Microsoft 365 outage: one provider's list of services that may be hit

One provider, many tools

Many companies may not think of their email, chat, file storage and security tools as one system. A Microsoft 365 outage on Thursday raised the question of how closely they are tied together.

Wired reported that Microsoft's 365 business and enterprise services had a major outage on Thursday, November 12. Users could not get into their accounts.

Microsoft's status page named five services that "may include" those affected: Exchange Online, Teams, SharePoint Online, Copilot 365 and Defender XDR. That is a list of possible impact, not proof that all five failed.

A Microsoft spokesperson told Wired the issues were "primarily issues with Exchange Online." The detailed reports Wired relayed concern only Teams and Exchange.

Even so, the list points to a possible dependency. A company may think it has bought five tools. It may in fact rely on one provider's infrastructure for all of them. That is the publication's view, not Microsoft's.

What happened, in order

Wired's timeline begins around 10:40 am ET. Microsoft said it was investigating Teams problems in North America. By just before 1 pm ET, it reported trouble reaching Exchange Online mailboxes "through multiple connection methods in the Americas."

Microsoft later said its network had "elevated load across multiple regions." It said the issue was servers going offline, not extra traffic.

Roughly an hour after that, Microsoft described a change of tack: it was sending connection traffic through other parts of its global network, hoping to even out the demand on the affected systems.

Microsoft also said it had paused all deployments and was restarting affected servers. A pause on deployments means engineers stop pushing new software changes while they work on the fault.

Downdetector tracks user reports of outages. It showed a large spike around noon ET. Reports began falling around 1:15 pm ET.

The problem was not over. In an update just before 3 pm ET, Microsoft said servers were still failing automated health checks and dropping out of service. It was trying to reboot them.

What a health check does

Large cloud services spread work across many servers. A health check is an automated test that asks each server whether it is working. Servers that fail are taken out of service, and traffic goes to the rest.

This design usually protects users when one server breaks. When many servers drop out, the remaining ones may have to carry more traffic. That is a general possibility, not something Microsoft said.

What Microsoft did say is narrower. Servers were failing health checks and going offline. It moved traffic to balance the load. It did not state why the checks were failing.

Wired reported that the cause was unclear. It also reported no indication of a cyberattack. Configuration errors, Domain Name System problems and attacks can all cause connection errors. Any of those is a guess until Microsoft says more, so this story does not pick one.

The people on the other end

Think of the person whose job runs on email. Then think of the incident team that would normally say "mail is down" in Teams. Teams had reported problems too.

Microsoft also listed Defender XDR, a security product, among services that may be affected. Neither Microsoft's list nor Wired's reporting says whether Defender XDR failed.

That raises a question for anyone who signs off on a platform bundle. Consolidation can cut licence cost and admin work. It can also mean the tools you use to coordinate a response share a provider with the thing that broke. That is the publication's observation, not a finding from the outage.

Wired put the event in context. Major Microsoft outages are relatively rare, it said, but not unprecedented. In July 2024, a CrowdStrike update caused Windows crashes worldwide. A separate Azure outage hit around the same time. Wired later clarified that the two were unrelated.

This is one incident. It shows what a failure can look like, not how often one occurs.

Questions to put to your team

First, if email and Teams are down, how does the incident team reach each other? A phone tree or a second messaging tool outside Microsoft would answer it.

Second, where do staff see status updates? Microsoft posted notices on its 365 status page on X. Does your IT team watch it?

Third, which security alerts and logs depend on the same provider as your mail? Defender XDR was on Microsoft's list of services that may be affected. Ask how your team would investigate during an outage.

Fourth, which customer-facing processes need an Exchange mailbox to work? In many organisations, order confirmations, approvals and password resets may. Check yours.

Fifth, what does your contract say about outages? Has anyone rehearsed running the business without these tools for an afternoon?

A cloud suite works like a power utility. Most days nobody thinks about it. On the day it fails, the only plan that helps is the one made beforehand.

Produced by the WebPulse Newsroom with AI assistance from the original reporting credited below, and checked against that source by our editorial review. How we use AI.
Original reporting: Wired.