On Tuesday, 20 October 2026, a power-related disruption at a Google Cloud facility in London knocked out services for millions of users across the UK and parts of Europe. Banks, delivery firms, property websites, the NHS app and even court video links went down for several hours before Google restored most services by the evening.
What Happened
Google Cloud reported issues with multiple services in its europe-west2 region (the London data-centre complex) starting around 07:00 BST. The problems were concentrated in availability zone europe-west2-b at a third-party facility. Google’s status page described “elevated errors and intermittent network failures” affecting dozens of products, including Compute Engine, Google Kubernetes Engine, Cloud Run, Cloud Storage, App Engine, BigQuery, Cloud SQL and others.
Google later confirmed the root cause was a “power event”. In a statement, a Google Cloud spokesperson said:
“We have resolved a disruption to Google Cloud services at a third-party facility in our europe-west2 cloud region due to a power event. Services have been restored.”
The company apologised to affected customers and said it would publish a full incident analysis on its public status dashboard.
Who Was Affected
Downdetector and company statements showed widespread disruption, mainly in the UK:
- Banks and fintech: Revolut (card payments, top-ups, ATM withdrawals and currency exchange were partially unavailable; more than 1,500 user reports shortly after 08:00 BST), Lloyds Bank, HSBC, Halifax and Bank of Scotland.
- Public and government services: HM Courts and Tribunals Service video platform for remote hearings; temporary glitches on the NHS app.
- Other businesses: Royal Mail, parcel firm DPD, property website Rightmove, e-bike service Forest, and various smaller retailers and logistics companies.
Revolut confirmed the issues were linked to a third-party problem and stressed that customers’ money and data remained safe throughout. Rightmove later said technical issues affecting Google Cloud across Europe had impacted some logged-in services before being resolved.
Timeline and Recovery
- Issues first flagged on Google’s status page around 07:00 BST (corresponding to late Monday evening Pacific time).
- Peak user reports mid-morning.
- Google said the power event had been mitigated and recovery of existing workloads was underway by midday.
- By mid-afternoon most customers were back online; only a “very small number” remained affected.
- Full resolution was confirmed by Tuesday evening. Google’s status page lists the incident as lasting roughly 5.5–8 hours depending on the exact measurement.
Why It Matters
The outage is a fresh reminder of how heavily the UK economy and public services depend on a handful of hyperscale cloud providers. In July 2026 the UK government designated Google Cloud, Amazon Web Services, Microsoft and Oracle as Critical Third Parties to the financial sector. This gives the Bank of England, Prudential Regulation Authority and Financial Conduct Authority direct oversight powers focused on resilience.
The Financial Conduct Authority said it was engaging with Google and affected firms. The incident also follows a major AWS outage roughly a year earlier that took down more than 1,000 sites, and a significant Google Cloud disruption in the Netherlands (europe-west4) in July 2026 caused by a voltage transient and cooling issues.
Experts noted that while individual zones are designed for resilience, a single power event at a third-party facility can still cascade into banking apps, court systems and everyday services. As one infrastructure specialist put it, hyperscalers have become “too big to not fail” and so deeply embedded that switching away is often impractical.
What Happens Next
Google has said it will publish a detailed post-incident analysis. UK regulators are already in contact with the company under the new Critical Third Party regime. For businesses, the episode reinforces the value of multi-region architectures, multi-cloud strategies where feasible, and robust offline or degraded-mode procedures—especially for critical payments, logistics and public services.
Services are now restored, but the episode is likely to feature in ongoing discussions about digital resilience and concentration risk in Europe’s cloud infrastructure.