Back

16/11/2026

The November 2026 Microsoft 365 outage: what happened, and how to prepare for the next one

A network change took Microsoft 365 email down for more than a day, and phishing followed within 24 hours. What happened, and how to prepare.

On Thursday 12 November 2026, a Microsoft 365 outage began that left organisations around the world without reliable email for more than a day. Microsoft says it was a technical fault, not an attack. But within a day, attackers were already using the confusion to send fake recovery emails. Here is what happened, what it shows, and what to prepare before the next outage.

What happened

The trouble started on Thursday 12 November in Exchange Online, the service behind Microsoft 365 mailboxes. Some users could not open their mailboxes or calendars. It then spread to the services that depend on Exchange.

ServiceWhat users saw
Exchange Online and OutlookMailboxes and calendars unavailable; email not sent or received
Microsoft TeamsCalendar and some chat and channel functions affected; calls and meetings kept working
Microsoft 365 CopilotChat failing to load; limited access to work data
Defender for Office 365Safe Links, Safe Attachments, quarantine and Threat Explorer affected
AI assistants connected to Microsoft 365OpenAI reported elevated errors in the ChatGPT connector for Microsoft 365

Public bodies were hit too. The clerk’s office of the US Court of Appeals for the Tenth Circuit could not send or receive email for much of Thursday. Japan’s Ministry of Finance said on Friday 13 November that it could not receive email because of the outage.

Microsoft restarted servers, redirected network traffic, added capacity and reverted a recent network configuration change. Early on Friday 13 November (UTC) it deployed a patch that improved stability, and recovery then moved region by region. It also extended email timeouts to 48 hours so that delayed messages were not lost.

Microsoft’s incident record puts the end of the main impact at 16:10 UTC on 13 November, about 30 hours after it began. Its final update, marking the incident resolved, followed at 04:15 UTC on 14 November. The incident was tracked as MO1550821 for Microsoft 365 as a whole and as EX1550543 for Exchange Online.

What caused it

According to Microsoft’s updates, a process that handles secure connections became overloaded, and the affected servers stopped responding. Microsoft’s preliminary finding is that the incident was introduced by an issue during an earlier network capacity upgrade, which surfaced after a more recent network change.

Microsoft described it as “a technical issue not related to cyber or AI / agents” and said there was no indication of malicious activity. It has said it will publish a preliminary post-incident report within two business days and a final report within five. Until the final report is out, treat the details above as preliminary.

The second wave: outage phishing

The outage was not an attack, but attackers used it. On 14 November, Microsoft Threat Intelligence warned of an active phishing campaign built on the disruption. The emails were first detected on 13 November, while recovery was still under way.

  • They copy real service health notices. The emails cite the genuine incident number, EX1550543, so they look official.
  • They create urgency. They claim that messages held during the outage will be deleted within 24 hours unless the recipient clicks a link to restore them.
  • They follow a pattern. The subject line is “Outage” followed by a six-character code. The sender name ends in “Outage Notification” or “Service Health”.
  • The link leads elsewhere. It passes through a click-tracking link unique to each recipient and lands on a site that has nothing to do with Microsoft.

Microsoft reported the sender domains moments-notice[.]com, planetmaison[.]com and newbooker[.]com, and the destination reallifeaccounting[.]com. The targets were organisations in several US sectors. Microsoft says the activity is consistent with credential phishing (stealing passwords), but it has not confirmed that passwords were collected or that malware was delivered.

The lesson for staff is simple. Recovery from a real outage never requires you to sign in through a link in an email. A genuine incident number does not make the link safe.

What this outage shows

  • One provider can be a single point of failure. When email, calendar, chat and files all sit with one vendor, one fault stops all of them.
  • Dependencies reach further than you think. Meeting invitations, security tools and AI assistants all rely on the same mail platform. Many teams only found that out when they stopped working.
  • Large providers fail too. This was a routine network change, not an exotic attack. No amount of scale removes that risk.
  • Confusion is an attack surface. During an outage people expect unusual emails and are impatient to get back to work. That is the best moment to phish them.

What to prepare before the next outage

  1. Agree on a backup channel now. Decide where staff will talk if email and chat are both down, and make sure it does not depend on the same provider. Test it once.
  2. Keep key contacts outside the affected system. A customer list or escalation sheet stored only in the mailbox is no help when the mailbox is down.
  3. Know how you will reach customers. Prepare a short status message and a way to publish it, such as your website or a phone line, that does not need your mail platform.
  4. Check status at the source. Use the Microsoft 365 admin center or Microsoft’s official status page, not links in emails.
  5. Warn staff early. As soon as a large outage starts, tell people that fake recovery emails will follow and that no real fix asks for a sign-in through an email link.
  6. Hunt for the campaign. Security teams can search Defender for Office 365 Threat Explorer and Defender XDR advanced hunting (the EmailEvents, EmailUrlInfo and UrlClickEvents tables) for the sender domains, subject pattern and destination above. A recorded click needs follow-up even if nothing else looks wrong.
  7. Confirm Safe Links is on. Safe Links checks a URL at the moment the user clicks it, which helps when a link turns malicious after delivery.
  8. Map your dependencies. List which systems, integrations and AI tools stop when your mail platform stops, and decide which of them need a manual fallback.

The bottom line

A cloud outage is a business continuity problem first and a security problem a few hours later. You cannot prevent your provider’s outage, but you can decide in advance how your people will communicate, how they will check what is real, and what they should never click.

Want help reviewing your continuity plan or your email security? Contact the LOGIQ team.

Sources

Photo: Brett Sayles / Pexels