Skip to content
By 5 min readNews

Six Outages in Six Days: What Each Organization Said Broke

Microsoft 365, the FAA, Workday, Chase, United and Apple all failed between November 12 and 17, 2026. Each cause as the organization stated it, and what to check.

Six Outages in Six Days: What Each Organization Said Broke, by Deepak Gupta on guptadeepak.com

Between November 12 and 17, 2026, six organizations had major outages: Microsoft 365, the FAA's air traffic network, Workday, JPMorgan Chase, United Airlines and Apple. No organization or investigator has said any of them share a cause. They happened on the same days, and so far that is all the public record supports.

I record incidents like these in Outage Atlas, with each cause in the organization's own words and its status. Here is each outage on those terms.

Microsoft 365: Thursday, November 12

Exchange Online, Teams, Copilot, SharePoint, OneDrive, Intune, Defender and other services failed worldwide. On November 13, Microsoft pointed to "an earlier network capacity upgrade," as quoted by Windows Latest. Its preliminary Post Incident Report of November 15 (publicly available only as a copy hosted by Next.ink) names "three main factors," starting with "a latent defect caused Windows servers to become overwhelmed and unresponsive under certain load conditions."

Per that report, the defect shipped in July in a fix for the networking gateway servers that front many Microsoft 365 services. A routine Teams failover moved load onto the gateway shared with Exchange Online while about 14% of its capacity was already out of service. Rerouting load to healthy servers spread the failure, and separation between capacity units did not hold. Microsoft reverted the code path. The cause is preliminary; the final report is due November 20.

Microsoft's Charles Lamanna said "there is no indication this is the result of malicious or cyber activity," per Al Jazeera. Unity Health Toronto declared a code grey and said "patient and resident care was not affected," per Fortune. On November 13, Microsoft Threat Intelligence saw a high-volume phishing campaign using fake Microsoft 365 service-health notifications as lures, per the Microsoft Security blog. Microsoft called it likely credential phishing, though its evidence did not confirm credential capture. Full record: Microsoft 365 incident page.

FAA nationwide ground stop: Friday, November 13

The FAA stopped departures nationwide at 3:54 p.m. Eastern and lifted the stop around 8:15 p.m., per Al Jazeera. The preliminary account: during a copper-to-fiber telecom upgrade by contractor L3Harris at a remote site, a router began flooding the air traffic control network, at "a thousand connections over every four to five seconds," per TechCrunch.

Whether the router or its installation was at fault is open. Transportation Secretary Sean Duffy asked, per ABC News, "installer error from L3Harris, or did they install a faulty router?" HPE said "the device functioned as it was configured to by the installer," per ABC News live updates. The FAA turned off that router type's ability to broadcast across the network and is running a root cause analysis. Asked about an attack, Homeland Security Secretary Markwayne Mullin said "it does not appear," adding "we're still looking, so we're not going to rule that out," per Government Executive. Every official quote here comes through news reporting.

Workday: Friday night, November 13

Workday went down around 5 p.m. Eastern and was back by about 11 a.m. Saturday, per UC Today. Its notice, quoted there, said "tenanted services within the DR Site unexpectedly connected to Production databases" while Workday was validating a weekly maintenance window for disaster recovery tenants. That left data "in an inconsistent state," and Workday placed "all WD1 Production tenants into Maintenance," as Washington University's IT status page reproduced it. Payroll, time tracking and, at some customers, Workday-based single sign-on were affected. Workday has not said why disaster recovery services could reach production.

JPMorgan Chase and United Airlines: Monday, November 16

Chase said it "has experienced a temporary issue related to a software bug in an on-premise hardware system," and that "it is not cyber-related," per AP via OPB. Chase said other banks were affected "only to the extent that they were a party to a transaction with us." By evening, "transactions are processing normally again," per FOX 5 New York.

United reported a "widespread system outage," then said a "technology-related issue that impacted some departing flights" was resolved, per FOX Weather. It told the AP there was "no indication that this was a cybersecurity-related issue," per CP24. Pilots and a union memo pointed to the system that provides final aircraft weights, per Airways Magazine. United has stated no cause. See the United Airlines incident page.

Apple: Tuesday, November 17

iCloud sign-in, the App Store, Apple Pay, Apple Music and device activation had problems. An Apple spokesperson told Engadget that "many of our systems are back up and running after an internal update caused some outages this morning," and that Apple was working to restore the rest. Apple has not said what the update changed, and its quoted statement does not address whether an attack was involved.

What the stated causes have in common

Four of the five stated causes begin with planned or routine work: a July fix met by a routine failover (Microsoft), contractor upgrade work (FAA), a disaster recovery maintenance check (Workday) and an internal update (Apple). Chase names a software bug. That is a pattern in what was said, not evidence of a connection, and none of it is final.

What security and engineering leaders should check

  1. Changes that have not met real load. Microsoft's July defect surfaced in November, during a failover. List recent changes that have not yet run through a failover or peak, and confirm each has a tested rollback.
  2. Failover headroom. Microsoft's failover landed with about 14% of capacity already out. Check what spare capacity your plan assumes, and whether failed health checks push load onto fewer servers until they fail too.
  3. Staged rollout for routine and vendor work. Apple's internal update and the FAA's contractor upgrade both reached far beyond one system. Give maintenance and contractor changes the same staging and monitoring as releases.
  4. Disaster recovery isolation. Test that your DR environment cannot reach production databases, using network and credential boundaries, not configuration alone.
  5. Dependency maps. Workday took single sign-on with it at some customers. Map which logins, payroll and payment flows rest on one vendor.
  6. Phishing after outages. Outage-themed phishing hit within a day of Microsoft's incident. Tell staff now where real status updates come from.

Frequently Asked Questions

Were the November 12 to 17, 2026 outages connected?

No organization or investigator has said they share a cause. Five gave different preliminary causes, and United gave none.

What caused the Microsoft 365 outage on November 12, 2026?

Microsoft's preliminary report names a latent defect, shipped in July, that made gateway servers unresponsive under certain load. A routine Teams failover triggered it, and the failure spread until more than 80% of servers were unresponsive.

Were any of these outages cyberattacks?

Microsoft and United said there was no indication of one, and Chase said its problem was not cyber-related. FAA officials said it did not appear to be an attack but would not rule it out. Apple and Workday have not addressed it on the record.

Get new Scams & Cybersecurity writing

Enjoyed this? Subscribe and tell us what you read most. Scams & Cybersecurity is already ticked for you. No tracking pixels, unsubscribe with one click.

Tell us what you read most (optional)

About DeepakPublicationsAnalysisAll tracks