Ten Hours Without a Trade

INDEPENDENT EVIDENCE

Thanksgiving night, 2025. Aurora, Illinois.

A chiller plant at CyrusOne's CHI1 data center failed and took multiple cooling units down with it. That building is the main hub for CME Group, the world's largest derivatives exchange.²

By Friday morning, markets around the world had stopped moving.

CME's trading was down for more than ten hours. Futures on the S&P 500, the Nasdaq 100 and U.S. Treasuries stopped updating. Currency trading went dark. Crude oil, gold and even palm oil in Malaysia felt it. Traders in Asia and Europe were hit first, because the halt landed in their working day.¹

Inside the building, temperatures reportedly climbed past 100°F. The site had redundancy. It even used free outside cooling when temperatures dropped below 30°F, and that night was cold. It still wasn't enough.⁴

CME had a disaster recovery site in the New York area. It chose not to switch, because early information pointed to a short outage. The outage was not short.

Afterward, CyrusOne said it added more backup cooling at the site.³

Be clear about what this was

This was a mechanical failure. Nobody has reported it as a cyberattack, and we don't suggest it was one.

So why open a security book with it?

Because it shows exactly what's on the line. One cooling plant in one suburban building stopped a slice of the global financial system for most of a day.

Ten hours. One cooling plant. No hacker.

Now imagine the same chiller plant run by a controller that someone else can reach.

The lesson

A cooling controller is a part of customer uptime. Treat it with the same seriousness you give the servers it protects.

And remember what slowed the response that night: uncertainty. Nobody knew how long the problem would last, so nobody made the big call. Knowing what depends on each system helps the team make that decision sooner.