Microsoft 365 Outage (Nov 12): Was AWS Down? What Happened, and How to Claim SLA Credits

Outlook, Teams and Copilot went down on Nov 12, 2026. AWS says it wasn't affected. What happened, how to claim Microsoft SLA credits, and the AWS comparison.

Lov Singh13 November 20269 min read0 viewsAWS
Microsoft 365 Outage (Nov 12): Was AWS Down? What Happened, and How to Claim SLA Credits

A large Microsoft 365 outage on Thursday, November 12, 2026 knocked out Outlook (Exchange Online), Teams, SharePoint Online, Microsoft 365 Copilot and Defender XDR for hours. AWS was not down. Amazon says its services ran normally and that some reports linking AWS to the disruption were inaccurate. If your company pays for Microsoft 365, you may be owed a service credit, but you have to claim it yourself.

This is a Microsoft story, so why is it on 9to5Amazon? Because "is AWS down?" was one of the questions people searched during the outage, and because many businesses run both Microsoft 365 and AWS. Below: what Microsoft has confirmed, how its SLA credits work, and how that compares with Amazon's own SLAs.

What happened in the Microsoft 365 outage?

Microsoft says the incident began at 1:26 p.m. ET on November 12 (about 11:56 p.m. IST). Its admin advisory MO1550821 said users might be unable to connect to:

  • Exchange Online (Outlook email and calendar)
  • Microsoft Teams
  • SharePoint Online
  • Microsoft 365 Copilot
  • Microsoft Defender XDR

The Exchange part is tracked under EX1550543. Microsoft first described the problem in the Americas. It later said users in any region could be affected.

On Downdetector, Outlook problem reports peaked at about 138,000 around 1 p.m. ET, according to Reuters. Reuters said services stayed disrupted for more than five hours. The outage also hit Microsoft 365 connectors in AI tools: OpenAI said it was investigating Microsoft connector issues, and Anthropic said an "upstream provider" issue affected its Microsoft 365 connector.

What caused it?

Microsoft's explanation changed as it investigated. These are its updates on the Microsoft 365 Status account on X, as reported by CNBC and BleepingComputer:

  1. It first blamed elevated load across multiple regions.
  2. It then said the load "isn't driven by an increase in traffic" but by a number of servers being unexpectedly unavailable, which strained the rest.
  3. It said the servers were failing automated health checks, and that a core process on affected servers was becoming overwhelmed, causing capacity loss.
  4. It validated an alternative health-check method, restarted affected servers, paused deployments and shifted traffic to other routes.

Microsoft has not reported a cyberattack. A full root cause should come in its post-incident report.

A separate Teams incident earlier that day

From 9:45 to 10:53 a.m. ET, Teams had an unrelated-looking problem (TM1550488): messages failed or were delayed in North and South America. Microsoft blamed resource contention and said it was fully resolved. It has not said whether the two incidents are linked.

Is it fixed?

By the evening of November 12 (ET), Microsoft reported limited recovery as servers came back, and Downdetector reports fell sharply. At the time of writing we could not find an official "fully resolved" notice. Check status.cloud.microsoft or the Service Health page in the Microsoft 365 admin center for the current state.

Was AWS down on November 12?

No. CNBC reported that AWS said its services were operating normally. AWS put the inaccurate reports down to an "unrelated disruption elsewhere on the internet" and pointed customers to the AWS Health Dashboard for accurate status.

Why do outages get mixed up? Downdetector counts complaints from users, not confirmed failures. When Outlook and Teams break, people report problems with whatever they think is involved, so unrelated services can show spikes too. For AWS, the reliable source is always the AWS Health Dashboard, not social media or complaint charts.

Microsoft SLA credits: who can claim, and how much?

Microsoft's commercial Online Services SLA promises a monthly uptime figure for each service. For Exchange Online, SharePoint Online and Teams, that figure is 99.9%. In a 30-day month, 99.9% allows about 43 minutes of downtime.

The credit tiers commonly cited for these services are below. We could not open Microsoft's current SLA document directly, so check the version for your agreement before you file.

Monthly uptime Service credit
Below 99.9% 25%
Below 99% 50%
Below 95% 100%

99% in a 30-day month is about 7.2 hours of downtime. So a single long outage does not automatically put you in the 50% tier.

How to claim a Microsoft 365 SLA credit

  1. Find the incident IDs. In the Microsoft 365 admin center, open Health > Service health and note EX1550543, MO1550821 and, if Teams was hit in the morning, TM1550488.
  2. Open a support request from the admin center and say you are requesting an SLA service credit.
  3. Include the details: your tenant ID, the incident IDs, the services affected, when they stopped and came back for your users, and how many users were affected.
  4. Bought through a partner? If your licences come from a CSP partner or reseller, file through them. Pax8, a large Microsoft distributor, says approved credits go to the partner's account and the partner passes them on.
  5. Meet the deadline. Pax8's guidance says claims are due by the end of the calendar month after the incident. For a November 12 incident, that means December 31, 2026.

Things that surprise people

  • It's not a cash refund. Credits reduce a future bill.
  • Only the affected service counts. The credit applies to the fee for that service, not your whole Microsoft 365 bundle.
  • Downtime is measured per user. If only some of your users were affected, your uptime may still be above the threshold.
  • Microsoft reviews every claim. Approval is not automatic.

How do AWS SLA credits compare?

Amazon's nearest product to Exchange Online is Amazon WorkMail, its business email service. AWS publishes SLAs for its services, and they work in a similar way.

Microsoft 365 (Exchange Online) Amazon WorkMail Amazon EC2 (region level)
Uptime commitment 99.9% 99.9% 99.99%
First credit tier 25% 10% 10%
Middle tier 50% (below 99%) 25% (below 99%) 30% (below 99%)
Full credit 100% (below 95%) 100% (below 95%) 100% (below 95%)
How to claim Support request in the admin center, or via your partner Case in the AWS Support Center Case in the AWS Support Center
Deadline End of the month after the incident (per Pax8) End of the second billing cycle after the incident End of the second billing cycle after the incident
Cash refund? No, bill credit No, bill credit No, bill credit

Two differences stand out:

  • Microsoft's first tier is more generous (25% against WorkMail's 10%) for the same 99.9% promise.
  • AWS gives you more time to claim. For a WorkMail claim, AWS asks for the subject line "SLA Credit Request", the dates and times of each incident, and request logs that show the errors.

In every case, an SLA credit is small next to the cost of a lost working day. It is compensation, not insurance.

Editor's take

This outage is a reminder that almost every company now depends on one or two cloud providers for its most basic work, email and chat. It also showed how quickly a Microsoft problem turned into "is AWS down too?" The lessons are the same whichever cloud you use:

  • Know where the real status page is. For Microsoft, status.cloud.microsoft and your admin center. For AWS, the AWS Health Dashboard.
  • Write down the incident IDs on the day. You need them for any credit claim.
  • Have a fallback for urgent communication. A phone tree, SMS or a second messaging tool costs little and matters most on days like this.
  • Read your SLA before you need it. Most people discover the deadlines and exclusions only after an outage.

New to cloud basics? Our guide to what AWS is explains regions and availability, and our AWS re:Invent 2026 guide covers the event where AWS usually talks about reliability.

What readers are asking

We didn't find community comments we could attribute and link reliably for this story. These are the questions people were searching during and after the outage, answered below.

FAQ

Is Microsoft 365 down right now?

Microsoft reported limited recovery on the evening of November 12 (ET). For the current state, check status.cloud.microsoft or the Service health page in the Microsoft 365 admin center.

Was AWS down on November 12, 2026?

No. AWS said its services were operating normally and that some reports were caused by an unrelated disruption elsewhere on the internet.

Why is Outlook not working today?

On November 12, Outlook's back end, Exchange Online, was hit by servers becoming unavailable and failing health checks, according to Microsoft. If Outlook fails on another day, check Microsoft's status page first.

Did the Microsoft outage affect India and Europe?

Microsoft first reported problems in the Americas, then said users in any region could be affected. In India the outage began around midnight IST.

Can I get a refund for the Microsoft 365 outage?

Not a cash refund. Business customers can request a service credit on a future bill if monthly uptime falls below the SLA.

What is the deadline to claim Microsoft SLA credits?

Partner guidance from Pax8 says the end of the calendar month after the incident, which is December 31, 2026 for this outage. Confirm against your own agreement.

What are the incident IDs for the November 12 outage?

MO1550821 (Microsoft 365 services), EX1550543 (Exchange Online) and, for the earlier Teams problem, TM1550488.

Does AWS give credits for outages too?

Yes. AWS services such as Amazon WorkMail and Amazon EC2 have published SLAs with credit tiers, claimed by opening a case in the AWS Support Center by the end of the second billing cycle after the incident.

Sources

This article is about Microsoft services. 9to5Amazon is independent and not affiliated with Amazon or Microsoft.

Featured image: Microsoft Building 92 on the Redmond, Washington campus. Photo by Coolcaesar, CC BY-SA 4.0, via Wikimedia Commons. Source

Share:

Comments

Leave a comment