Fire Destroys OVH's Strasbourg Data Center, Takes Millions of Sites Offline

A fire tore through OVHcloud’s SBG2 data center in Strasbourg, France yesterday, and by the time it was contained the building was gone. Not damaged — gone. OVH also had to shut down the neighboring SBG1 facility as a precaution, which means two of the company’s Strasbourg buildings are effectively out of commission at once. The result: millions of websites across Europe and beyond dropped offline with no warning.

If you don’t recognize the name, OVH (now OVHcloud) is one of the biggest hosting and cloud providers in Europe, the kind of company that quietly sits underneath a huge chunk of the internet you never think about — small business sites, forums, SaaS backends, government services, you name it. That’s exactly why this is such a big deal. When infrastructure at this scale disappears, it doesn’t just take down one company’s product, it takes down whoever happened to be renting space on those particular racks.

Why this hurts so much

The uncomfortable part of this story isn’t the fire itself — buildings catch fire, it happens — it’s what the fire exposed about how a lot of businesses handle disaster recovery. Plenty of the sites and services that vanished yesterday were running entirely out of Strasbourg, with no failover region, no live backups sitting somewhere else, nothing. For those customers, “our data center burned down” isn’t an outage, it’s potentially a full data-loss event. We don’t have final numbers yet on how much is unrecoverable, but the fact that the question is even open should be setting off alarms in a lot of ops teams this morning.

It’s a blunt reminder of a lesson every infrastructure team claims to know and a shocking number don’t actually practice: a backup that lives in the same building as the thing it’s backing up isn’t a backup, it’s a second copy of the same risk. Multi-region replication, offsite backups, tested restore procedures — none of it is exciting work, and it’s exactly the kind of thing that gets deprioritized right up until the day a data center literally burns to the ground.

What happens next

Expect a scramble over the next few days as affected customers try to figure out what’s recoverable and what isn’t, and a longer conversation industry-wide about concentration risk in cloud hosting. OVH built a reputation partly on being a large, France-based alternative to the American hyperscalers, which made it an attractive option for European companies wanting data sovereignty and cost efficiency. That reputation is going to take a hit regardless of how well the response goes, simply because “a data center burned down and stayed down” is about as bad as infrastructure headlines get.

For anyone reading this who runs anything in the cloud, single region or otherwise, today’s a reasonable day to go check when your backups last ran and where they actually live.