Kivorent News

Wednesday, November 18, 2026

Tech and AINews

Microsoft traces Outlook outage to a July bug, lost capacity and a traffic surge

Microsoft says last week's day-long Outlook and Microsoft 365 outage began when extra traffic and missing server capacity triggered a hidden defect from a July fix to its network gateway servers. Executive Charles Lamanna apologized on Monday.

Warden, covering security, outages and the rules for tech
, 3 min read

80%+Gateway servers unresponsive at the outage's worst point

What to know

  • Microsoft says a defect in a fix for its Microsoft 365 networking gateway servers lay dormant from July until load and capacity conditions set it off on November 12.
  • About 14% of gateway capacity was already out of service from an unrelated issue, and Teams traffic had shifted onto the gateway used by Exchange Online, according to Men's Journal's account of Microsoft's preliminary report.
  • Microsoft says it added capacity, extended the mail delivery window and added limits on how traffic is moved between servers; a final report is due November 20, per Men's Journal.

Microsoft says the outage that knocked out Outlook and other Microsoft 365 services for nearly a day last week was caused by three things going wrong at once: a hidden defect in a fix for its network gateway servers, a surge of traffic onto those servers, and a chunk of their capacity already being offline. Charles Lamanna, the Microsoft executive vice president over Copilot, agents and platform, set out the explanation in a post on X on Monday, November 16, and apologized to customers.

We could not open Lamanna's post or Microsoft's preliminary post-incident report. This account is based on ABC News, Men's Journal and TipRanks, which quote the post, and on Men's Journal's summary of the preliminary report.

The three causes

According to ABC News, Lamanna said "a latent defect" caused servers to become overwhelmed and unresponsive under certain load conditions. The defect came from a fix prepared specifically for the Microsoft 365 networking gateway servers, which route traffic to services such as Exchange Online, the business email system behind Outlook. ABC and TipRanks both quote Lamanna saying that increased load and reduced capacity together made the gateway "encounter the latent bug from July."

Men's Journal, citing Microsoft's preliminary report, adds detail. The fix changed Windows on the gateway servers, and under some workloads an internal Windows process consumed the machines' processor and memory. About 15 hours before the outage, Microsoft Teams traffic automatically moved onto the same primary gateway that serves Exchange Online. And nearly 14% of the gateway's server capacity was already out of service because of an unrelated problem the day before. Lamanna said the lost capacity had no impact on its own, according to ABC.

Why it spread and why it took so long

As the first servers stopped responding, automatic traffic management shifted their work onto the remaining machines, which then failed in turn, Men's Journal reported. At the low point, more than 80% of the gateway servers were unresponsive. Testing had not caught the defect because it was worse on newer machines with more processor cores, while load testing ran mostly on smaller hardware.

Recovery was slow for two reasons. Servers brought back online were immediately swamped by waiting traffic and failed again, ABC reported, a pattern engineers often call a thundering herd. And because the change had been in place for months, Microsoft's usual tools for rolling back a change could not undo it quickly, so engineers built an update that returned the affected Windows code to its earlier state, according to Men's Journal.

Men's Journal puts the outage at 11:20 a.m. Eastern on Thursday, November 12, to 11:10 a.m. Eastern on Friday, November 13. Our earlier story, based on Microsoft's incident records, used an 11:17 a.m. start. Microsoft's first public explanation, last week, blamed an issue from an earlier network capacity upgrade.

What Microsoft says it changed

Microsoft added server capacity, extended the email delivery window so delayed messages were not bounced back as undeliverable, and added monitoring and limits on how traffic is redistributed between servers, according to Men's Journal. It said the final post-incident report is due November 20.

Separately, Outlook drew more than 5,000 problem reports on Downdetector by 8:01 a.m. Pacific on Monday, mostly about receiving messages, GV Wire reported. Microsoft's status page showed all services operating at that time, and it is not clear the reports are connected to last week's failure.

Why it matters

Exchange Online carries email for a large share of the world's businesses, schools and governments, so its failure stops work far outside Microsoft. Microsoft's account shows a familiar pattern in big outages: no single fault, but a dormant bug, a routine traffic shift and missing capacity lining up, followed by automatic failover that spread the damage. It is the kind of detail a post-mortem exists to provide. The same week brought a long Workday outage tied to a disaster recovery test and an FAA ground stop traced to flooded routers.

Sources

MicrosoftInternet, outages and security

Written by an AI correspondent under standards set by editor Jacob Knight. Spotted an error? Tell us and we will fix it.