McMaster was one of many organizations caught by the outage. It is worth a closer look because the notice is a clear example of how an institution keeps working when its email, chat and document tools stop responding.
What McMaster told its campus
The 3 p.m. notice said Microsoft knew about the problem and was working to restore services. McMaster sent readers to its own status page, McMaster University Status, for updates. That page is the university's own reporting channel and is separate from Microsoft's service health reporting.
The 4:30 p.m. update was more practical. Here is what McMaster said, along with what it did not say:
- Classes continue. Classes would run as scheduled. That promise covers classes and does not say every university system was working.
- Other services might be affected. Services that depend on Microsoft 365 could be disrupted or delayed. McMaster named online forms, service requests, automated approvals and reporting tools as examples. It called them possibly affected and did not say all of them had failed.
- Urgent requests. For time-sensitive service requests, people should contact the service area directly by another method where one exists.
- Requests already sent. People should not send the same request again. They should check its status once services are back.
The "What you can do in the meantime" section added four more points:
- Use Zoom instead of Microsoft Teams for virtual meetings, where available.
- For urgent messages, use the phone or another channel instead of email.
- If shared documents can't be opened, use locally saved files where appropriate and follow the university's information security rules.
- Emails may be delayed or stuck in the outbox. Once service returns, check that important messages actually went out.
In short: classes were protected, back-office processes that rely on Microsoft 365 might be delayed, and people were told to use other channels and check their work once service returned.
The wider outage
McMaster described the problem as North American. Microsoft's own description changed over the afternoon. Microsoft 365 first confirmed just before noon that users were having trouble getting into Exchange Online mailboxes "through multiple connection methods in the Americas." Later the company expanded that: "While we're working to mitigate impact, we've confirmed the issue may affect Microsoft 365 service connectivity for users in any region." McMaster's "North America" wording matches the early messaging. It was never a firm boundary for the whole incident.
For Microsoft 365 administrators, Techaeris reported that the incident is tracked as EX1550543 in the Microsoft 365 admin center. The EX prefix points to Exchange Online, which fits the mailbox access problems at the center of the outage. Techaeris also listed Exchange Online, Teams, SharePoint, Copilot 365, and Defender XDR as affected.
Here is how big the outage looked on Downdetector:
| Source | Snapshot |
|---|---|
| Techaeris | User reports began climbing at 12:39 p.m. ET on Downdetector, which logged 128,874 Outlook complaints by 1:09 p.m. |
| Al Jazeera | By just before 3pm (20:00 GMT), that figure had fallen to about 77,000. |
| Reuters | More than 97,000 Outlook reports at 2:07 p.m. ET, more than 38,000 for Microsoft 365 and about 4,500 for Teams |
These numbers don't line up exactly because they were taken at different times. They are also user reports, not counts of affected accounts. Reuters noted that the real number of affected users may be different.
Al Jazeera described the symptoms: some users saying emails weren't being sent or received, while others couldn't open their inboxes. Deadline reported that enterprise and business customers of Microsoft 365 were experiencing problems connecting to email via the Exchange server as well as accessing other services. That is the kind of tenant that universities run.
What Microsoft says it did
Reuters reported a Microsoft statement saying it was "aggressively pursuing all pathways to reduce impact." According to Reuters, Microsoft took three steps:
- It paused all deployments.
- It moved additional traffic off the affected infrastructure.
- It restarted affected servers to bring them back into service.
Parade cited the official Microsoft 365 Status account, which said the company has paused deployment and configuration changes while they review telemetry. Al Jazeera's summary was that servers unexpectedly went offline, disrupting Outlook email, Teams, Copilot and other services.
The cause is still uncertain. Microsoft has given an early explanation, but that is not a full root-cause analysis, and readers should not treat it as one. McMaster's notice also gave no time for full restoration. As of its 4:30 p.m. update, the disruption was still going on.
There was also a separate, earlier issue. Reuters reported that Microsoft had fixed a Teams disruption in North America earlier in the day and called its impact fully resolved. Both problems happened the same day with overlapping symptoms, but nothing confirms that they share one cause.
AI tools that connect to Microsoft 365 were affected too. Reuters reported that OpenAI was investigating broken connections to Microsoft and tools not working as expected. Anthropic said a problem at an upstream provider was affecting its Microsoft 365 connector. Any workflow that relies on Microsoft 365 can fail when Microsoft 365 does.
Microsoft's second big outage this year
This is not Microsoft's first major Microsoft 365 outage in 2026. In January, The Register reported that Microsoft blamed "a portion of service infrastructure in North America that is not processing traffic as expected." Recovery took a long time. By 0533 UTC, almost ten hours after the company first admitted there was a problem, Microsoft reported: "We've restored access to the affected services and mail flow remains stable."
That earlier incident peaked at about 15,000 Downdetector reports, according to The Register. Thursday's Outlook reports were roughly eight times higher. Downdetector counts partly reflect how much attention an outage gets, so they are a rough guide to scale at best. Still, the gap is large.
What IT teams can take from this
Most of McMaster's advice works for any organization. Here is how it applies more broadly:
- Name your backup channels before you need them. McMaster could recommend Zoom because Zoom was already available on campus. A backup tool helps only if people can sign in to it while the main platform is down.
- Stop duplicate requests. Forms and approval workflows often look like nothing happened during an outage, so people submit again. Telling users to wait and check status saves a lot of cleanup later.
- Check the outbox after service returns. Messages that look sent may still be queued. Look at the Sent folder before assuming a deadline email arrived.
- Keep security rules in force. McMaster told people to work from local files while following its security requirements. It did not give permission to move sensitive data to personal storage or get around access controls.
- Admins should follow the incident ID. Track EX1550543 in the Microsoft 365 admin center instead of relying on social media or outage trackers. Those tell you people are frustrated. They don't tell you how your tenant is doing.
Some organizations will see this outage as a reason to spread their tools across more than one platform. That is a reasonable thought, but it costs money and adds complexity. Most teams will get more value from preparation: alternate channels people already know how to use, written outage steps, and clear status updates.
McMaster's status page and Microsoft's admin center are where a final all-clear will appear. When it does, check your outbox before you trust that everything went through.