Deutsch
English
Español
Français
Nederlands
Português
Share
On Tuesday 20 October 2026, a Google Cloud outage caused by a power event in the London europe-west2-b availability zone disrupted UK banking apps, paused some court hearings and triggered temporary glitches on the NHS app. The disruption began shortly after 07:00 BST and was largely resolved by mid-afternoon, but it renewed urgent warnings about how heavily British institutions depend on a small number of cloud providers.
What caused the Google Cloud outage?
Google's status page said a data centre hosting the europe-west2-b zone lost external electrical power, and emergency generators failed to start because of a PLC voltage-sensing misconfiguration that did not verify both undervoltage and physical breaker position. Uninterruptible power supply batteries carried the load until they were depleted, leading to complete power loss for servers, storage and network equipment. Traffic that automatically failed over to other zones and regions caused transient latency and errors across Compute Engine, Google Kubernetes Engine, Cloud SQL, Cloud Storage, BigQuery, Cloud Run and other services.
Google reported that primary impact was mitigated after roughly six hours, with most Compute Engine virtual machines healthy by 12:10 UTC and full service availability restored through the afternoon. The company apologised to affected customers and said it would publish an incident analysis. We apologise to affected customers, Google said in its status update.
Which UK services were affected?
The Cloud Video Platform courts used by HM Courts and Tribunals Service went down, preventing barristers and litigants from joining remote hearings in England and Wales. HMCTS confirmed it was working with technology partners and later said the platform was available and operating normally.
- Revolut: card payments, ATM withdrawals, bank card top-ups, currency exchange and sign-in were disrupted; the company said customer money and data were safe.
- High street banks: Lloyds, Halifax, HSBC and Bank of Scotland drew user complaints about online and mobile banking problems and card declines.
- Public services: the NHS app showed temporary glitches, while Royal Mail, Rightmove, DPD, Spotify and Screwfix also reported issues.
Why cloud concentration risk is growing
Cloud concentration risk refers to the danger that many critical services rely on one hyperscale provider, so a single failure becomes widely felt. The UK financial resilience rules were already under scrutiny after Google Cloud EMEA was designated a critical third party to the UK financial sector in July 2026, jointly overseen by the Financial Conduct Authority, Bank of England and Prudential Regulation Authority.
The FCA said it was engaging with Google and affected firms, while Treasury Committee chair Dame Meg Hillier requested details of the response by 4 November. The incident highlights how heavily economies depend on a few cloud providers, an academic said. The disruption came exactly one year after the AWS October 2025 outage knocked UK banks, airlines and HMRC offline for hours, illustrating how cloud concentration risk has become a recurring concern.
2025 AWS outage vs 2026 Google Cloud outage
| Feature | AWS October 2025 | Google Cloud October 2026 |
|---|---|---|
| Root cause | DynamoDB DNS failure in US-East-1 | Power event and generator PLC fault in London europe-west2-b |
| Duration | Up to 15 hours | About 6 hours, with recovery into the afternoon |
| UK impact | Banks, airlines, HMRC | Revolut, high street banks, courts, NHS app |
How should institutions respond?
Technology and resilience experts say organisations should map dependencies, test multi-zone or multi-cloud strategy failover and rehearse crisis communications. The incident also highlighted un-mapped supply-chain dependencies, because many firms were hit through software-as-a-service partners running on Google Cloud.
FAQ
What caused the Google Cloud outage on 20 October 2026?
A power event at Google's London europe-west2-b data centre zone caused external electricity loss; emergency generators failed because of a PLC voltage-sensing misconfiguration, and UPS batteries depleted, bringing down compute and storage services.
Was Revolut customer money at risk?
No. Revolut said customer money and data were safe, although card payments, ATM withdrawals and sign-in were disrupted for several hours.
How long did the Google Cloud outage last?
Impact began shortly after 07:00 BST and services were largely restored by 3pm, with primary mitigation after about six hours and some workloads recovering into the afternoon.
Did the outage affect the NHS and courts?
Yes. The NHS app showed temporary glitches, and HM Courts and Tribunals Service's Cloud Video Platform went down, pausing some remote hearings before service was restored.
What is cloud concentration risk?
It is the risk that many institutions depend on one provider, so a failure at that company causes widespread disruption. Regulators now treat Google Cloud as a critical third party to UK finance.
Conclusion and future outlook
The outage was resolved the same day, but its simultaneity across banks, courts and health services showed how a single power event can ripple through national infrastructure. There was no evidence of a cyberattack. With Google Cloud now under joint regulator oversight, businesses and policymakers are likely to face tougher questions about failover testing, exit plans and whether the UK's digital backbone is too concentrated.
Follow Discussion