The facts

  1. The disruption affected Outlook, Teams, Copilot and other Microsoft 365 services. AP reported that Outlook was working again for many users on Friday morning in North America, while some users continued to report intermittent problems or delayed mail. [3][5]

  2. In Britain, police forces said online crime reports were unavailable and directed urgent reports to telephone emergency responders. The National Pharmacy Association warned of possible prescription delays linked to NHS email problems; hospitals asked patients to attend appointments as normal, AP reported. [5]

  3. Microsoft said an earlier network-capacity upgrade issue surfaced after a later network change and said the incident was not related to cyber activity or AI/agents. A PennO365 extract of Microsoft's status notice says a post-incident report was published on Nov. 16; the extract does not include the report's technical details. [4][7][12]

  4. AP reported that Downdetector received outage reports from 74 countries and territories within about an hour of the disruption beginning. Those reports establish reported locations, not the number of affected customers or the share of service unavailable. [5]

  5. ABC News Australia reported at 01:04 UTC Nov. 14 that Microsoft said email queues were significantly reduced and approaching normal levels, while some users might still experience intermittent access issues or delayed delivery. [6]

  6. ABC reported NHS England activated paper or other IT contingencies. A union leader said staff could not consult patient details through usual systems, while emergency life-threatening callouts appeared unaffected. [6]

  7. PennO365's 04:50 UTC Nov. 14 extract listed both Microsoft 365 incident MO1550821 and Exchange Online incident EX1550543 as serviceRestored, with EX1550543 ending Nov. 13 at 16:10 UTC. Its displayed status text still described restoration for remaining customers and regions; that relay predates Microsoft's later all-clear. [7]

  8. Reuters reported at 08:23 UTC Nov. 14 that Microsoft said it had fully recovered service across Teams, Outlook and other products, restoring the remaining impacted customers and regions. Reuters attributed the update to Microsoft; it does not independently test every customer's service. [8]

  9. Anadolu quoted Microsoft's service-status update saying services across previously affected environments had fully recovered and the email backlog had been processed, and that the issue was remediated. This is the company's recovery claim, not independent customer-by-customer telemetry. [9]

  10. AP reported that the outage disrupted government, court, health, business, airline and banking operations in multiple places. It cited more than 430 million Microsoft 365 users as the platform’s user base, not as people affected; total outage reach remains unknown. [10]

  11. A PennO365 extract generated at 11:05 UTC on Nov. 15 showed the original Microsoft 365 and Exchange incidents restored and a separate Teams Pro Management Portal device-monitoring advisory degraded at about 95% recovery. A later Nov. 16 extract records the advisory’s restoration. [11]

  12. A PennO365 extract generated at 20:35 UTC on Nov. 16 shows the original Microsoft 365 and Exchange incidents as post-incident-review-published, with a report published at 04:47 UTC. It separately marks Teams advisory TM1551945 serviceRestored at 18:48:12 UTC. Microsoft’s final update says fixes rebalanced the service and redirected request traffic; the advisory’s root cause was a traffic spike after the earlier outage. This extract does not include the full post-incident report. [12]

  13. Ars Technica’s account of Microsoft’s preliminary post-incident report says the outage involved a latent gateway software defect, a Teams traffic shift after failover about 15 hours before the outage, and an unrelated fault that left gateway capacity about 14% below optimal. Microsoft’s report said more than 80% of affected gateway servers froze. These are preliminary company findings; Ars reported the final report was expected Nov. 20. [13]

Still unconfirmed

  • The total number and share of customers affected by the broad Microsoft 365 outage.
  • Whether any isolated customer or feature reports continued after Microsoft's broad-service all-clear; independent customer-by-customer recovery data were not available.
  • The contents of the post-incident report identified in the Nov. 16 service-health extract, including its full technical analysis and mitigation details.
  • Whether the full post-incident report contains further technical findings or mitigation details not shown in the PennO365 service-health extract.

Corrections & updates

Update, 13 November: the PennO365 mirror of Microsoft service health showed EX1550543 and MO1550821 remained in service degradation in a 01:48 UTC update. Microsoft reported positive results from server restarts but continued monitoring before expanding remediation. The mirror is identified as a relay, not independent telemetry; Downdetector reports are not a customer count.

Substantive update, 13 November: added AP's second-day account of effects on emergency calls and court operations, plus Microsoft's 13:00 UTC status reporting uneven recovery across North America, Europe and Asia-Pacific. Some access problems and delayed mail remain; cause and customer total are unknown. Original URL and publication date are unchanged.

Update, 13 November: added Microsoft’s 20:01–20:02 UTC notices that email backlogs remain and no full-recovery time is available, plus its initial explanation involving a network upgrade and later change. Clarified the reported British public-service effects: online police reports and potential prescription delays, rather than implying emergency telephone lines were unavailable. Expanded the location from Americas to Worldwide. The original URL, publication date and earlier updates are retained.

Substantive update, 13 November: added the latest 22:00 UTC PennO365 mirror of Microsoft’s service-health notices. Both incidents still showed service degradation while Microsoft reported sustained recovery and said it was monitoring an accumulated mail backlog; no resolution time was given. The AP-reported partial recovery and British service impacts remain attributed, and the article does not imply the outage is fully active or resolved. Original URL and publication date remain unchanged.

Substantive update, 14 November UTC: checked the latest available Microsoft service-health relay, which still listed service degradation and backlog monitoring, against AP and ABC reports of broad but incomplete recovery. Some access problems or delayed mail may persist; the incident is not described as fully active or resolved. Original URL and publication date remain unchanged.

Substantive update, 14 November UTC: replaced the prior 22:00 UTC provider update with Microsoft’s 02:17 UTC notice, relayed in PennO365’s 02:50 UTC extract. Microsoft reported continued recovery while monitoring open impact scenarios and an accumulated email backlog; status remained service degradation and its next update was due 04:30 UTC. No resolution was reported. Original URL and publication date remain unchanged.

Substantive update, 14 November UTC: refreshed status using PennO365’s 04:05 UTC extract. It lists EX1550543 as service degradation and not ended; record metadata last updated 03:58 UTC, while newest displayed status text is timestamped 02:13 UTC and reports continued recovery and backlog/open-scenario monitoring. No later status body was available in the extract at 04:33 UTC. Recovery language remains; the article does not imply fully active or resolved. Original URL and publication date are unchanged.

Substantive update, 14 November UTC: PennO365’s 04:50 UTC extract now marks both Microsoft service records serviceRestored, with EX1550543 ending Nov. 13 at 16:10 UTC. Microsoft’s latest notes say the underlying issue was addressed but restoration continued for remaining customers and regions. Replaced the stale 04:05 UTC degradation status; retained attributed public-service effects and the provider’s qualified network explanation. The article does not imply that every user or mail queue recovered. Original URL and publication date are unchanged.

Substantive update, 14 November: added Reuters' 08:23 UTC report and Anadolu's later account quoting Microsoft's service-status update that previously affected environments had fully recovered and the email backlog had been processed. Replaced the earlier qualified residual-restoration status with Microsoft's reported all-clear. Independent customer-by-customer recovery and the formal technical incident report remain unavailable; the original URL, publication date and prior history are retained.

Substantive update, 14 November UTC: added AP’s report of disruption across government, court, health, business, airline and banking services in multiple countries. Its figure of more than 430 million Microsoft 365 users is the platform-wide user base, not an affected-user count. Microsoft’s full-restoration statement remains attributed; independent customer-level recovery and the full technical post-incident report remain unknown.

Substantive update, 15 November UTC: added the current Microsoft service-health extract. The original Exchange Online and Microsoft 365 outage incidents remain marked restored; a separate Teams Pro Management Portal degradation linked to post-outage request traffic remains open, with Microsoft reporting 95% recovery of affected Windows and Android device-monitoring status at 10:59 UTC. This narrower administrative issue does not establish a return of the broad outage.

Nov. 16: Updated to reflect Microsoft’s status record that a post-incident report had been published. The extract shows the broad outage restored while a separate Teams Pro device-monitoring issue remained degraded at about 95% recovery.

Status update, 16 November UTC: the PennO365 extract generated at 20:35 UTC relays Microsoft post-incident-review publication for the original Microsoft 365/Exchange incidents and marks the separate Teams Pro Management Portal advisory restored at 18:48:12 UTC. The broad outage remains restored; the full report contents and independent customer-level recovery data are not inferred.

Substantive update: added Ars Technica’s account of Microsoft’s preliminary post-incident report, including the latent gateway defect, traffic shift after Teams failover and pre-existing 14% capacity shortfall. Microsoft’s explanation remains preliminary; Ars reported a final report was expected Nov. 20.

Update, Nov. 17 UTC: an earlier service-health extract showed the separate Teams Pro Management Portal advisory at 95% recovery. PennO365’s later Nov. 16 extract records it restored at 18:48 UTC. The current summary reflects that later provider-status update; the broad Microsoft 365 outage and email backlog were already reported restored.

Sources

References support the numbered facts above. Source checks: 17 Nov 2026, 18:34 UTC.

PennO365’s Nov. 16 20:35 UTC extract relaying Microsoft Admin Center records shows MO1550821 and EX1550543 as post-incident-review-published. It separately marks Teams Pro Management Portal advisory TM1551945 serviceRestored at 18:48:12 UTC Nov. 16. Microsoft attributed that fix to rebalancing the service and redirecting traffic after a request spike. This provider relay is not independent customer telemetry. Ars’s account of the preliminary post-incident findings is attributed to Microsoft; Ars reported a final report was expected Nov. 20.

  1. Is Outlook down? Widespread outage hits Microsoft email and other apps

    The Associated Press · Reporting · 12 Nov 2026

    AP reports the outage, Microsoft's account of the Exchange Online disruption and fix work, and user reports on Downdetector. The user-report total is attributed and is not treated as a count of all affected customers.

  2. Microsoft service-health extracts for Exchange Online EX1550543 and Microsoft 365 MO1550821

    University of Pennsylvania PennO365 · Reporting · 14 Nov 2026

    PennO365 relays Microsoft Admin Center notices and refreshes about every 15 minutes. Its 04:05 UTC Nov. 14 extract lists EX1550543 serviceDegradation/not ended; record metadata last updated 03:58 UTC. Newest displayed Microsoft status body: 02:13 UTC, continued recovery, most scenarios addressed, remaining impact and email backlog monitored, next update 04:30 UTC. No later body was accessible in this extract at 04:33 UTC. Provider status relay, not independent telemetry.

  3. Microsoft reports progress as it works to recover email and other services in a global outage

    The Associated Press · Reporting · 13 Nov 2026

    Reports second-day impacts on businesses, police emergency calls, courts, health and government services, alongside Microsoft's regional recovery updates. Affected-user total remains unknown.

  4. Microsoft outage live — Outlook and Teams returning to 'normal levels' as tech giant says there's 'no indication' of cyberattack

    TechRadar · Reporting · 13 Nov 2026

    Read the 16:57 UTC entry reproducing Microsoft’s network-upgrade explanation. Its confident recovery and cyberattack language is treated as the provider’s assessment, not independent proof.

  5. The Latest: Microsoft says recovery efforts are making progress on worldwide outage

    The Associated Press via WSB-TV · Reporting · 13 Nov 2026

    Read the AP dispatch carried by WSB-TV, updated 15:46 UTC. Specific public-service effects concern online crime reporting and possible prescription delays; telephone emergency reporting is recommended. This is the same AP reporting organization as sources 1 and 3.

  6. Microsoft outage hits users worldwide, affecting Teams, Outlook and Copilot - as it happened

    ABC News (Australia) · Reporting · 13 Nov 2026

    ABC reported at 01:04 UTC Nov. 14 that Microsoft said email queues were significantly reduced and approaching normal levels while intermittent access or delayed delivery could remain. It also reported NHS England activated contingencies and attributed staff-impact details to a union leader; emergency life-threatening callouts appeared unaffected.

  7. Microsoft service-health extracts for Exchange Online EX1550543 and Microsoft 365 MO1550821

    University of Pennsylvania PennO365, relaying Microsoft Admin Center service-health notices · Reporting · 14 Nov 2026

    PennO365’s 04:50 UTC Nov. 14 extract lists both records as serviceRestored; EX1550543 gives an end time of Nov. 13 16:10 UTC. The displayed Microsoft status text was updated 04:15:46 UTC for EX1550543 and 04:17:52 UTC for MO1550821. Microsoft said the underlying issue was addressed but it was continuing to restore service to remaining customers and regions; it attributed the issue to an earlier capacity upgrade surfacing after a later network change and planned preliminary and final incident reports. This is a provider-status relay, not independent user telemetry or proof all accounts and mail queues recovered.

  8. Microsoft restores service to Outlook and Teams after global outage

    Reuters (republished by Euronext) · Reporting · 14 Nov 2026

    Opened Reuters' 08:23 UTC Nov. 14 dispatch carried by Euronext. It reports Microsoft's statement that services had fully recovered across Teams, Outlook and other products and that remaining impacted customers and regions were restored. The recovery account is attributed to Microsoft.

  9. Microsoft says consumer products, services fully recovered as of Friday

    Anadolu Agency · Reporting · 14 Nov 2026

    Opened Anadolu's Nov. 14 report. It says consumer products recovered by 16:10 UTC Friday while Microsoft 365 enterprise services showed sustained recovery, and also quotes the company's status-page statement that previously affected environments recovered and the email backlog was processed. These are company statements, not independent service telemetry.

  10. Microsoft says services fully restored after hourslong global outage

    The Associated Press · Reporting · 14 Nov 2026

    Opened AP’s Nov. 14 report. Microsoft said it restored remaining impacted customers and regions; AP says Downdetector received reports from more than 70 countries and describes disruption to government, court, healthcare, business, airline and banking services. It cites 430 million Microsoft 365 users as the platform-wide user base, not people affected by this outage.

  11. Microsoft service-health extracts for Microsoft 365 MO1550821, Exchange Online EX1550543 and Teams PMP TM1551945

    University of Pennsylvania PennO365 (Microsoft service-health extract) · Primary source · 15 Nov 2026

    Opened the extract generated at 11:05 UTC Nov. 15. It marks MO1550821 and EX1550543 serviceRestored, ended Nov. 13 at 16:10 UTC; a distinct Teams Pro Management Portal advisory remains serviceDegradation after an outage-related request spike. Microsoft’s 10:59 UTC update says 95% of affected Windows and Android device-monitoring status has recovered; the next update is due Nov. 16 at 20:30 UTC.

  12. Microsoft service-health extract for incidents MO1550821, EX1550543 and TM1551945

    University of Pennsylvania (PennO365, relaying Microsoft Admin Center service-health notices) · Primary source · 16 Nov 2026

    Opened PennO365’s Microsoft service-health extract at 20:35 UTC on 16 November. It shows MO1550821 and EX1550543 as post-incident-review-published and says a report was published at 04:47 UTC. It marks separate Teams Pro Management Portal advisory TM1551945 serviceRestored at 18:48:12 UTC; Microsoft’s final update attributes the advisory to a post-outage traffic spike and says its fix rebalanced service and redirected traffic. This is a provider-status relay, not independent customer telemetry; the full post-incident report text is not included in the extract.

  13. Three small failures combined to take down Microsoft 365

    Ars Technica · Reporting · 16 Nov 2026

    Ars reports details from Microsoft’s preliminary post-incident report and reproduces its explanation of the latent gateway bug, Teams failover traffic shift, 14% capacity shortfall and frozen servers. The full report requires Microsoft 365 Admin Center access; the causes remain Microsoft’s preliminary account, not independent technical findings. Ars says a final report was expected Nov. 20.

Report a factual error