Google Cloud Outage in London Leaves Revolut Down

X (Twitter) Facebook Pinterest LinkedIn E-mail

An electrical incident at one of Google Cloud’s data centers in London left Revolut without much of its service in Spain and the rest of Europe for several hours on October 20. The outage began at 08:08 Spanish time in the europe-west2-b zone of the London region and affected dozens of Google Cloud products. Card payments, transfers, and cash withdrawals at Revolut were blocked while Google worked to restore the zone.

The Google Cloud London outage in 30 seconds

A power outage in Google Cloud’s europe-west2-b zone in London caused errors across dozens of Google Cloud products starting at 08:08 on October 20.
Revolut reported problems with 31 services in Spain, including card payments, transfers, and ATM withdrawals.
British banks and companies such as Lloyds, HSBC, Royal Mail, and BT were also affected, according to Downdetector.
Google has not yet published a full root-cause report.

According to Google Cloud’s status dashboard, the incident started at 23:08 Pacific time on October 19, or 06:08 UTC. In its first updates, the company described elevated errors and intermittent network failures across several services in the europe-west2 region, and advised customers who could do so to move their workloads to another zone or region. Its first public communication came roughly two and a half hours after the start, according to the timeline reconstructed by StatusStack. By around 13:15 Spanish time, Google reported that the power issue had been fixed and that service recovery would be gradual, as reported by elDiario.es.

Revolut, the most visible customer in Spain

In Spain, the impact was felt above all at Revolut. According to information from El País picked up by DCD, the company identified problems with as many as 31 services, including card payments, transfers, ATM withdrawals, balance top-ups, and investment operations, as well as its business product, Revolut Business. Many users couldn’t log into the app or pay at stores during the morning.

Revolut sent a reassuring message, saying “all customer funds and data are completely safe,” and that its teams were working to restore normal service. According to the company, this was a systems availability issue, not a security breach or data loss. The digital bank is one of Google Cloud’s flagship customers and had expanded its contract with the provider to support its international expansion.

Downdetector also logged incidents among CaixaBank and Spotify customers in Spain during the same window, though it hasn’t been confirmed that they were related to the same cause. In the UK, The Register cited problems at Lloyds Bank, HSBC, Royal Mail, Rightmove, DPD, BT, and Screwfix, and providers such as Hostinger, MongoDB Atlas, and JFrog reported disruptions to services hosted in that zone.

One zone down, dozens of products affected

Google Cloud regions are split into several availability zones — separate data centers designed so that a failure in one doesn’t take down the rest. This time the problem was concentrated in europe-west2-b, one of London’s three zones, but its effect was far broader than that would suggest. By mid-morning, Google’s dashboard listed as many as 37 affected products in the region, including Compute Engine, Cloud SQL, Cloud Storage, GKE, BigQuery, Cloud Run, and Pub/Sub.

Google has not yet published the root cause or a post-incident report. The first clue about the origin came from Hostinger, which said Google had told it about a power failure at one of its data centers — something the company’s own updates later confirmed when they referred to an electrical event. It isn’t the first time: in October 2024, an electrical arc in a power distribution unit knocked out part of the europe-west3-c zone in Frankfurt for more than seven hours, according to the analysis Google published at the time.

The same old lesson: depending on a single zone

The episode once again highlights how dependent large digital platforms are on a handful of cloud providers and, within them, on a single region. Architectures spread across multiple zones, or across multiple regions and providers, cost more and are more complex to run, but they’re what makes it possible to absorb a failure like this one.

In European banking, the question is no longer just technical. Since January 2025, the DORA digital operational resilience regulation has required EU financial institutions to manage the risk posed by their critical technology providers and test their continuity plans. Lawyer Ryan Gracey, of the law firm Gordons, warned in The Guardian that many companies would discover that day that their cloud contracts protect the provider more than the customer — a debate that connects with how data centers have become targets that threaten business continuity and with the bet many organizations are making on multicloud strategies.

Frequently asked questions

Why did Revolut go down on October 20?

Because of a Google Cloud outage in the europe-west2-b zone in London, caused by an electrical incident at a data center. Revolut hosts part of its systems on that infrastructure.

Was Revolut customers’ money at risk?

According to Revolut, no. The company said customer funds and data were safe and that this was a service disruption, not a security breach.

Which Google Cloud services were affected?

Up to 37 products in the London region, including Compute Engine, Cloud SQL, Cloud Storage, GKE, BigQuery, and Cloud Run, according to Google’s status dashboard.

Has Google explained the cause?

It has referred to an electrical event in the europe-west2-b zone, which has since been fixed, but it hasn’t yet published the full incident analysis.

Sources: