US outage week: Microsoft 365, FAA, banks, NSANet and Apple, what each cause is and what Rubio said
Rubio says there is zero evidence the November outages are linked; here is each stated cause, who gave it, and what is still unexplained.
· 6 min read
Written by AI from the sources linked below. How we use AI
On this page
Five separate US technology failures in six days, across Microsoft 365, air traffic control, three large banks, an NSA network and Apple services, have led to public talk of a coordinated attack. On 18 November, Secretary of State Marco Rubio said there is "zero evidence" the incidents are connected. Here is what each company and agency has actually said, where the accounts conflict, and what remains unexplained.
Key points
- Microsoft 365 failed on Thursday 12 November; Microsoft later blamed a software defect introduced by a July fix, according to The Register.
- The FAA's nationwide ground stop on Friday 13 November has been linked by officials to contractor L3Harris's work on routers; the exact technical cause is unconfirmed.
- On Monday 16 November customers of JPMorgan Chase, Bank of America and US Bank reported problems; Chase said the cause was "a software bug in on-premise hardware systems" and that it saw no evidence of a cyber incident.
- On Tuesday 17 November the NSA's classified network went offline; the agency called it a "temporary degradation during scheduled routine maintenance."
- Rubio said on 18 November there is "zero evidence" the outages are linked or that the US is under systemic attack.
What happened
Microsoft 365, Thursday 12 November. A cloud outage took down Exchange Online, Teams and Intune. Services were reported restored to all customers on 13 November. Microsoft said the cause was a technical issue and not malicious activity, and The Register reported on 16 November that Microsoft blamed a software defect introduced in a fix shipped in July. At the Ignite keynote on 17 November, Satya Nadella referred to "last week's massive M365 outage." We covered the first day of it in our Microsoft 365 outage report.
FAA ground stop, Friday 13 November. All departing US flights were halted for several hours, and flights were moving again on Saturday, though delays and cancellations continued. Transportation Secretary Sean Duffy said work by the FAA's telecommunications contractor L3Harris may have triggered it. Two routers were taken from remote FAA sites in Texas and Arkansas to L3Harris's headquarters in Florida for testing, with engineers from the FAA, Juniper Networks and Cisco helping, according to Fox 4. Officials said the FBI and federal partners are investigating and that there was no sign of a foreign cyberattack. Our earlier coverage is here.
Banks, Monday 16 November. Customers of JPMorgan Chase, Bank of America and US Bank reported service disruptions. CBS says all have since been resolved and gives no cause. The Register quotes Chase's post attributing its outage to a software bug in on-premise hardware, with no evidence it was cyber-related.
NSANet, Tuesday 17 November. CBS reported, citing multiple US officials, that the internal network used to store and share classified information went offline. The Register reports the NSA's description: a temporary degradation during scheduled routine maintenance. Neither gives a technical cause.
Apple. Apple Support, iCloud, the App Store and Apple Pay had problems and some retail stores were affected, according to The Register, which gives no cause. CBS says services were disrupted and have since been restored.
Background
None of these is the first outage of its kind, and the sources do not say they share a cause. What is new is the clustering in time, and the sensitivity of the systems involved: air traffic, retail banking and a classified government network. Public distrust grew enough that Bloomberg ran a piece headlined on "tech paranoia" after the outages, and the Rubio statement came in answer to reporters' questions while he was travelling with President Trump in Manila.
For the FAA incident, the public record has a gap worth noting. The Register's summary says the stated cause was faulty copper wiring during an L3Harris network upgrade. Other reports describe two newly installed Juniper routers flooding the agency's ageing network with connection requests; Duffy was reported as saying the routers made over 1,000 connection attempts every four to five seconds, though we could only read that figure in a secondary summary. The Wall Street Journal, as relayed by Silicon Report, said FAA technicians turned off a specific networking feature on all the agency's Juniper routers. An HPE spokesperson, whose company owns Juniper, said the routers appeared to work as designed: "They did not fail." These accounts may describe the same event from different angles, but they are not identical, and the official root cause has not been published.
How the explanations compare
| Incident | Date | Stated cause | Who says so |
|---|---|---|---|
| Microsoft 365 | 12-13 Nov | Software defect from a July fix | Microsoft, via The Register |
| FAA ground stop | 13 Nov | Contractor work on routers; cause not confirmed | Transportation Secretary Duffy, via Fox 4 |
| Chase and other banks | 16 Nov | Chase: software bug in on-premise hardware | Chase, via The Register |
| NSANet | 17 Nov | Scheduled routine maintenance | NSA, via The Register |
| Apple services | around 17 Nov | Not given | Not stated |
Read down the column and the pattern the officials describe is ordinary: a software defect, a contractor upgrade, a bug, maintenance. That pattern is also what makes people uneasy, because ordinary failures in different organisations landing in the same week looks like something more.
What it means
For engineers, the most useful reading is about change management. Three of the four stated causes (a July software fix, a router upgrade, scheduled maintenance) are changes made by the operators or their contractors, not attacks. A change that sat in production for months before failing, as with the Microsoft defect, is a reminder that testing after release matters as much as testing before it.
There is also an argument about speed. Senator Mark Warner told CBS there was no indication of an attack, and said AI is pressuring companies and agencies to patch vulnerabilities quickly, straining systems not built for that pace. Rubio, asked about AI's role, declined to speculate and deferred to the individual companies. A separate report in the Washington Sun, relayed by Techmeme, said Wall Street executives and the NSA told Congress that this week's outages came from systems overwhelmed by a rush to patch vulnerabilities found by AI. We could not open that article and found no second outlet carrying it, so treat it as unconfirmed. If it holds, it ties the week's events to the patch-volume surge we have followed in our coverage of the 2026 CVE count passing 100,000.
For the public, the practical effect is that each failure is individually small and recoverable, but the combined effect is a loss of trust. Rubio's blanket statement is useful for ruling out a single attacker; it does not say whether the shared cause might be a common practice, such as how quickly updates are pushed.
What we don't know yet
- The confirmed root cause of the FAA ground stop, including whether it was a router feature, a wiring issue or both.
- The cause of the Apple disruption and of the other banks' outages; only Chase has given one.
- What failed on the NSA network beyond "maintenance."
- Whether the AI-driven patching explanation reported by the Washington Sun is accurate. Warner's version is a general view and not a finding about any of these incidents.
- Whether investigations by the FBI and others will publish findings.
What to watch next
Watch for the FAA's and L3Harris's post-incident report, Microsoft's full post-incident review, statements from the banks, and any congressional hearing referred to in reports about executives and the NSA briefing Congress. Also watch whether any of the companies describe a common factor in how changes were rolled out.
Sources
- CBS News: Rubio says tech outages are not connected
- The Register: Daily disruptive outages that ground planes and halt work
- The Register: Satya sorry for SaaS-y SNAFU
- Fox 4: FAA ground stop, routers in Texas and Arkansas being tested
- Silicon Report: FAA disables Juniper router feature, citing the Wall Street Journal
- Washington Sun via Techmeme: bank CEOs and NSA tell Congress AI-driven patches behind outages (unconfirmed)
Written by
The official account of the people who run audiotale.net: announcements, conversation starters and the newsroom's drafts. Posts written with AI say so. Every other account here is a real person.
Read next
NGINX 1.31.9 and 1.30.7: CVE-2026-123007 HTTP/2 RCE and four more flaws fixed
F5 patched five NGINX memory-safety bugs on 18 November; the HTTP/2 flaw CVE-2026-123007 dates to 2015 and Wiz says it can give unauthenticated code execution.
audiotale.net team8 min read
agent-framework 1.26.1 and 1.26.2 on PyPI steal credentials: what to do
Two malicious releases of Microsoft's agent-framework package ran a credential stealer at Python startup on 17 November before PyPI quarantined them; 1.26.0 is the clean version.
audiotale.net team6 min read
CVE-2026-129448: Apache Roller XML-RPC RCE exploited in attacks, fixed in 6.1.6
Apache says unauthenticated attackers are already exploiting a critical deserialization flaw in Roller 4.0.0 through 6.1.4; here is how to patch or mitigate.
audiotale.net team8 min read