Email on Microsoft 365 went down from the early hours of Friday 13 November 2026, and many Queensland offices opened to an inbox that would not load. Microsoft says the cause was technical, not a cyber attack.
Some follow-up is time-sensitive: the incident record in the admin centre goes back only 30 days. The rest is the plan to have before the next outage.
What happened
From the early hours of Friday 13 November 2026, Exchange Online, which carries Outlook mail and calendars, was disrupted for many customers worldwide, including Australian federal agencies, courts and councils. Microsoft's notice (MO1550821) also lists Teams, SharePoint Online, Microsoft 365 Copilot and Microsoft Defender, and gives a start of 1:26 pm US Eastern on Thursday 12 November, which is 4:26 am AEST (Queensland time) on Friday; ABC said user reports peaked at about 4:20 am AEST. Microsoft said broad recovery was under way by the early hours of Saturday 14 November, about 24 hours after it began. It called the cause a technical issue not related to cyber activity or AI, introduced during an earlier network capacity upgrade and surfacing after a recent network change, and promised a full post-incident report.
Outage or local fault? A two-minute test
Check Microsoft's side before anyone touches the office network.
- 1Open status.cloud.microsoft. It needs no sign-in, and Microsoft points admins to it when they cannot sign in to the admin centre.
- 2In the Microsoft 365 admin centre, open Health, then Service health. The Service Support Administrator and Helpdesk Administrator roles can see it. Exchange incidents start with EX.
- 3On a phone, open the Microsoft 365 Admin app, which shows Service health and sends push notifications.
- 4Try the same sign-in on a phone using mobile data. If it works there but not in the office, suspect the office link. If it fails on both and Service health lists an incident, the problem is on Microsoft's side.
If only one person is affected, suspect their device or account. Treat an unexpected sign-in prompt or approval request as a possible compromise.
Backup is not continuity
A backup is a copy of data that can be restored after something is lost or deleted. This outage was an access problem: staff could not read, send or receive mail on time. A backup does not give anyone a working inbox while the service is down.
Keeping mail flowing is a separate job, called email continuity. A continuity service is a second system that stays reachable when Exchange Online does not. In the usual design it holds incoming mail while Microsoft 365 cannot accept it, lets staff read and send from a page that does not depend on Microsoft, and hands the stored mail back afterwards. Products differ, so ask how failover starts, how staff sign in and how long mail is held.
Microsoft extended its email timeout window to 48 hours to protect delayed messages, and delayed mail was reported arriving as service recovered. That helps with late mail, not with a team that must answer a customer at 10 am.
Backup still has a job: restoring deleted files and recovering from ransomware. That is a separate question from an outage, and the check below covers it.
An out-of-band plan
Out-of-band means it works when Microsoft 365 does not. Set these up now, somewhere a dead inbox cannot hide them.
- A contact list on paper and in every manager's phone: staff mobiles, key customers, suppliers, the IT provider and the internet provider.
- An SMS group that reaches every staff member, with one named person who sends the first message.
- A non-Microsoft chat group agreed in advance, with customer personal information kept out of it.
- Service health alerts sent to an address outside Microsoft 365. Under Customize, then Email, each admin can enter up to two addresses.
- Key documents available offline: the continuity plan, price lists, rosters and safety paperwork, as printed copies and as synced files set to Always keep on this device.
- A check on where other systems send sign-in codes and password resets. If it is a Microsoft 365 mailbox, an email outage locks those systems too.
What kept working, and what did not
| What | Result | Why |
|---|---|---|
| Classic Outlook in Cached Exchange Mode | Synced mail stays readable | The PC holds a copy of the mailbox, 12 months by default, so it works when the connection is unavailable. |
| Files kept on the device | Kept working | OneDrive and synced SharePoint files set to Always keep on this device are stored locally. Online-only files need a connection. |
| Mobile calls and SMS | Kept working | They do not run through Microsoft 365. |
| Teams chat and online meetings | Largely kept working | Microsoft's own description. Calendar and scheduling depend on Exchange Online and were affected. |
| Outlook on the web and phone mail apps | Did not work reliably | Every route into the mailbox goes through Exchange Online. |
| Admin centre pages | Some did not load | Use the public status page when the admin centre will not open. |
Cached Exchange Mode is not available in new Outlook for Windows, so check which version staff actually run.
Break-glass admin accounts
Sign-in was not among the services Microsoft listed as affected. The next outage might hit it. If a federation or multifactor service fails, every administrator can be locked out when access matters most. Microsoft's guidance is at least two emergency access accounts:
- Cloud-only accounts on the onmicrosoft.com domain, not federated or synchronised from an on-premises directory.
- The Global Administrator role, permanently active rather than eligible.
- A passkey (FIDO2) or certificate for sign-in, different from the method normal administrators use.
- Excluded from Conditional Access policies that block or restrict sign-in.
- Credentials stored securely in separate locations, with an alert on every sign-in and a test at least every 90 days.
After the outage: record, save, claim
Record the business impact
Write it down now: when each team lost access, what stopped, the workarounds used, hours lost and customer commitments missed. Microsoft's claim guidance asks for the downtime, the affected users and what was done to resolve it.
Save the post-incident report
For incidents with broad impact, Microsoft says a preliminary report arrives in Service health within 48 hours of resolution and a final report within five business days, with start and end times, a timeline and the root cause. Service health keeps history for 30 days, so save the incident and the report.
Claim the service credit
If the Service Level Agreement applies, this is the process Microsoft publishes for its Cloud Solution Provider partners:
- Credits depend on which services were affected and are prorated by the length of the outage. Advisories are not typically eligible, so check whether the outage is listed in Service health as an incident.
- The request needs the tenant ID, the incident ID from Service health, and evidence that the business was affected and asked for the credit. The evidence email must come from the tenant's own domain.
- For services other than Azure, the request must arrive within one month of the end of the billing month in which the incident occurred.
- Credit requests are accepted from providers, so a business that buys through one should ask it to lodge the claim.
If the site runs on one link
A depot or site office on one fixed link, mobile broadband or satellite has a second way to lose email, and from the desk both look identical: the page will not load. The two-minute test separates them, and the plan has to cover both. A second path only counts if it fails differently from the first and has been tested.
In cyclone season a site can lose power and connectivity at once, so the plan has to work with neither.
Before the next outage: a one-page checklist
- 1Name who checks Service health, and give them a role that can see it.
- 2Send Service health email alerts outside Microsoft 365, and install the Admin app on two phones.
- 3Print the contact list and store it on every manager's phone.
- 4Set up an SMS group and a non-Microsoft chat group, and agree who triggers them.
- 5Mark the folders needed in a bad week as Always keep on this device.
- 6Confirm classic Outlook runs in Cached Exchange Mode for staff who live in email.
- 7Ask the IT provider whether email continuity is in place and what happens to mail on day one.
- 8Create two emergency access accounts, store the credentials separately and test them every 90 days.
- 9Write a one-page manual process for orders, bookings, rosters and payroll, keep a paper copy and rehearse it yearly.
- 10If one link or a satellite serves the site, add an independent second path and test the failover.