An IT administrator investigates a suspicious Microsoft 365 outage email flagged as likely phishing.
A week of outages has produced a lot of scary headlines and not much evidence of a common cause. For Windows and Microsoft 365 administrators, the sharpest lesson is a small one. Real incident numbers can be turned into phishing bait.

This roundup is based on MLQ.ai's running tracker, which covers Nov. 12 to Nov. 17, 2026. I added Microsoft's own statements and independent security reporting. The tracker lists Microsoft 365, the FAA's air traffic network, Workday, JPMorgan Chase, United Airlines, NTT Docomo, NSA Net and Apple's iCloud. Two other items sit alongside the outages: separate cyber incidents at Equinor, Ipsos, Infoblox and Optimizely.

An IT administrator investigates a suspicious Microsoft 365 outage email flagged as likely phishing. Microsoft 365: a latent bug, not an attack​

According to the tracker, the outage began at about 11:20 a.m. ET on Nov. 12. It hit Outlook, Teams and Copilot, and Microsoft declared full restoration at about 1:09 a.m. ET on Nov. 14.

Microsoft's Nov. 15 preliminary post-incident report covers more than mail. The listed impacted services include Exchange, Copilot, Teams, SharePoint, OneDrive, Defender, Purview, Fabric, Intune, Planner, Power Apps, To Do, Power BI and Universal Print. Microsoft confirmed problems across several Exchange connection methods, including Outlook, Outlook on the web, Exchange ActiveSync, and Messaging and REST APIs.

The cause was a defect and a capacity squeeze, not an intrusion. The tracker says the report points to a latent defect in a fix for its network gateway, applied in July. Microsoft's status account, as relayed in a post, describes a bug that sat quiet for three months until a Teams failover and 14% of gateway capacity being offline pushed load past it. The same post says load tests ran on smaller hardware, so they never saw it. The tracker adds that more than 80% of servers became unresponsive. Reporting from SecurityWeek says Microsoft stated the outage was not caused by a cyberattack.

Admins can find the report in the Microsoft 365 admin center. Microsoft says it is available under MO1550821 and EX1550543. A final report is expected later.

The phishing campaign riding on the outage​

Attackers moved quickly. Fraudulent emails mimic Microsoft 365 service health notices and cite the real incident, EX1550543. Some falsely say messages were held during the outage and will be deleted within 24 hours unless the recipient clicks a "Restore messages" link. The attackers personalized the recipient's name and organization, with phrases such as "Outage notification" or "Service health."

Details reported so far:

  • Sending domains: Microsoft linked three domains to the campaign: moments-notice[.]com, planetmaison[.]com and newbooker[.]com.
  • Redirect chain: Customized SendGrid click-tracking links sent recipients on to reallifeaccounting.com.
  • Subject lines: The emails generally had subjects beginning with "Outage -" followed by a six-character code.
  • Assessment: Microsoft says the activity is consistent with credential phishing, but credential collection and malware delivery have not been confirmed.
  • Protection: Microsoft Threat Intelligence said Defender blocks the malicious domains and Defender for Office 365 blocks and quarantines the emails. It also published a blog with indicators of compromise and Defender XDR hunting queries.

What admins should do​

  1. Tell users that Microsoft does not require a "restore" click. Outage notices are not a reason to enter credentials.
  2. Point staff at the admin center for status. Microsoft advised organizations to verify service incidents directly through the Microsoft 365 admin center.
  3. Search mail logs for the three domains above and for the "Outage -" subject pattern. Block the domains where possible.
  4. Check for clicks. If anyone clicked, treat the account as at risk. Reset the password, revoke sessions and review sign-in logs. This is standard incident-response practice, not Microsoft's guidance for this specific campaign.
  5. Apply phishing-resistant MFA where you can. That is general industry advice.

A fake outage notice works because the real outage makes users expect a service notice. It is also good timing for the attacker, since a legitimate incident ID makes the lure feel official.

The FAA ground stop: the biggest outage, with its own cause​

The FAA outage is not a Microsoft story, but it dominated the week. The tracker puts the nationwide ground stop on Nov. 13, with departures resuming at about 8:15 p.m. ET. Officials blame overloaded legacy switches and routers in the FAA's communications network.

Transportation Secretary Sean Duffy tied the investigation to a router installed in El Dorado, Arkansas. It was installed during L3Harris-managed work to replace copper lines with fiber. The FAA has since paused installation of new Juniper routers while it investigates. Thirteen Senate Commerce Committee Democrats, led by Maria Cantwell, asked it to pause pending testing.

The FAA's own account, as summarized in my research notes, describes legacy equipment destabilized by a flood of connection attempts. It reports about 1,000 attempts every four to five seconds. It puts instability at 17 of 22 Air Route Traffic Control Centers and about 12,462 cancellations. The agency said its initial information was preliminary. I could not open that FAA page directly, so treat these figures as the agency's reported numbers.

Two points need care:

  • The FAA expanded its AI-supported SMART planning tool to all 22 centers three days earlier. Officials and the developer see no link, so the timing alone proves nothing.
  • CBS News found that the Transportation Department's inspector general had warned about the L3Harris-run network's reliability since 2006. That is context for the network's condition, not a finding on the root cause.

The rest of the week​

IncidentWhat was reportedCause, as stated
WorkdayRestored by 11 a.m. Nov. 14 after employees were locked out of login, timekeeping and payThe tracker gives no detailed cause
JPMorgan ChaseTransactions posting normally by about 6:30 p.m. ET Nov. 16Software bug in on-premises hardware, not a cyberattack, per Chase
United Airlines"Widespread system outage" just before 5 p.m. ET Nov. 16 delayed departures; resolved that eveningNot specified in the tracker
NTT Docomo5G SA service restored 3:58 p.m. JST Nov. 16, its third nationwide outage this monthSwitch congestion, per Docomo
NSA NetDisrupted Nov. 17, per CNN citing six peopleA US official cited "temporary degradation during scheduled routine maintenance"
AppleiCloud sign-in issue for all users from 11:45 a.m. ET Nov. 17, plus App Store and Apple Pay problemsNo cause given in the tracker
GmailThousands of Downdetector reports Nov. 17Google says it sees no evidence of a broad disruption

Apple's status was still ongoing at the tracker's snapshot, so don't read it as the current state.

Separate cyber incidents​

These are distinct events, and nothing in the evidence links them to the outages:

  • Equinor says it detected an ongoing cyberattack on some IT systems. It reports no effect on operations or production and no indication that data was stolen.
  • Ipsos reported an incident affecting one internal system and filed a complaint with authorities.
  • Infoblox confirmed hackers reached company information through a compromised third-party data migration vendor.
  • Optimizely says hackers used a signup flaw to plant card- and crypto-stealing code on customer sites. Salesforce disabled its Optimizely scripts as a precaution, and Equifax stopped using the service.

For developers and web admins, the Optimizely and Infoblox cases are a reminder about third-party scripts and vendors. Both attacks came through suppliers, not the victims' own perimeters.

Is it all connected?​

Probably not, going by what's on record. White House National Cyber Director Sean Cairncross told Semafor the FAA stop and other outages were not a coordinated cyberattack. Microsoft, Chase and Docomo each named a technical cause. The NSA explanation is maintenance. A one-week cluster of failures invites a single-villain story, but the stated causes here are software bugs, capacity limits and legacy hardware.

The cyber incidents are real, but they involve different victims and methods. Treat any claim of a shared cause as unsupported until investigators say otherwise. Several causes remain open, including the full FAA root cause and the Apple and Workday explanations.

Takeaways for IT teams​

  • Microsoft 365's outage came from a latent bug and reduced capacity, not an attack. The preliminary report is in the admin center.
  • The phishing lure uses a real incident ID. Verify outage notices only through the admin center.
  • Block the three listed domains and check mail logs for the campaign.
  • Test failover and capacity on production-scale hardware. Microsoft's load tests reportedly missed the bug because they ran on smaller hardware.
 

References

  1. Microsoft Rules Out Cyberattack in Outage, Warns of Phishing Campaign - SecurityWeek securityweek.com
  2. The outages: latest news and analysis - MLQ.ai MLQ.ai 2026-11-17T19:58:11.824962+00:00
  3. Microsoft 365 Status on X: "We have released the Preliminary Post-Incident Report of last week’s outage, which is accessible via the Admin center under MO1550821 and EX1550543." / X x.com