Australia's domestic intelligence agency and police confirmed a global Microsoft 365 outage disrupted their operations, as Outlook, Teams and Copilot stayed unstable into Friday.
Governments, hospitals and airlines across several continents spent the day describing email and online services that had slowed or stopped, while Tony Burke, Australia's Home Affairs Minister, said the disruption was "very much on email systems".
The fault began late on Thursday 12 November and was still disrupting services on Friday 13 November, with Outlook, Teams and Copilot affected for more than half a day. The trouble spilled into Asia-Pacific business hours.
Microsoft said it was not aware of any malicious activity causing the incident.
As of now, we are not aware of any malicious activity that is causing this incident. We will continue to treat this incident with urgency and priority until full resolution
The company's account of what went wrong shifted during the day. Microsoft first blamed connectivity problems on "elevated load" across its infrastructure, then said servers had unexpectedly become unavailable, pushing demand onto the machines still running.
The servers we returned to service can become unresponsive again once they receive traffic; we're analyzing diagnostic data to determine what's triggering the behavior
Microsoft paused all deployments, moved traffic away from affected infrastructure and restarted affected servers, it said. Some regions improved after traffic protocols were adjusted, traffic was redistributed and some traffic was temporarily blocked, but the gains didn't last.
Recovery came in stages, and did not hold
- Thu, before 1 p.m. ETFirst public post
Microsoft said it was investigating reports that users could not access online mailboxes.
- Thu night into FriOutage runs on
Outlook, Teams and Copilot stayed degraded into the second day.
- OvernightPartial fixes fail
Traffic protocol changes, redistribution and temporary blocks brought recovery in some regions that did not hold.
- 6 a.m. ET FriAsia Pacific restored
Microsoft said traffic and key services were being restored to affected users in the region.
- Early FriTeams and Copilot core back
The core functions of both services were online again.
- FriCapacity short
Service health stayed variable and capacity had not fully returned to baseline.
Service health stayed variable, and capacity hadn't fully returned to baseline, Microsoft said.
Embassies, hospitals and visa queues
Several banks' email services and some government systems in Singapore were hit, while Singpass and emergency services continued to operate. The tax authority said its email capabilities were also disrupted.
| Place | What was reported |
|---|---|
| Australia | Domestic intelligence agency and police; email systems |
| Singapore | Some government systems; email at several banks; tax authority email |
| New Zealand | All visa processing paused |
| Ireland | Mater Hospital staff email; embassy in London |
| Japan | Government IT infrastructure |
| Britain | Email at some schools |
| United States | New York City email and online services |
| Airlines | Lufthansa, IndiGo and Akasa Air customer email |
New Zealand's immigration authority stopped work on visa applications altogether.
Immigration New Zealand has paused the processing of all visa applications until further notice because of the global Microsoft outage
Dublin's Mater Hospital said staff could not reach their emails and asked people to phone instead, while saying it was operating as normal.
The Mater Hospital has been affected by the global outages relating to Microsoft
The outage affected government IT infrastructure in Japan, Deputy Chief Cabinet Secretary Masanao Ozaki said. Ireland's embassy in London was among the embassies affected.
New York City's Emergency Management Department warned residents of possible delays reaching city agencies by email or through online services that rely on Microsoft systems. In Britain, some schools asked parents to call because their email systems were down.
Lufthansa said on its website that sending or receiving emails could be significantly delayed, including booking confirmations and customer service requests, and asked customers to check current flight information through its apps, websites and other digital channels. IndiGo and Akasa Air also faced limits on customer service by email.
Microsoft shares were down by just over 2% in pre-market trading.
Downdetector's map showed user reports in the Americas, as well as in areas of Europe, Asia and Africa.
You can't outsource risk
Organisations need plans for disruptions beyond their control, said Matthew Hardman, Asia-Pacific chief technology officer at Hitachi Vantara, who previously worked at Microsoft.
Risk is something you have to own as an organisation.
Redundancy cannot cover every risk, said Liu Yang, executive director of NTU's Cyber Security Research Centre.
Redundancy does not necessarily guarantee resilience, particularly when multiple services share underlying infrastructure or dependencies
Boh Wai Fong, a professor at NTU, called the mismatch between what works for one organisation and what works for the whole ecosystem a "paradox of digital transformation".
Anupam Chattopadhyay, also at NTU, said a detailed root-cause analysis was needed before assessing possible vulnerabilities. The outage highlights people's reliance on cloud systems, said Timothy Wood, who teaches computer science at George Washington University.
Microsoft was still investigating why the servers went offline, it said.
Excluded, and unexplained
Ruled out
What Microsoft has said
- Malicious activity: the company said it was not aware of any causing the incident
Not explained
Still under investigation
- Why returned servers go unresponsive once they take traffic
- Why the traffic changes that brought partial recovery were not sustained
- When full service will be restored
Workhorse Group, a US maker of trucks and buses, moved the release of its business figures from Thursday to Monday.
