and the equipment struggling to keep up supported the feeds controllers use. Investigators reproduced what happened at El Dorado and were still working out why it happened. The FAA said it had found no indication of a cyberattack. (DOT’s November 17 update · Reuters on the FAA investigation)
The part I keep coming back to is Friday afternoon. The Washington Post reported that an FAA official tested and signed off on the work. The trouble started about eight minutes later. DOT says the old copper line was being discontinued. That explains why the work needed doing. The public record still leaves me wondering why it needed doing that afternoon, a week before the forecast Thanksgiving travel period began. What did the test check, and how did the FAA expect to contain a problem at that site? (The Washington Post on the FAA sign-off · Airlines for America’s holiday forecast · Senate Commerce oversight letter)
Microsoft 365 and the two-day gap
Microsoft patched an actively exploited Outlook flaw on November 10. Two days later, people couldn’t get to their mail. I wanted to know whether the dates meant anything. The advisory covers the Outlook app, while Microsoft’s preliminary incident report traces the outage to the cloud gateway that handles connections to Exchange Online. A software defect had been sitting there since July. Teams traffic shifted onto that gateway after a problem elsewhere, while about 14% of its capacity was already offline. (Microsoft’s Outlook security advisory · Microsoft’s preliminary incident report (PDF))
Picture a grocery store sending everyone to the registers on the other side of the store. A few are already out of service, and the remaining ones have a bug that only shows up when the lines get long. Bringing one back gives you about five seconds of relief if everyone rushes it at once. Microsoft’s report describes that kind of recovery problem. Its security team also documented phishing emails pretending to be outage notices. That’s a real security incident with its own timeline. (Microsoft Threat Intelligence on the phishing campaign)
Technical discovery & auditing
The public page doesn’t tell you much about the machinery behind it. Raymond Tec audits inherited and long-running projects to uncover the plugins, integrations, data, dependencies, and old decisions that determine what the next change will really involve.
N-able took its own service offline
N-central is software IT providers use to manage customers’ computers remotely. On November 9, N-able took the hosted version offline after vulnerabilities were disclosed publicly. That software has broad access by design. Think about how you’d handle a master key that could be copied. You’d want to change the lock before trusting it again. N-able restored hosted service after applying a hotfix. (Health-ISAC’s N-central bulletin · N-able’s remediation updates)
A later disclosure led to evidence of actual attacks on November 12. Security researchers observed exploitation attempts, and Huntback reported finding cryptominers on 21 production N-central servers. The count comes from the researchers. The earlier hosted shutdown was N-able’s decision to contain the risk before it had confirmed exploitation. (BleepingComputer’s exploitation reporting)
Chase, other banks, and Zelle
On November 16, Chase said a software bug in its own systems delayed transactions. Customers at other banks reported trouble, too. U.S. Bank said its systems were unaffected, and Bank of America and Capital One said theirs were operating normally. Early Warning Services, which runs Zelle, said the Zelle network was operational. (Reuters on Chase’s software bug · Fortune’s Bank of America statement · Banking Dive on Capital One · AP on the bank and Zelle delays)

Think about a package that’s stuck at the sender’s depot. The destination depot can be open all day and still have nothing to put on your truck. That’s how a Chase problem could show up on somebody else’s bank app when a transaction involved Chase. That explains a documented part of the disruption. The public record is thinner on what caused individual login or balance complaints at other banks, including Wells Fargo. (Reuters on Chase’s software bug · AP on the bank and Zelle delays)
Workday’s backup test reached the live system
Workday said a backup test uncovered a problem on November 13. A notice relayed to customers said services in the disaster recovery environment unexpectedly connected to live databases and left some data inconsistent. Workday put affected customers’ systems into maintenance while it worked out how to recover them. For the people trying to submit time or run payroll, that meant waiting for access to the system they use to do the job. (San Francisco Chronicle on Workday’s backup test · StatusStack’s archive of customer notices · Florida Atlantic University’s Workday advisory)
The Workday incident gives us a concrete reason to ask how a recovery drill was able to reach the live data. (StatusStack’s archive of customer notices)
AWS had a regional sign-in and connectivity problem
On November 15, Amazon Web Services reported sign-in and connectivity failures in its Northern Virginia region. Some customers couldn’t authenticate through IAM Identity Center, the service that handles sign-in for the console and some AWS tools. A smaller group still had load-balancer trouble after sign-in recovered. A load balancer directs requests to available servers, a bit like the person at a busy checkout pointing you toward an open register. (AWS incident updates archived by StatusWire)
The recovery order is worth noticing. A working sign-in screen told you the front door was open again. You still had to check whether the particular service behind it was working. AWS’s archived updates give us that sequence. The underlying trigger is still unclear in the public account.
Backups & recovery
Backups are comforting right up until you need one. Raymond Tec helps small businesses build practical backup and recovery plans — including the decidedly unglamorous part where we make sure the thing can actually be restored.
United had its own departure delays
On November 16, United Airlines said an internal technology outage delayed departing flights. A pilots’ union memo obtained by Reuters said crews couldn’t get final aircraft weights and had problems with crew communications. Air traffic control audio also captured a dispatch-data problem. A crew can have a fueled plane full of passengers and still need those numbers to complete its takeoff calculations. Flights already airborne and United Express flights were unaffected, according to United. The company hasn’t publicly identified the underlying fault. (Reuters on United and the pilots’ memo · NBC Chicago on the ATC audio)
The NSA’s systems didn’t come back as expected
The NSA said it experienced disruption during scheduled maintenance on November 17 and that mission-critical capabilities kept running. Director of National Intelligence Jay Clayton said some systems didn’t come back as expected after patching software for PHP vulnerabilities. Think of shutting off the water to replace a valve. The time you planned for the water to be off and the time it actually stays off can be two very different numbers. (The NSA’s statement · The Record on the patch and network disruption · The NSA’s PHP patch advisory)

The affected network carries classified internal communications. The Record reported intermittent access and staggered staff arrivals to reduce sign-in load. CNN reported that the NSA later restricted its campus to essential personnel that night. Those details give us a better sense of how the outage affected the workday. The public record still leaves the full scope and restoration time unclear. (The Record on the patch and network disruption · CNN’s Fort Meade reporting via KRDO)
Apple’s update reached well beyond iCloud
On November 17, Apple attributed a wide outage to an internal update and said a cyberattack hadn’t caused it. Its status pages recorded trouble across iCloud sign-in, the App Store, Apple Music, and business and school services. People reported problems activating new phones. Some Apple stores couldn’t complete purchases, and Apple Pay complaints were mixed. Apple’s public dashboard was clear by about 5:46 p.m. Eastern, according to the company, while some developer-facing problems continued into the evening. (CBS News on Apple’s recovery · MacRumors on retail and device activation · AppleInsider’s status-page timeline · Apple’s statement to TechCrunch)

Apple’s public explanation hasn’t identified the component it changed or how that change spread across services.
Business IT goes well beyond the website
Your business also depends on workstations, cloud accounts, browsers, Wi-Fi, remote access, collaboration tools, and all the other technology that quietly becomes infrastructure. Raymond Tec works across that whole stack, whether the problem lives on a server, on a desk, or somewhere in between.
What I’d ask after the status page turns green
The question I’d put to the FAA is the same one I’d ask of a much smaller system. If this change goes sideways, what else goes with it? I want to know what the acceptance test covered, how far the team expected a failure to spread, and whether they could get back to the old setup in time. Those are questions we can ask now, even while investigators are working out the exact failure. (Senate Commerce oversight letter)
For a smaller business, start with a process you need on your busiest day. If both your normal checkout and your backup plan rely on the same sign-in or internet connection, try the backup with that dependency unavailable. Finding out on an ordinary Wednesday is a lot less exciting. That’s the point. (How we tested deployment and recovery on Longtail Forge)
Photo credits
Featured highway overpass photo by Drone Visionary on Pexels and Runway photo by 1 cnrdmroglu on Pexels (Pexels license).
Chase branch photo by Harrison Keely via Wikimedia Commons (CC BY 4.0, resized). Fort Meade photo by the National Security Agency via Wikimedia Commons (public domain). Apple Park photo by Daniel L. Lu via Wikimedia Commons (CC BY-SA 4.0, resized).
Technology has an annoying habit of connecting to everything else. Browse the News archive for current stories, or dig into Raymond Tec Insights for longer pieces about what building, fixing, maintaining, and living with this stuff actually looks like.
