Trebron Welcomes Productive to the Team! // Click to Learn More >

Cybersecurity

When the Cloud Goes Dark: Lessons from 2026’s Year of Outages…(so far)

Christopher Tress 4 min read
In this article

On Thursday, November 12, the cloud went dark for millions of workers. A Microsoft 365 outage disrupted Outlook, Teams, and Copilot for users in more than 70 countries, and Outlook outage reports passed 130,000 at the peak (Maintech). For many organizations, the disruption carried into a second workday. Microsoft confirmed it was not a cyberattack. Its preliminary review traces the outage to a software bug introduced in July, which surfaced when failover traffic hit servers already running at reduced capacity (Maintech). 

The year in context 

November was not the only dark day this year. Through mid-November, 2026 has recorded 19 major outages at AWS, Azure, Google Cloud, and Cloudflare, already more than the 16 in all of 2025 (Control Plane Cloud Outage Tracker). Other critical systems went down too, including two in the same week as the Microsoft outage: 

  • A nationwide FAA ground stop on November 13, when a router fault during a routine network upgrade halted departures for more than four hours (Maintech) 
  • A software bug in a major US bank’s own systems that disrupted online banking for thousands of customers on November 16 (Maintech) 

Physical failures also made a comeback. Power and cooling outages were less common than change-related ones, but they took roughly four times as long to resolve (Control Plane 2026 Cloud Outage Report). 

What the data tells us 

Providers often share limited detail after an incident, and full postmortems can take months. The data we do have points in one direction. Of the 63 major cloud outages Control Plane has documented since 2020, 54% started with a configuration change, a software bug, or automation acting on bad input (Control Plane Cloud Outage Tracker). 

The most common cause of downtime is routine work that does not go as planned. Maintech’s review of the year reaches the same conclusion: staying online depends more on operating discipline than on the size of the organization (Maintech). 

Three questions worth asking 

  1. If Microsoft 365 went down again tomorrow, how would your team communicate?

    If your plan relies on email or Teams, it fails at the same moment they do. A current phone list, a text tree, or a messaging channel outside Microsoft 365 gives your team a path that does not depend on the system that failed.

  2. When did you last restore from a backup?

    Veeam’s 2026 resilience research found that 90% of organizations are confident they can recover within their target window. Only 28% actually recover their data in full. A backup is proven only after it has been restored, so schedule restore tests and document the results. It also matters where those copies live. Keep key IT and business information in more than one location, such as on-premises and in the cloud, or across two different cloud providers, with the same encryption and access controls in each. That way, one outage cannot take every copy offline at once. 

  3. Does your recovery plan depend on the system that failed?

    Many cloud recovery steps run through the provider’s management console or API. If that control plane is part of the outage, the plan cannot run. Rehearsing failover ahead of time shows whether it will work under real conditions. 

Back to basics, whatever the cause 

None of this year’s biggest outages were cyberattacks, but the lesson applies to security just the same. In the first few minutes, an outage and an attack look alike: systems go quiet, people can’t log in, and no one is sure what is happening. The same fundamentals carry you through either one: a way to communicate, backups kept in more than one place and actually restored, and a recovery plan your team has practiced. 

Disruptions also open the door for attackers. When a major service goes down, phishing emails posing as IT support or password resets often follow, and stretched teams are easier to catch off guard. 

That is why continuous monitoring belongs on the list of basics. Managed detection and response, with security analysts watching your environment around the clock, gives you a steady set of eyes when your own team is stretched. When something goes wrong, it helps confirm whether you are dealing with a provider outage or a real threat, so you are not working through it alone. 

Preparing for cloudy days 

Cloud providers will keep updating their platforms, and occasionally the cloud will go dark. You cannot control the forecast. You can control how ready your organization is when it changes: a communication fallback, tested backups, and a recovery plan your team has practiced. 

Were you affected by the Microsoft outage? Reply to this email or reach out to your Trebron team. A review of your continuity plan and security monitoring is a practical place to start, and we can help you find the gaps before the next disruption does. 

Sources 

Share

Ready To Learn More?