Somewhere in western Europe this morning, a data center had a very bad Tuesday, and thanks to the miracle of modern cloud dependency, so did a sizable chunk of the UK. Banking apps blinked, playlists paused, and a lot of people learned just how many of their favorite services share the same landlord.

One Power Event, One Very Long Downdetector List

Google Cloud suffered an outage on October 20 that it attributed to a "power event," with problems first reported around 7am UK time and tied to its western Europe region. Downdetector lit up with user reports for Spotify, Revolut, Lloyds, Bank of Scotland, Halifax, Royal Mail, DPD, Rightmove and NHS England, among others. Google later said the power event had been mitigated and services were recovering.

One caveat worth keeping: Downdetector counts user complaints, not confirmed root causes, so not every name on that list was necessarily knocked over by Google directly. Google's own updates also flagged session connection failures for its Google Security Operations product while teams worked to "restore full availability." A full post-incident report hadn't been published at the time of writing.

The Cloud Still Runs on Extension Cords

For all the talk of infinitely elastic infrastructure, the cloud remains a collection of very physical buildings that need very physical electricity. When one region hiccups, everything pinned to that region hiccups with it, and in 2026 that includes your bank, your parcels and your doctor's appointment system.

The angle most people miss is that this isn't really a "Google problem." AWS, Azure and Cloudflare have all had their own spectacular mornings, and they will again. The real question is architectural: how many businesses parked everything in a single region because multi-region felt like overkill, and how many even know which region their critical apps live in? Resilience is a design decision you make long before the lights flicker.

Outages are inevitable. Being caught completely flat-footed by one is optional.

If this morning made you wonder what happens to your own site when your host's region goes dark, our team can review your hosting, failover and backup setup before the next "power event" runs that test for you, so get in touch.

Source: Tom's Guide