A Microsoft 365 outage that began on Thursday evening, West Africa time, was still being fixed on Friday. Users worldwide, including in South Africa, struggled to open Outlook mailboxes, and features in Teams and other Microsoft 365 apps that rely on Exchange Online were disrupted too.
Key facts
- Problems with Exchange Online, the service behind Outlook email, started at 16:17 UTC (17:17 WAT) on Thursday 12 November, according to LazyAdmin’s timeline of Microsoft’s incident EX1550543. Microsoft opened a wider Microsoft 365 incident, MO1550821, at 18:26 UTC (19:26 WAT) (Microsoft incident log).
- Microsoft says “a number of servers being unexpectedly unavailable” overloaded the rest of its infrastructure (Microsoft 365 Status on X).
- Outlook problem reports on Downdetector peaked worldwide at nearly 140,000, around 18:00 GMT on Thursday, the BBC reported via KBC.
Microsoft lists Exchange Online, Teams, SharePoint Online, Microsoft Copilot 365, Defender XDR, Planner, To Do and Intune among the services that may be affected, ABC7 reported from Microsoft’s status page. Only features that rely on Exchange Online and email were hit. Microsoft said: “Elements of those services that do not depend on interaction with Exchange Online remain functional” (Microsoft 365 service health). In Teams, chats and meetings mostly kept working while calendar and meeting scheduling were affected, LazyAdmin reported.
Why it matters: Friday is a working day across Africa, and the outage ran into it. In South Africa, TechCentral and Sunday World reported local users were hit as the business day began. We found no reports yet that put a number on the impact in Nigeria, Kenya or elsewhere on the continent.
What went wrong
Microsoft initially reported “elevated load” across several regions. It later said the load was not caused by more traffic, but by servers that had dropped out and failed automated health checks (Microsoft 365 Status on X).
Fixes did not hold at first. Microsoft said servers it returned to service “can become unresponsive again once they receive traffic” (Microsoft 365 Status on X). It also said it found “a core process on affected servers is becoming overwhelmed” (Microsoft 365 Status on X).
On Friday morning, Microsoft said it was “seeing positive results for server stability” after a patch. By about 09:50 UTC (10:50 WAT) it reported “early stabilization and capacity improvements”, but warned that “service capacity has not yet returned to normal levels”, according to TechRadar’s live blog. It said it was not aware of “any malicious activity” behind the incident.
Later on Friday, Microsoft said it was “seeing encouraging recovery progress, with service availability and email delivery performance improving”, but warned that “some users may still experience intermittent issues as recovery actions continue”, Windows Central reported at 12:33 UTC (13:33 WAT).
At 17:57 UTC (18:57 WAT) on Friday, Microsoft said it was “seeing broad recovery and stability across our services”, but it had not yet declared the incident resolved (Microsoft incident log).
For email, Microsoft said “the extended 48-hour email timeout window remains in place to help protect delayed messages as delivery catches up” (Microsoft incident log). That is a setting on its Exchange Online mail service, not a guarantee that every delayed message arrives.
Who else was hit
Australian government agencies, Japan’s government information system and some users at the US House of Representatives were among those affected, Sunday World reported. OpenAI and Anthropic reported problems with their Microsoft 365 integrations, not with ChatGPT or Claude as a whole (OpenAI status).
What to do if your organisation is affected
- Check Microsoft’s public status page at status.cloud.microsoft and, for IT admins, incidents EX1550543 and MO1550821 in the Microsoft 365 admin centre.
- Expect a slow return: Microsoft says users “may experience a gradual recovery as their clients reconnect” (Blueprint).
- Some paying Microsoft 365 customers may qualify for a service credit if the affected service’s November availability falls below its monthly service-level threshold. Credits are not automatic: document your organisation’s impact, and check your contract or ask your Microsoft provider how to file. Microsoft’s window for non-Azure claims from a November incident ends on 31 December 2026 (Microsoft, LazyAdmin).
What’s next
Microsoft has not given a time for full recovery. It has promised a full root-cause report once the problem is mitigated, Forbes reported.
Last updated: Friday 13 November 2026, 23:00 WAT. This story is developing.
