Google datacentre power outage in London disrupts banks, courts and NHS app
A power outage at a Google datacentre in London's europe-west2-b zone disrupted banking services, paused court hearings and caused temporary glitches on the NHS app, with problems peaking around 9am and mostly resolved by mid-afternoon.
Google datacentre power outage in London disrupts banks, courts and NHS app
A power outage at a Google datacentre in London disrupted banking apps, put court hearings on hold and caused temporary glitches on the NHS app. The affected facility sits in Google's europe-west2-b zone, one of more than a dozen zones the company operates in Europe.
The knock-on effects were mostly confined to UK businesses, since Google Cloud runs operations for many of them. According to Downdetector, the disruption peaked around 9am and was somewhat resolved by mid-afternoon.
Services knocked offline
The outage faltered multiple banking services, put court hearings on hold and drew patient reports of temporary glitches on the NHS app. Downdetector listed Royal Mail, Revolut, HSBC and Lloyds bank among the affected services.
What Google Cloud's customers said
Revolut confirmed it was dealing with "a service disruption affecting multiple services", but added "all our customers' money and data are safe".
Lloyds said its app was working, while acknowledging "we understand some [people] are having issues with certain online services this morning".
HM Courts and Tribunals Service, which runs remote video hearings in courts across England and Wales, reported an issue affecting its Cloud Video Platform and said it was working with technology partners to restore service.
Smaller than the AWS outage
The disruption is so far not on the scale of a similar cascade a year earlier, when Amazon Web Services went down for nearly half a day. That incident lasted about 15 hours and originated in Amazon's us-east1 region in Virginia, and was later found to have been caused not by a power outage but by a bug in automation software.
Virginia hosts the greatest density of datacentres on Earth, and the AWS failure took out the Roblox game and the Signal messaging app while causing smart mattresses worldwide to suddenly heat overnight.
A separate AWS outage was linked to an AI tool that Amazon adopted after laying off tens of thousands of staff last autumn.
Concentration concerns
Michał 'rysiek' Woźniak, an internet infrastructure expert, said the outages are a reminder of the increasing concentration of the internet and vital services on the infrastructure of a few key companies.
Hyperscalers have become too big to not fail. But worse, they've become so deeply ingrained into the very fabric of our digital lives.
Woźniak continued that switching away from the hyperscalers "seems unimaginable". He added that while he was sure Google's engineers were doing everything they could to mitigate the problems, it is unreasonable to have a handful of gigantic companies repeatedly cause public institutions or large parts of the digital economy to stop working en masse.
Woźniak also noted a Google announcement earlier this year that 75% of all new code is "now AI-generated and approved by engineers", and wondered how that affects the reliability and quality of Google's services, pointing out that multiple large outages at AWS a year ago were linked to internal AI tooling.
The concentration question extends beyond Google: the Guardian has reported that British police keep vital data on Microsoft's Azure cloud despite apparently being aware this exposes sensitive information to US government insiders, hostile foreign actors and outages. Up to 60% of the UK government's infrastructure is hosted on cloud services, mostly Amazon, Microsoft and Google.
Google has been approached for comment.
