Google Cloud London Outage Takes Down Revolut and Banks
Google blames a power event in its London zone europe-west2-b. Revolut customers could not pay by card or withdraw cash.
In brief
- The outage began on October 20, 2026 at 06:08 UTC in the London zone europe-west2-b.
- Google cites a power event, with no further detail.
- Revolut, British banks, delivery and postal services, and even video feeds from a trial went down.
- Revolut says customer money and data are safe.
Why did Revolut fail at the checkout on Tuesday morning, and half of Britain’s banks with it? The answer lies in a single zone of Google’s cloud in London, where, according to the company, a power event took place. Google’s latest update says the problem has been mitigated for most users.
When did it start and what went down?
The outage began at 06:08 UTC. That is 07:08 British Summer Time and 08:08 in Serbia, since clocks do not go back until October 25. Australia’s ABC News reports that it began “a little after 6 a.m. British time,” which does not match the UTC time. For the exact time we rely on the figure from Google’s status page, as relayed by The Register.
The affected zone is europe-west2-b, one of three availability zones in Google Cloud’s London region (europe-west2). The status page lists these services as affected:
- BigQuery Data Transfer Service
- Cloud Filestore
- Cloud Run
- Google App Engine
- Google Cloud NetApp Volumes
- Google Kubernetes Engine
- Managed Service for Apache Kafka
Google posted the following on its status page:
“A power event caused an outage in the europe-west2-b zone. The power event has been mitigated. We are seeing recovery of services and are prioritizing recovery of existing workloads. Customers may still be unable to create new workloads.”
The Register asked for further explanation, but Google replied that the only information available is what appears on the status page.
Who felt the outage?
According to British media, Lloyds Bank, HSBC, Royal Mail, Rightmove, DPD, BT, Screwfix and Revolut were hit. British outlets also mention Halifax and Microsoft. Downdetector logged hundreds of reports, including for Rightmove and Royal Mail. ABC writes that the outage reached everything from electric bike rentals to budgeting apps.
Halifax wrote on X that customers were “having problems with our app and online banking.”
Court feeds stalled too. His Majesty’s Courts and Tribunals Service (HMCTS) posted on X:
“We are aware of an issue affecting the Cloud Video Platform (CVP), which is used for some remote hearings.”
Among the feeds that dropped was the one from the trial of comedian Russell Brand.
What happened with Revolut?
If you have a Revolut card, and plenty of people in Serbia and the region do, Tuesday morning looked like this: the card was declined, the app would not let you log in, sending money did not work, and neither did cash withdrawals at ATMs. Reports came in from the United Kingdom, Germany, Romania, Greece and Croatia.
Revolut responded on social media:
“We currently have an outage affecting multiple services.”
The company added:
“We can confirm that all of our customers’ money and data are safe.”
An unavailable app is not the same as missing money. Kurir advises checking the status of a transaction before trying again, so you do not pay twice. The same outlet says the latest available reports did not confirm full normalization of all Revolut services, including card payments and cash withdrawals, even though Google had meanwhile reported recovery.
When did Google say it was over?
At 11:58 UTC (13:58 in Serbia), Google said full recovery was expected by roughly 13:00 UTC (15:00 in Serbia). At 14:03 UTC (16:03 in Serbia), the status page reported that the problem had been mitigated for most users.
In the final notice, which The Register relayed at 17:37 UTC, Google wrote:
“The issue with the affected GCP products has been mitigated for the majority of users and is believed to be affecting only a very small number of users at this time, and our engineering team is working on it.”
The same notice carries two sentences about what comes next: “No further updates will be posted here. We will publish an analysis of this incident once we have completed our internal investigation.”
What does an expert say?
Dr Amro Al-Said Ahmad, a lecturer in computing at Keele University in the United Kingdom, told The Register that 38 services went down in the outage. He added:
“The incident shows the dependency effect of modern cloud ecosystems.”
He follows that with a lesson for anyone building on the cloud:
“Resilience is not achieved by simply moving to the cloud. True resilience requires designing for regional failures through multi-region deployment, workload portability and tested disaster recovery procedures.”
What exactly failed at the London facility is left to Google’s internal investigation. The company has promised to publish its analysis once it is finished.
