Microsoft’s Outage Shows the Cost of One Vendor Running Everything

After Microsoft’s global email outage, open source rivals say the real problem is putting everything in one basket.
Warp Terminal

You are probably already aware of the recent Microsoft outage. For more than a day last week, a large share of the world’s office workers could not reliably send or receive email.

The trouble started on the afternoon of Nov. 12, when Outlook users found they could no longer reach their mailboxes. Within about an hour, Downdetector had logged reports from 74 countries and territories. Teams and Copilot features that rely on Exchange Online broke too.

Government systems in Singapore and Japan were hit, as were communications at the Supreme Court of the Philippines and, in Australia, the federal police and the country’s spy agency.

Microsoft said the outage was not a cyberattack. In a preliminary explanation, it pointed to an issue from an earlier network capacity upgrade that surfaced after a more recent network change. The company declared service fully restored early on Nov. 14, and has promised a full post-incident report.

The disruption is the latest in a run of Exchange Online failures this year. On Aug. 31, a bad authentication configuration knocked out email and spilled over into Teams, SharePoint, OneDrive and Defender for about 12 hours. Some estimated that lost productivity from the November outage could approach $7 billion.

For the companies selling open source alternatives to Microsoft’s products, the episode is a chance to make an argument they have been making for years. It’s FOSS asked several of them what governments and businesses should be thinking about as they recover. Their answer, broadly, was not that their software never fails. It was that no single company should be able to stop an entire economy’s email at once.

“Errors can happen to any provider,” said Peter Ganten, the chief executive of Univention, a German company that makes open source identity management software. “The more important question is concentration.”

“When hundreds of millions of people in thousands of organizations worldwide depend on the same infrastructure, running the same software and managed by the same company, a single fault in the wrong place can bring public life to a standstill,” Ganten said. This failure appears to have been technical, he noted, but “an external attack could have had equally severe consequences.” The same dependency, he added, leaves entire economies exposed to sudden price increases or politically motivated service restrictions.

The scale he describes is real. Microsoft reported more than 450 million paid commercial seats for Microsoft 365 in January this year.

Nextcloud, which makes a self-hostable file sharing and collaboration suite, struck a gentler note while the outage was still unfolding. “Mistakes happen in the IT industry,” Jos Poortvliet, a co-founder of the company, said in a statement on Friday. “This holds true for companies of all sizes, including us. Our heart goes out to all the admins today.”

But he too pointed to concentration. “Especially in critical infrastructure, monopolies are increasing the risk of stalling businesses globally if an incident occurs,” Poortvliet said. “When everything is hanging on a single vendor, companies can’t switch providers quickly to get back up and running, or have a second setup as backup, or pull the plug when under attack and keep working air-gapped.”

Michael Meeks, the chief executive of Collabora Productivity, which makes the open source Collabora Office suite, said the outage should change how people think about sovereignty. “The recent outage is a reminder that digital sovereignty is about more than where data is stored,” he said. “It is also about having choice and control over the services and infrastructure organisations depend on.”

He recommended spreading critical workloads across independently managed infrastructure and using open source software that organizations can run themselves. That, he said, can “help prevent one failure from disrupting everything.”

The weak point is the login

The outages this year have shown how one fault can spread. In August, a problem with Microsoft’s authentication configuration did not just break email. It took down other Microsoft 365 services that sit on the same sign-in layer.

Ganten said that was the risk worth worrying about most. “When authentication fails, every service that depends on it fails with it,” he said. Identity, in his view, should be an open and separate layer that different providers can run, so that services stay reachable when one of them goes down. Univention sells an identity platform, Nubus, built on that idea.

‘Outsourced operations’ is not ‘outsourced risk’

Galina Goduhina, the commercial director of ONLYOFFICE, an open source office suite, said organizations had misread what the cloud bought them. “The cloud has made infrastructure someone else’s problem, but it hasn’t made downtime someone else’s consequence,” she said. “Organizations need to stop confusing outsourced operations with outsourced risk.”

The question, she said, is whether an organization has “enough control, portability and independence to keep working when their primary provider cannot.”

She was candid that open source software is not immune. “The difference isn’t that open source fails less,” Ms. Goduhina said. “It’s that failure doesn’t necessarily leave you waiting for a single vendor to act.” Whether organizations can actually use that freedom, she added, depends on their preparation.

That raises an obvious objection: moving from Microsoft’s cloud to another company’s cloud could simply trade one dependency for another. Ganten agreed that it can. “It does, unless the software is open source and built on open standards,” he said. Such software can be run without its original vendor and, in the worst case, maintained by someone else entirely.

Leaving is harder than it looks

None of the companies suggested that switching is quick. Asked what it would take for a 500-person organization to move off Exchange, Goduhina said the mailboxes were the simple part.

“Moving 500 mailboxes is the easy part to estimate,” she said. “Replacing years of accumulated integrations, permissions and habits is not.” She would plan for several months, starting with an audit of dependencies and a pilot group. “The biggest migration risk is discovering too late that Exchange was doing far more than handling email.”

Some governments have already done it. The German state of Schleswig-Holstein moved more than 44,000 mailboxes from Exchange to Open-Xchange by October 2025. Poortvliet said open source alternatives “exist, are ready to use and have been proven at scale.”

Collabora says interest is already rising. Meeks said the company had seen a strong uptick in demand from large public and private sector organizations since the start of the year. That followed Microsoft’s announcement that it would retire Office Online Server, the on-premises software that lets SharePoint users edit documents in a browser. Support ends on Dec. 31. As Microsoft pushes customers toward the cloud, he said, “the need for alternatives that give organisations greater autonomy is becoming more apparent.”

For governments weighing a move, ZenDiS recommends starting with a map of where they depend on a single vendor. The agency is responsible for openDesk, an open source workplace suite already running on about 100,000 public administration workstations. “We recommend a step-by-step approach,” said Lutz Niemeyer, a spokesman for ZenDiS. “Start with an analysis: where are the greatest dependencies? Do they affect critical components or services? What would happen if they were taken offline?”

Organizations should then work through those dependencies in order of how critical they are and evaluate alternatives, he said. He pointed to the ZenDiS sovereignty criteria and the European Commission’s Cloud Sovereignty Framework as tools for the assessment.

For organizations that are not ready to leave, Ganten suggested a smaller step. Where critical services are involved, he said, organizations should be able to keep at least a minimal operation running without their usual provider.

“This fallback,” he said, “must be rehearsed regularly.”

Enjoyed this update? Support independent Linux news coverage

It's FOSS has been helping people use Linux for the past 14 years. Help us stay independent from big tech. Become a Plus member, enjoy ad-free reading and get 5 eBooks.

Plus yearly

Ad-free, FREE ebooks

Join yearly

Buy us a coffee

Any amount, no commitment

Support on Ko-fi
About the author
Abhishek Prakash

Abhishek Prakash

Created It's FOSS 13 years ago to share my Linux adventures. Have a Master's degree in Engineering and years of IT industry experience. Huge fan of Agatha Christie detective mysteries 🕵️‍♂️

Become a Better Linux User

With the FOSS Weekly Newsletter, you learn useful Linux tips, discover applications, explore new distros and stay updated with the latest from Linux world

itsfoss happy penguin

Great! You’ve successfully signed up.

Welcome back! You've successfully signed in.

You've successfully subscribed to It's FOSS.

Success! Check your email for magic link to sign-in.

Success! Your billing info has been updated.

Your billing was not updated.