Microsoft 365 Outage: Windows Lock Contention Left Over 80% of Gateway Servers Unresponsive, Exchange Online Email Hit for 23 Hours and 50 Minutes, and What to Plan For
Microsoft's preliminary report blames lock contention in an internal Windows process for leaving over 80% of its gateway servers unresponsive. Exchange Online email was disrupted for 23 hours and 50 minutes. What is confirmed, what is not, and what to plan for.
Where Worldwide
On this page 7 sections
A problem inside Microsoft's own servers caused 23 hours and 50 minutes of customer impact for Microsoft 365, mostly on email, and Microsoft says it was not a cyberattack. If your business runs on Exchange Online, this is a reminder that your email can stop working for reasons you cannot see, patch or fix yourself, so you need a plan for that day before it comes.
What happened in the Microsoft 365 outage
BleepingComputer reported that the outage began Thursday at 1:26 PM ET. Microsoft later published a preliminary incident report, dated November 15, and GBHackers reported what it says.
According to that reporting, the cause was lock contention inside an internal Windows process, which exhausted CPU and memory resources. Within less than two hours, over 80% of gateway servers were unresponsive.
In general, a gateway is a server that traffic passes through on its way to the service behind it. When gateways stop answering, the service behind them is out of reach, even if parts of it are still running.
Email took most of the damage. GBHackers reports that the outage primarily disrupted access to Exchange Online mailboxes and delayed email delivery, with customer impact totaling 23 hours and 50 minutes.
Was the Microsoft 365 outage a cyberattack?
Microsoft says no. In a statement quoted by BleepingComputer, the company said: "This was a technical issue not related to cyber or AI / agents."
That is Microsoft describing its own incident. Nothing in the sources we checked contradicts it.
What is confirmed, and what is not
First, a note on sourcing: we are not quoting Microsoft's incident report directly. The facts below come from two news outlets' reporting of Microsoft's statement and report, which is the best source available to us. The report is labelled preliminary, so details may change.
| What is reported | Where it comes from |
|---|---|
| The outage began Thursday at 1:26 PM ET | BleepingComputer |
| It was not a cyber or AI-related issue | Microsoft statement, quoted by BleepingComputer |
| Lock contention in an internal Windows process exhausted CPU and memory resources | Microsoft's preliminary report, via GBHackers |
| Over 80% of gateway servers were unresponsive within less than two hours | Microsoft's preliminary report, via GBHackers |
| Exchange Online mailbox access and email delivery were the main services disrupted | GBHackers |
| Customer impact totaled 23 hours and 50 minutes | GBHackers |
What the sources we checked do not tell us:
- What started the lock contention. The reporting describes the mechanism, not the trigger.
- How the fault spread from the first servers it hit to most of the gateway fleet.
- How many customers were affected, or where. We found no customer count and no list of countries.
- Whether any email was lost. Delivery was reported as delayed. We found no statement either way about lost messages.
- How badly other Microsoft 365 apps were affected. Email was the main impact. We could not confirm details for anything else.
- What Microsoft will change to stop it happening again.
If you see a number or a cause somewhere else that is not on this list, treat it as unconfirmed until Microsoft's own report says it.
Why a Microsoft 365 outage matters for your business
You did nothing to cause this. Nobody skipped a patch or misconfigured a firewall. That is exactly why it matters: some outages are entirely in your provider's hands, and all you control is how ready you are.
For most businesses on Microsoft 365, email is more than a tool. Orders arrive by email, invoices go out by email, vendors confirm deliveries by email, and staff reset passwords for other systems by email. A day of blocked mailboxes and delayed delivery hits all of that at once.
Lock contention, in plain English, is what happens when many tasks need the same resource and have to wait their turn for it, like a store with one checkout lane on its busiest day. The waiting itself uses up CPU and memory until the machine has nothing left for real work. The sources we checked do not explain exactly how this spread across Microsoft's servers, so we will not guess. The general pattern is still worth knowing: when servers share the load, a fault that takes some of them out can push more work onto the rest.
Your own setup has the same weak points on a smaller scale: one server, one firewall, one internet line, one admin account. In the cloud or on your own hardware, the question is the same: what happens when the one thing everything passes through stops answering? Running your own servers does not make you immune. It moves the responsibility to you, and our self-hosting and infrastructure page explains what that involves.
What to do this week: Microsoft 365 outage checklist
Nothing in the reporting asks customers to install an update or change a setting. This week's work is about recovery and planning.
| Task | Why it matters | How to do it |
|---|---|---|
| Confirm that time-sensitive email from the outage window arrived | Delivery was delayed | Ask key customers and vendors to confirm the orders, invoices and approvals they sent or expected |
| Keep a contact list outside Microsoft 365 | You cannot look up a phone number in a mailbox you cannot open | Staff, top customers and vendors, on paper or in a separate system |
| Choose a backup channel that does not run through the same provider | Your team still needs to talk to each other | A phone tree, text messages or a separate messaging service, agreed in advance |
| List everything that depends on your email | An email outage spreads past the inbox | Website contact forms, online store order alerts, password reset emails, supplier portals |
| Prepare a notice for customers | Customers notice the problem before you explain it | A short website banner and phone message saying email is delayed and how to reach you instead |
| Know where to check status without your own email | Your inbox may be the thing that is down | Bookmark your provider's status page on a phone and name who checks it |
| Review this outage with your team | A real incident is the best drill you will get | How long it took to notice, what stopped working, what you would do differently |
Microsoft 365 outage questions people are asking
What caused the Microsoft 365 outage?
According to reporting on Microsoft's preliminary incident report, lock contention inside an internal Windows process exhausted CPU and memory resources. Within less than two hours, over 80% of gateway servers were unresponsive. The sources we checked do not say what triggered the contention.
Was the Microsoft 365 outage a hack?
Microsoft says it was not: "This was a technical issue not related to cyber or AI / agents."
How long did the Microsoft 365 outage last?
Customer impact totaled 23 hours and 50 minutes, according to GBHackers' reporting of Microsoft's report.
When did the Microsoft 365 outage start?
BleepingComputer reported that it began Thursday at 1:26 PM ET.
Which Microsoft 365 services were affected?
Mainly email: access to Exchange Online mailboxes, and email delivery. We could not confirm details for other apps.
Were emails lost in the Microsoft 365 outage?
Delivery was reported as delayed. We found no statement on whether any messages were lost, so check important messages with the people who sent them.
Do I need to install a fix?
No update or setting change for customers is mentioned in the reporting.
Is Microsoft 365 down right now?
This post covers an incident with a reported total impact of 23 hours and 50 minutes. For current status, check Microsoft's service status, not a news article.
The bigger lesson
Most businesses chose Microsoft 365 so they would not have to run their own mail servers, and that is a sensible choice. But handing over the servers does not hand over the consequences. When over 80% of a provider's gateway servers stop answering, every customer behind them has the same problem at the same time, and the only thing each business controls is how ready it was. Write the email-down plan before the day you need it.
For help mapping what your business depends on and hardening it, see security hardening or request a quote.
Get security alerts by email, with the free checklist, at our security alerts page.
Sources: BleepingComputer; GBHackers.
Have a question about this article?
Ask it here and we will answer you directly by email. Questions are private, never posted.
Security alerts by email
Hear about the next one first.
When attackers start using a flaw in software businesses run on, we email what it is, whether it touches you and what to do that week, with a round-up of our latest security writing. Free, one-click unsubscribe. You get the checklist straight away: 12 things to secure this week.
Want a result like this?
Security alerts, breaches and the technology a business runs on: what happened, who it affects and what to do.
Get a Tech QuoteKeep reading
Microsoft 365 Outage Phishing: Fake Held-Mail Emails Used the November 12 Disruption as Bait, What Is Confirmed and How to Check Who Clicked
Phishing emails used the November 12 Microsoft 365 outage as cover, falsely claiming messages held during the...
Fireworks AI Data Breach: Uber Notifies 149,346 People After Hackers Stole Driver License Data From Its AI Vendor, What Is Confirmed and What Businesses Should Check
Uber says hackers stole drivers' names, addresses, dates of birth and license numbers in a breach at vendor Fi...
PostgreSQL Security Update: 52 Vulnerabilities Fixed Including CVE-2026-6460, PostgreSQL 14 Reaches End of Life, and the Versions to Install
PostgreSQL's November 12 update fixes 52 security vulnerabilities, including CVE-2026-6460, a code execution f...