Tech · Microsoft 365 outageContinuing story · report 1 of 5
Microsoft’s Outlook recovery falters as restored servers fail under traffic
Microsoft’s email recovery remained unstable early on November 13, with consequences reaching hospitals, public agencies and connected AI tools. Getting servers online was only half the job.
Written 6 min read · 51 sources
What counts as recovery if the servers fail again when people return to use them? That became the central question in Microsoft’s November 12 outage: restoring capacity was not enough if that capacity could not reliably carry the traffic sent back to it.
Early on November 13, the recovery remained incomplete. A Microsoft update timestamped 03:13 GMT and reproduced by RR Tech Pulse said applied changes were not taking as expected, with alternative ways to increase capacity being pursued amid continuing traffic-management imbalances. That reported update was not independently authenticated as an original Microsoft post. No confirmed full-restoration time was available.38
The consequences reached well beyond an inaccessible inbox. Workplace email and calendars failed, a Toronto hospital network activated an infrastructure-failure response, public agencies warned of delayed communications, and AI tools connected to Microsoft began returning errors. Yet this was not a uniform shutdown of every Microsoft application: the company said messaging and online meetings were largely functional even while other features remained impaired.1245
The servers returned, but not reliably
Exchange Online, the service handling workplace email, was at the center of the disruption. Users could not reliably load inboxes, send or receive messages, or retrieve calendar information. Microsoft’s initial warning concerned mailbox access through multiple connection methods in the Americas; its warning subsequently widened to potential Microsoft 365 connectivity problems in any region. Users reported disruption in countries including the United States, Canada, Ireland, France, Japan and Singapore.31476161
Microsoft’s explanation became more specific as the incident progressed. Servers had become unexpectedly unavailable after failing automated health checks, leaving the remaining infrastructure under greater pressure. Crucially, Microsoft said the elevated load was not caused by an increase in traffic. The problem was reduced capacity, not an established surge in customer demand.38
That describes a failure mechanism, not a definitive root cause. CBC reported that Microsoft was still trying to determine the cause after more than eight hours. Why the checks failed, and why returned servers could not reliably handle traffic, remained unresolved. Microsoft said it had no indication of a cyberattack. RMIT University associate professor Mark Gregory suggested a network-layer problem, but that was his assessment, not a confirmed finding.45
We've paused all deployments, we're moving additional traffic off the affected infrastructure, and we're restarting affected servers to return them to service.
The company also tested a load-rebalancing configuration and an alternative health-check method. The latter allowed a subset of affected servers to return, and Microsoft began extending it more broadly. Other reported measures included limiting particular traffic types in some locations, testing a previous server configuration, and preparing automated detection and restarts for servers that stopped responding. These were recovery attempts, not proof of recovery.131138
Around 6 p.m. Eastern on November 12, Microsoft said a portion of the affected servers had returned and appeared stable. Later, it warned they could become unresponsive again once traffic reached them. CBC reported that Microsoft restricted traffic to servers previously believed restored after finding they remained vulnerable. The later warning superseded the earlier reassurance.24
Microsoft subsequently described “positive results” from restarting infrastructure and recovering server availability in isolation, while monitoring stability before expanding the work. But some efforts had failed to provide the anticipated relief, and the reported 03:13 GMT update still described unresolved capacity problems. The meaningful test was no longer whether a server could restart; it was whether service could stay available under returning traffic.238
When email becomes public infrastructure
At Unity Health Toronto, teams could not reliably send and receive email. The network declared a Code Grey, its response to infrastructure loss or failure. It explicitly said patient and resident care was unaffected and its hospitals remained safe and open. A communications emergency response should not be mistaken for evidence of hospital closures or harm to patients.45
Patients with urgent questions about upcoming appointments were directed to call their clinic or department through the hospital’s main telephone line. Non-urgent email matters could wait until service returned. That was a practical workaround for a failure affecting how people reached care, even though care itself continued.45
VIA Rail warned that passengers might not receive trip-related emails. The University of Toronto, Seneca Polytechnic and George Brown Polytechnic reported email disruption, while classes appeared to continue without disruption. Michigan’s Department of Technology, Management and Budget reported effects on state email and applications. None of those accounts establishes that every underlying service stopped operating.78
New York City Emergency Management warned that agency email replies and Microsoft-dependent online services could be delayed or temporarily disrupted. Residents with urgent requests were advised to use another contact method, with 311 for non-emergency city services and 911 for emergencies. The failure of a shared technology supplier had become a problem of public access.2
Complaint counts are not customer counts
The scale was unmistakable; the number of people actually affected was not. The BBC, citing Downdetector, reported a global Outlook peak of nearly 140,000 problem reports around 18:00 GMT—1 p.m. Eastern—on November 12. Those are submitted complaints, not an audited count of people or organizations that lost access.1
Al Jazeera reported about 130,000 Outlook complaints shortly after 1 p.m. Eastern and about 77,000 shortly before 3 p.m. The Guardian reported about 13,000 five hours after the early-afternoon peak. Falling complaints showed fewer incoming reports; they did not establish that every affected customer had recovered.135
A separate Canada–U.S. figure remains unresolved: CTV’s recorded account gave nearly 150,000 Outlook reports at 1:30 p.m. Eastern, while an accessible CTV-attributed account carried by Canada News Media gave nearly 140,000. Neither should be silently substituted for the BBC’s global figure, which has a different geographic description.732
Other services produced separate snapshots. Euronews’ November 12 report cited more than 46,000 Microsoft 365 complaints and around 3,000 Teams complaints; The Independent’s later November 12 report cited more than 38,000 and roughly 4,500 respectively. These were different measurements, not a synchronized total or necessarily contradictory accounts. Adding them together would not yield a count of unique customers.1511
The failures traveled through connected tools
Microsoft separately announced that mitigations had restored service health and functionality after an earlier North American Teams issue, although some users continued reporting difficulties. That announcement did not establish that Outlook or Exchange Online had recovered. Microsoft said services without an Exchange Online dependency should function normally; the dependency mattered more than the product name.14154
That distinction also explains the disruption reaching other companies’ AI tools. OpenAI reported elevated errors affecting its Microsoft 365 connector, warning that new connections might be broken and tools might not work as expected. Its available official status excerpt did not name individual tools or confirm restoration. This was not evidence that all of ChatGPT was unavailable.19
Anthropic’s official status update at 22:23 UTC on November 12 said mail, calendar and search tools in its Microsoft 365 connector were still failing for most requests, and Claude for Outlook remained affected. It attributed the problem to a Microsoft 365 service incident. No confirmed connector-restoration time was established.20
One business consequence now has a new date: electric truck maker Workhorse Group postponed its third-quarter earnings conference call, citing disruption from the Microsoft outage. Originally scheduled for November 12, the call was moved to November 16 at 4:30 p.m. Eastern. That is a rescheduled business event, not a deadline for Microsoft’s recovery.29
The incident’s unresolved questions are substantial: full restoration, durability under normal traffic, a definitive technical explanation, and the actual population affected. Restarting servers and reducing complaints are progress indicators. Neither, on its own, answers the question that matters to people relying on the service: will it work when they need it?
Timeline
Day 4 · 5 reports
We aired · Update 1You are here
Microsoft’s Outlook recovery falters as restored servers fail under trafficThe Irish Times reported that servers restored repeatedly overnight had crashed again. The earlier stability account had been overtaken by renewed failures.
Microsoft reported better service stability and capacity after traffic optimization, while broadening the measures across affected regions in pursuit of persistent recovery—not declaring it achieved.
We aired · Update 2Newer
Microsoft 365 outage disrupts government services as recovery strugglesWe aired · Update 3Newer
Microsoft outage delays medical scans and a final job interview as recovery remains partialWe aired · Update 4Newer
Microsoft protects delayed mail for 48 hours as outage recovery remains incompleteMicrosoft's stated end of active customer impact.
Microsoft says service is restored to the remaining affected customers and regions and keeps monitoring delayed mail.
Preliminary report published; Lamanna apologises on X.
Sources49
Continuing story · report 1 of 5
Every source, by outlet
51 sources from 49 outlets· Most cited first
Google News Canada (CA:en)ctvnews.ca2
Google News Singapore (SG:en)channelnewsasia.com2
ABC Newsabcnews.com1
- First reported
- Microsoft Outlook outage stretches on for hours
Al Jazeeraaljazeera.com1
- First reported
- Microsoft outage hits tens of thousands of users
BBCbbc.com1
Canada News Mediacanadanewsmedia.ca1
- First reported
- Outlook outage hits Microsoft 365 users in Canada, U.S.
Car Dealership Guy Newsdealershipguy.com1
- First reported
- Microsoft notes ‘positive results’ amid worldwide outage
CBC News1 video1
- First reported
- Microsoft says services slowly being restored after widespread outages | Hanomansing Live · video
CBS Minnesotacbsnews.com1
Claudeclaude.com1
cloud.microsoft1
CNBCcnbc.com1
As aired
- Microsoft's email servers failed again when traffic returned.
- The disruption reaches workplace email, hospital communications and tools connected to Microsoft's systems. Restarts have brought some capacity back, but recovery hasn't held reliably. Can Microsoft restore service and keep it working under normal traffic?
- Users in the United States, Canada, Ireland, France, Japan and Singapore reported problems. Microsoft initially warned about mailbox access across the Americas, then said Microsoft 365 connections could be affected in any region.
- The outage began on November 12 and continued into November 13. People couldn't reliably open inboxes, send or receive email, or retrieve calendar information. Microsoft's Exchange Online service, which handles workplace email, was at the center of the disruption.
- Outlook problem reports peaked globally at nearly 140,000 around one p.m. Eastern. Those are complaints submitted about service problems. They can't tell us how many people or organizations actually lost access.
- A separate Canada-U.S. Outlook count ranges from nearly 140,000 to nearly 150,000, with that discrepancy still unresolved. Its geographic scope also differs from the global peak.
- Other snapshots put Outlook complaints at about 130,000 around one p.m., then about 77,000 before three. Five hours after the early-afternoon peak, the count was about 13,000. Fewer incoming complaints didn't mean everyone had recovered.
- Late that evening, the Canada-U.S. count still exceeded 35,000 across Microsoft services, including nearly 6,700 in Canada. That broader measure includes multiple services, so it can't be directly compared with an Outlook-only peak.
- Microsoft 365 complaints moved from more than 46,000 to more than 38,000 across two snapshots. Teams reports moved from around 3,000 to roughly 4,500. These separate service counts can't be added into a total of unique customers.
- The failures also reached OneDrive and Copilot, Microsoft's AI assistant. Users could have trouble viewing particular files or calendar details. But Microsoft said messaging and online meetings were largely functional.
- Microsoft said servers failed automated health checks and became unexpectedly unavailable. The remaining systems then carried more pressure. The company traced the elevated load to lost capacity. Servers returned to service could fail again when traffic reached them.
- More than eight hours into the investigation, Microsoft still hadn't explained the underlying cause. Why the checks failed, and why the servers couldn't reliably handle returning traffic, remained unresolved. The company said it had no indication of a cyberattack.
- RMIT University associate professor Mark Gregory suggested a problem in the network layer, the systems that carry data between services. That was his assessment. Microsoft hadn't confirmed it as the cause.
- Microsoft paused deployments and shifted traffic away from the affected infrastructure. It also redistributed traffic across its worldwide systems, restarted servers and tested a configuration change to rebalance the load.
- An alternative health-check method allowed a subset of servers to return. Microsoft began extending that change and preparing automated detection and restarts for servers that stopped responding. Those measures still left it managing an unstable recovery.
- Microsoft also limited certain traffic types in some locations and rebuilt some servers using a previous configuration. It isolated infrastructure again to restart components suspected of contributing to the problem.
- Around six p.m. Eastern, Microsoft said a portion of returned servers appeared stable. Later, it warned they could become unresponsive under traffic. It restricted traffic to servers previously thought restored after finding they were still vulnerable.
- Some attempts to open more capacity failed to deliver the expected relief. Microsoft later described positive results from restarting isolated infrastructure, but said it was monitoring stability before expanding that recovery.
- Subscribe to this channel to follow Microsoft's recovery and whether restored services stay reliable.
- Unity Health Toronto activated a Code Grey, its response to infrastructure loss or failure. Teams couldn't reliably send and receive email. The network said patient and resident care was unaffected, and its hospitals remained safe and open.
- Patients with urgent questions about upcoming appointments were directed to call their clinic or department through the hospital's main telephone line. Non-urgent email matters could wait until service returned.
- VIA Rail warned passengers they might not receive trip-related emails. The University of Toronto, Seneca Polytechnic and George Brown Polytechnic also reported email disruption. Classes appeared to continue without disruption.
- Michigan's Department of Technology, Management and Budget reported effects on state email and applications. Individual workers also missed meetings. A Vancouver animation worker said calendar failures caused her to miss meetings on a Disney film project with a major deadline the next day.
- New York City Emergency Management warned that agency email replies and Microsoft-dependent online services could be delayed or temporarily disrupted. It advised residents with urgent requests to use another contact method, with 311 for non-emergency city services and 911 for emergencies.
- Microsoft separately said it had restored service health and functionality after an earlier North American Teams issue. Some users still reported connection problems. Outlook and Exchange Online recovery continued, while services without an Exchange Online dependency were expected to function normally.
- OpenAI said new Microsoft connections could be broken and some connected tools weren't working as expected. Anthropic reported errors and slow responses across its Microsoft 365 connector, including possible failures involving mail, SharePoint, Teams chat and calendars.
- Anthropic initially flagged the problem at 17:52 UTC on November 12 and identified an upstream provider issue later that evening. New connections could fail to complete.
- At 22:23 UTC, Anthropic said mail, calendar and search tools were still failing for most requests. Claude for Outlook remained affected. Its upstream provider was restarting servers as part of the recovery effort.
- Microsoft's recovery was still incomplete early on November 13. In its 03:13 GMT update, the company said applied changes weren't taking as expected. It was pursuing alternative ways to increase capacity while traffic-management imbalances continued. Microsoft hadn't given a full-restoration time.
- Electric truck maker Workhorse Group postponed its third-quarter results event, citing business disruption from the Microsoft outage. Its earnings conference call, originally scheduled for November 12, was moved to November 16 at four-thirty p.m. Eastern.





