A major Microsoft outage hit Outlook, Teams, SharePoint, Copilot and a long list of Microsoft 365 services on Thursday 12 November 2026 (early Friday morning in Australia). Downdetector logged nearly 130,000 Outlook problem reports worldwide at the peak. Australian businesses, schools and government teams woke up to failed logins, stuck inboxes and silent Teams chats.
Microsoft tracked the Exchange Online impact as incident EX1550543, with wider Microsoft 365 impact under MO1550821. As of Microsoft’s public service-health update (last refreshed late Thursday UTC / Friday morning AEDT), the company still listed Microsoft 365 as degraded. A portion of affected servers had been returned to service and remained stable, with limited recovery in some regions. Full restoration was still underway and may feel gradual as clients reconnect.
What went down
This was not a single app glitch. Microsoft’s status page listed connectivity problems across core Microsoft 365 workloads, including:
- Exchange Online (Outlook / business email)
- Microsoft Teams
- SharePoint Online
- Microsoft 365 Copilot
- Planner, To Do and Intune
- Defender XDR and Defender for Office 365
- Multiple Microsoft Purview services (DLP, labelling, compliance)
- Power BI sensitivity-label PDF exports, Universal Print, and parts of the Microsoft 365 Apps admin centre
Many users also reported trouble reaching Word, Excel and PowerPoint when those apps depended on cloud sign-in or cloud files. OpenAI and Anthropic both reported knock-on issues with Microsoft 365 connectors into ChatGPT and Claude.
What Microsoft said caused it
Microsoft first described “elevated load” across infrastructure. It later clarified the load was not from a traffic spike. A large number of servers became unexpectedly unavailable, which pushed remaining capacity harder and created a cascading failure.
We’re testing a configuration change to address a problem that’s causing servers to fail automated health checks and become unavailable. In parallel, we’ve initiated a restart of all affected servers and are monitoring health as they return to service. More info under EX1550543…
Microsoft 365 Status (@MSFT365Status), 12 November 2026
In practical terms:
- Servers started failing automated health checks and were taken out of service.
- A core process on affected servers became overwhelmed and stopped responding.
- Microsoft paused deployments and configuration changes while investigating recent updates.
- Early restarts and reroutes did not deliver the relief Microsoft expected.
- Recovery then focused on alternative health checks, selective isolation, traffic shifting, protective traffic management, and bringing servers back in stages.
RMIT associate professor Mark Gregory told The Guardian the pattern looks like a network-layer configuration problem that shrinks the healthy server pool, overloads what is left, then cascades. That matches what many Australian IT teams saw: intermittent access, then broad failures, then slow regional recovery.
Australia impact
Local time mattered. The Exchange Online incident (EX1550543) began around 3.17am AEDT on Friday 13 November. Microsoft flagged mailbox issues in the Americas around 4.54am AEDT, then confirmed users in any region could be affected. By mid-morning AEDT, Australian Downdetector reports for Outlook were in the thousands. Schools, councils, airlines-adjacent ops and ordinary SMEs all rely on the same Outlook and Teams stack for the start of the day.
If your business runs customer email, booking confirmations, or staff coordination through Microsoft 365 only, an outage like this is not a tech inconvenience. It is lost sales, delayed support tickets, and meetings that never start.
Current status (as of late 12 Nov UTC)
Microsoft’s public Microsoft 365 service health page still listed service degradation after the overnight incident window. The company said:
- A portion of affected servers had returned and stayed stable.
- Traffic and connections had been shifted to alternative pathways where possible.
- Limited recovery had been observed in some regions.
- Protective traffic management was being trialled in a specific location to reduce load.
- Users should expect gradual recovery as capacity returns and clients reconnect.
Always check live sources rather than social media rumours:
- Microsoft service health status
- Microsoft 365 admin centre → Health → Service health (incidents EX1550543 / MO1550821)
- Microsoft 365 Status on X (@MSFT365Status)
What Australian businesses should do right now
- Confirm the outage before changing DNS or passwords. If Outlook and Teams are both failing for many tenants, local config changes usually make recovery harder.
- Use a backup channel for urgent customer contact. Phone, SMS, a secondary mailbox, or a contact form on your website that is not hosted inside Microsoft 365.
- Keep an offline copy of critical templates and contact lists. Anything that only lives in SharePoint or OneDrive is unreachable during a cloud outage.
- Tell staff what “good” looks like when services return. Gradual recovery means some people come online before others. Avoid everyone hammering reconnect at once if Microsoft asks for restraint.
- Review vendor concentration after the dust settles. Productivity suite outages and website/email hosting failures are different risks. Spreading them reduces the chance that one vendor event stops the whole business.
The hosting lesson from this Microsoft outage
Microsoft 365 is excellent when it is up. This incident is a reminder that “the cloud” is still someone else’s computers, health checks and configuration pipeline. Australian businesses that put customer websites, transactional email and collaboration into one vendor’s control plane feel every failure twice.
A practical split many of our customers already use:
- Keep Microsoft 365 for staff collaboration if that is your standard.
- Host the public website and customer-facing webmail or contact forms on Australian hosting you control, with known backup and restore paths.
- Document a short outage playbook: who posts the status note, which alternate inbox gets used, and how support tickets are captured if Outlook is dark.
If your site still needs a Sydney-based home that is not tied to a global SaaS control plane, start with shared cPanel hosting. For teams that outgrew shared and need dedicated resources, see Linux VPS hosting. This article is the news briefing. Those pages are the product detail.
Quick FAQ
Is the Microsoft outage over?
Microsoft reported partial recovery and stable returns for some servers late on 12 November UTC, with gradual restoration expected. Treat any “resolved” claim as provisional until Service health clears EX1550543 / MO1550821 for your tenant.
Was Azure Front Door the cause?
Early social posts mixed Azure edge rumours with the Microsoft 365 incident. Microsoft’s own updates for this event focused on servers failing health checks, an overwhelmed core process, and configuration investigation — not a confirmed public root-cause report yet. Wait for Microsoft’s post-incident summary before locking that narrative in.
Why did Australia feel it so hard?
The failure window lined up with the Australian morning. Outlook and Teams are default tools for large parts of education, government and SME work here, so a global Microsoft outage becomes a national productivity event by breakfast.
We will update this post if Microsoft publishes a formal root-cause write-up. For hosting that stays under your control in Australia, start at HSJ Host.
