Blog

13 Tech Outages in One Week: What Really Happened, With Sources

November 18, 2026By Russell Wyatt, Chief Executive Officer

If you have asked us this week whether the internet is under attack, you are in good company. Our service desk has fielded that question over and over from clients, friends and neighbors, and it is a fair one. In seven days, Microsoft 365, AWS, the FAA, Workday, Chase, United, Apple and others all had trouble at once.

So we did the homework. We went back to each provider's own status page or official statement and wrote down what they actually said. Here is the short version: as TechCrunch reported on November 17, none of the outages in the past week have been attributed to cyberattacks. [22]

The Short Answer

Most of this week was ordinary technology having a bad day: a software update, a backup test, a hidden bug, a busy server, routine maintenance. A few providers have not published a cause yet, and we say so plainly below rather than guess. Every fact links to its source, so you can check our work.

What Happened, Outage by Outage

Times are Eastern. Reporting is as of November 17, and providers may update their findings.

Nov 11: AWS Tokyo

When: 9:46 AM to 3:31 PM ET (overnight in Tokyo). [1]
What happened: New EC2 servers failed to launch or connect, mostly in one zone. Servers already running kept working. [1]
Root cause: An internal system that passes along network setting changes fell behind. [1]
Status: AWS: Networking Delay

Nov 12 to 13: Microsoft 365

When: Nov 12, 11:20 AM to Nov 13, 11:10 AM ET. [3]
What happened: Outlook email was hit hardest. Microsoft published a detailed account of what went wrong. [2][3]
Root cause: A hidden bug from a July update, triggered by extra traffic from a routine Teams failover while about 14% of gateway servers were already out. [2][3][4]
Status: Microsoft: Not Cyber

Nov 12 to 13: Claude M365 Connector

When: Nov 12, 11:03 AM to Nov 13, 11:40 AM ET. [5]
What happened: Claude's Outlook and Microsoft 365 tools were waiting on Microsoft. When Microsoft recovered, so did they. [5]
Root cause: The Microsoft 365 service incident above. [5]
Status: Downstream of Microsoft

Nov 13: Claude

When: Friday morning, resolved by 10:38 AM ET. [5]
What happened: Separate sign-in and error issues on claude.ai, the API and Claude Code. Anthropic deployed fixes. [5][6]
Root cause: Not publicly disclosed. Anthropic did not link it to Microsoft. [6]
Status: Cause Not Disclosed

Nov 13 to 14: Workday

When: Friday afternoon to 11 AM ET Saturday. [8]
What happened: A backup safety drill went sideways. Workday placed systems into maintenance and restored service Saturday. [7][8]
Root cause: During a weekly disaster recovery check, backup-site services connected to live production databases. [7][8]
Status: Workday: Not Cyber

Nov 13: FAA

When: About 3:30 PM to 8:15 PM ET. [9]
What happened: Safety first: the FAA paused flights out of an abundance of caution, then resumed once systems were stable. [9]
Root cause: A flood of network traffic. Early findings point to a radar site in Texas and a radio site in Arkansas, where a new router was being installed. Investigation ongoing. [9][10][11]
Status: Zero Evidence of Cyber

Nov 15: AWS US-EAST-1

When: 6:05 AM to 9:21 AM ET. [12]
What happened: Sign-in trouble for IAM Identity Center and some load balancers. Engineers were engaged within 5 minutes. [12]
Root cause: Not published by AWS. [12]
Status: Cause Not Disclosed

Nov 16: JPMorgan Chase

When: From about 10 AM ET, fixed Monday evening. [14]
What happened: Some transactions ran slow, but customers could still reach their accounts and funds. [13]
Root cause: A software bug in Chase's on-premises hardware systems. [13]
Status: Chase: Not Cyber

Nov 16: United Airlines

When: About 4 PM to 6 PM ET. [15]
What happened: Some departures waited at the gate. Flights already in the air were not affected. [15]
Root cause: A technology issue that blocked final aircraft weight figures, per the pilots' union. Full cause not published. [15][16]
Status: United: No Sign of Cyber

Nov 17: Apple

When: From about 11:45 AM ET. [18]
What happened: iCloud sign-in and other services slowed. Many systems were back up the same day. [17][18]
Root cause: An internal update, according to Apple. [17]
Status: Apple: Not Cyber

Nov 17: NSA

When: Times not disclosed. [19]
What happened: Some internal systems degraded. Mission-critical capabilities stayed fully operational. [19]
Root cause: Temporary degradation during scheduled routine maintenance. [19]
Status: NSA: Routine Maintenance

Nov 17: Gmail

When: Reports peaked around noon ET. [20]
What happened: Users reported mail trouble. Google said it saw no evidence of a broad Gmail disruption. [20]
Root cause: No root cause or broad outage established. [20]
Status: No Broad Outage Found

Nov 17: Charles Schwab

When: Working again around 10:38 AM ET. [21]
What happened: A brief webpage loading issue. Trading stayed available and the mobile app worked. [21]
Root cause: Not publicly identified. [21]
Status: Cause Not Disclosed

The Good News Inside a Rough Week

Behind every one of these headlines was a team of people who found the problem, fixed it and told the public what they knew. Microsoft published a detailed account of what went wrong. [2][3] The FAA paused flights out of an abundance of caution and resumed once systems were stable. [9] Chase customers could still reach their accounts and funds. [13] United flights already in the air were not affected. [15] AWS engineers were engaged within five minutes. [12]

It is also a reminder of how connected our tools are. When Microsoft 365 struggled, Claude's Outlook and Microsoft 365 tools waited on it and recovered when Microsoft did. [5] A problem at one vendor can ripple into everything that depends on it, which is why planning beats panic.

Is Your Business Ready for the Next One

Outages will happen again. The businesses that ride them out calmly usually have four things in place:

  • A second way to communicate when email is down
  • Backups you have actually tested
  • A simple written downtime plan
  • A map of the vendors your work depends on

Want the whole week on one page? Download the one-page brief (PDF), with a source for every fact. And if you would like a second set of eyes on how ready your business is, talk with our team.

Sources

  1. AWS Health Dashboard, AP-NORTHEAST-1 'Elevated API issues and Launch Failures' (archived). statusstack.com
  2. Microsoft preliminary post-incident report EX1550543 / MO1550821, quoted by Gadgets 360. www.gadgets360.com
  3. Ars Technica, 'Three small failures combined to take down Microsoft 365'. arstechnica.com
  4. CNBC, Microsoft statement: no malicious activity detected. www.cnbc.com
  5. Anthropic, Claude status page (status.claude.com). status.claude.com
  6. Gulf Business, Claude connector and separate sign-in incident. gulfbusiness.com
  7. Workday Community Status notice, as reposted by Washington University IT. statusstack.com
  8. ABC10, Workday spokesperson statement. www.abc10.com
  9. FAA, 'Nationwide Ground Stop: Questions and Answers'. www.faa.gov
  10. U.S. DHS, Secretaries Mullin and Duffy briefing, Nov. 13. www.dhs.gov
  11. AVweb, investigation status and router pause. avweb.com
  12. AWS Health Dashboard, US-EAST-1 'Increased Connectivity issues' (archived). statusstack.com
  13. CNBC, JPMorgan Chase statement. www.cnbc.com
  14. Banking Dive, Chase timing and resolution. www.bankingdive.com
  15. AirlineGeeks, United Airlines statement. airlinegeeks.com
  16. NBC News, United outage and pilots' union statement. www.nbcnews.com
  17. CBS News, Apple spokesperson statement. www.cbsnews.com
  18. MacRumors, Apple outage timing. www.macrumors.com
  19. NSA, 'NSA statement on network issues', Nov. 17. www.nsa.gov
  20. Anadolu Agency, Google spokesperson statement on Gmail. www.aa.com.tr
  21. Business Insider, Charles Schwab spokesperson statement. www.businessinsider.com
  22. TechCrunch, 'none of the outages in the past week have been attributed to cyberattacks'. techcrunch.com
Russell Wyatt
Written byRussell Wyatt
Chief Executive Officer, PC Techware

Russell leads PC Techware and its cybersecurity and compliance practice, including CMMC and NIST readiness.

Connect on LinkedIn

Ready for Security That Holds Up Under Scrutiny?

Pick the next step that fits. Every one of them starts with a real conversation, not a sales script.

CallFree Scan