Outlook Outage EX1550543: Microsoft Says Unavailable Servers, Not Traffic Surge, Strained Exchange Online

What happened​

The Verge's Richard Lawler reported that, within about an hour, people around the world said they couldn't reach their Outlook email in either the web client or the apps. The Verge quoted Microsoft's Microsoft 365 Status account saying that "a portion of infrastructure is experiencing elevated load across multiple regions," which the company said was likely contributing to the connectivity errors.

The first public acknowledgment was narrower. According to Yahoo Tech, Microsoft posted at 12:54 p.m. ET that it was investigating "reports of users experiencing issues accessing Exchange Online mailboxes through multiple connection methods in the Americas". The scope then grew. ABC News reported that Microsoft later said "we've confirmed the issue may affect Microsoft 365 service connectivity for users in any region."

The incident ID admins need​

Tenant administrators should look for this ID in the Microsoft 365 admin center: more details can be found in the Microsoft 365 admin center under EX1550543. How-To Geek reported that the affected connection methods could include Outlook on the web, Messaging API (MAPI), Exchange ActiveSync (EAS), and the Representational State Transfer (REST) API.

That list explains why the outage reached so many devices:

• Outlook on the web: browser access to mail.
• MAPI: the protocol classic Outlook for Windows uses to connect to Exchange Online.
• Exchange ActiveSync: used by many phone mail apps, including built-in iOS and Android mail clients connected to work accounts.
• REST API: used by newer clients and many third-party integrations.

When the web client, desktop app, phone and integrations fail at the same moment, the problem is almost always on the service side. Rebuilding Outlook profiles wouldn't have helped.

Section summary: Exchange Online, logged as incident EX1550543, had connectivity failures across several access methods. Microsoft first described the problem as limited to the Americas, then said users in any region could be affected.

How Microsoft's explanation changed​

This part of the incident deserves the most attention, because the explanation changed over the afternoon.

1. Step one: elevated load. Microsoft first said some infrastructure was under elevated load in several regions.
2. Step two: moving traffic and restarting servers. In the update The Verge quoted, Microsoft said:
   • it had finished redistributing connection traffic across its worldwide infrastructure
   • it was moving more traffic off the affected infrastructure
   • it was restarting affected servers to bring them back into service
   • it was investigating whether the extra traffic was a real increase in requests or a change in how traffic was routed
   Reuters, carried by CNA, quoted Microsoft as saying "We're aggressively pursuing all pathways to reduce impact," and reported that the company has paused all deployments, is moving additional traffic off the affected infrastructure, and is restarting affected servers to return them to service.
3. Step three: a revised diagnosis. ABC News reported a clarification: "We determined the elevated load isn't driven by an increase in traffic, but by a number of servers being unexpectedly unavailable; straining the remaining infrastructure."

That third point matters. The question The Verge's update left open was whether the load came from a real surge in requests or from a routing change. Microsoft's later answer was neither, at least as the main driver. Some servers went down, and the healthy servers that remained couldn't absorb the extra work. Think of a supermarket where half the checkout lanes close at rush hour: the shoppers are the same, but the open lanes still back up.

Microsoft has not said why those servers became unavailable. The deployment pause suggests engineers were ruling out a recent change. That is my inference, not something Microsoft has stated. A full root cause will normally appear in the post-incident report in the admin center, not on social media.

Section summary: Microsoft first blamed elevated load. It later said the load came from servers that were unexpectedly unavailable, not from more traffic. The reason those servers failed hasn't been made public.

How big was it?​

Downdetector numbers were large, but treat them with care. They count user reports at particular moments, not confirmed affected customers. The snapshots below come from different outlets at different times, so don't add them together:

The Verge said Outlook reports began spiking just before 1 p.m. ET and hadn't fully dropped when it published. Reuters also noted that the actual number of affected users may vary. The Verge also struggled to load Microsoft's own status pages consistently. That's frustrating during an outage, though it's common when everyone checks the same page at once.

Section summary: Downdetector reports peaked above 100,000, but they are user reports at specific times and shouldn't be read as a count of affected customers.

Not just Outlook​

Teams: two separate stories​

Two Teams issues got mixed together in some coverage. Earlier that day, Microsoft reported a separate Teams problem in North America. ABC News reported that Microsoft later posted that "Our mitigative actions have successfully restored service health and functionality." During the Outlook incident, The Verge said some people reported Teams trouble but most said Teams still worked. The available reporting doesn't show that the two problems had the same cause.

SharePoint, Copilot and AI connectors​

Yahoo Tech reported that the outage also affected other Microsoft services, including Microsoft Teams, the Copilot 365 AI service and SharePoint Online. It quoted Microsoft's Service Health page as saying "Users may be unable to connect to multiple Microsoft 365 services". How-To Geek added that some users also faced issues accessing Microsoft's administrative portal during the outage, which left administrators with limited visibility.

The effects also reached AI products. According to Reuters, OpenAI said it was investigating an issue with Microsoft connectors in which new connections may be broken and tools are not working as expected. Anthropic said it had identified an issue with an upstream provider affecting its Microsoft 365 connector. ChatGPT and Claude themselves weren't down. Their features that read your Microsoft 365 mail and files were failing because Exchange was failing. Any agent or automation built on the Microsoft Graph and Exchange will break when Exchange does.

Section summary: The impact went beyond mail to some Microsoft 365 services and to third-party AI connectors. The earlier Teams problem in North America appears to be a separate, already-fixed incident.

What to do now​

No public report I could verify gives a guaranteed recovery time. Microsoft also hasn't published specific user workarounds, and none of the reporting establishes whether delayed mail will be delivered later. Treat this as a service-side incident:

For IT administrators:

1. Use the incident ID. Track EX1550543 in the Microsoft 365 admin center under Health > Service health. Gadget Review gives the same advice, saying admins should check the Microsoft 365 admin center for tenant-specific incident details before troubleshooting.
2. Expect the admin portal itself to be slow or unreliable. Have a backup way to communicate with staff, such as Teams if it works for your tenant, SMS or a phone tree.
3. Don't rebuild Outlook profiles or reset accounts in bulk. If OWA, MAPI, ActiveSync and REST are all failing at once, local fixes only create tickets you'll have to undo later.
4. Check your integrations. Mail flows, Power Automate jobs, AI connectors and line-of-business apps that read Exchange may have failed quietly. Recheck them once service is restored.
5. Save the post-incident report. Microsoft's final write-up is what you'll need to explain the outage to your own leadership or compliance teams.

For individual users: if Outlook is failing everywhere, waiting really is the fix. Changing passwords or reinstalling Office won't bring back a mailbox server that's down.

The bigger picture​

This is a familiar problem with Microsoft 365. Putting mail, files, meetings and AI assistants on one platform simplifies administration. It also means one fault inside Exchange Online can stop all of them at once. The Register has covered similar Exchange Online outages for years. In one from 2020, Microsoft eventually pinned the blame on a "recent configuration update", rolled it back.

Whether this incident has a similar cause remains unknown until Microsoft says so. Two lessons apply either way:

• Concentration risk is real. If your incident-response plan assumes email works, test that assumption.
• AI agents add dependencies. Each Copilot feature or third-party connector built on Exchange becomes another thing that breaks when Exchange does.

Microsoft deserves some credit for posting frequent updates and for correcting its own diagnosis in public. Still, the company was answering the question of where the load came from in real time, while more than 100,000 people reported problems on Downdetector. Admins should read the final root-cause analysis carefully when it arrives.

Key takeaways​

• What: an Exchange Online connectivity incident, EX1550543, affecting Outlook on the web, desktop and mobile apps, plus connections through MAPI, ActiveSync and REST.
• Where: first described as the Americas, later expanded to say users in any region could be affected.
• Why, so far: servers became unexpectedly unavailable and overloaded the remaining infrastructure. Microsoft says it wasn't a traffic surge. The underlying cause hasn't been disclosed.
• Microsoft's response: paused deployments, redistributed traffic and restarted affected servers.
• Also affected: some Microsoft 365 services, the admin portal for some users, and OpenAI and Anthropic's Microsoft 365 connectors.
• What admins should do: track the incident in Service health, avoid local "fixes," and audit your automations once service returns.