Google Cloud outage hits millions as UK courts and banks go offline

A London cloud fault with street-level consequences

A Google Cloud outage in the London-based europe-west2 region turned into a highly visible failure for UK banks, courts and online services on Tuesday, October 20, 2026. Google’s incident record says the disruption began at 06:08 UTC and ended at 14:00 UTC, with the most recent update posted at 14:03 UTC saying the issue had been mitigated for most customers while a very small number could still be affected [1]. The company attributed the disruption to a “power event” in europe-west2-b, one of the zones that make up its London region [1].

The impact was not confined to engineers watching dashboards. Customers of Revolut, Lloyds, Halifax, HSBC and Bank of Scotland reported trouble using apps, online banking and, in Revolut’s case, payment functions such as card payments and ATM withdrawals [2][3]. The Guardian reported that millions of customers were unable to access or move cash at affected banks, and that court hearings across England and Wales were disrupted when the cloud video platform used for remote attendance failed during the morning [2].

By early afternoon in London, recovery had started. Google said services were recovering for most workloads at 12:44 UTC and then marked the status as available in its 14:03 UTC update, while still saying its engineers would keep working with customers that remained affected [1]. Reuters later reported that Google said full service availability had been restored after the hours-long outage, although the episode had already exposed how widely one cloud region can ripple through financial and public infrastructure [3].

What Google says happened

The key technical phrase in Google’s update is “power event.” Google said the power event caused an outage inside europe-west2-b, after which Google Cloud services were unavailable in that zone and some services in the wider europe-west2 region experienced elevated errors and intermittent network issues [1]. The status history shows an escalation from investigation to mitigation: early updates described errors and network failures, later updates identified the power event, and the final update said the affected products had been mitigated for most customers [1].

Google’s own JSON incident feed listed “Multiple Products” as the affected service category, not a single failed component. The affected product list included infrastructure and platform services such as Google Compute Engine, Google Kubernetes Engine, Cloud Run, Cloud SQL, Cloud Storage, Cloud Load Balancing, Cloud Pub/Sub, BigQuery and many others [1]. That breadth matters because modern applications rarely depend on only one cloud product; a banking app, for example, may need compute, networking, databases, queues, identity and storage to all behave at once.

The incident page also said Google would publish an analysis once its internal investigation was complete [1]. Until that post-incident report appears, the public record supports a cautious conclusion: this was not a cyberattack claim, not a customer data breach claim and not merely a single website going down. It was a regional cloud infrastructure event that cut across multiple Google Cloud products and then surfaced through banks, courts and consumer-facing services.

Banks, payments and customers

Revolut was one of the most visible casualties because the outage touched services that customers feel immediately: sign-in, card payments, transfers, top-ups and ATM withdrawals [2][4]. ITV News reported that Revolut said multiple systems were unavailable or partially unavailable, while the company told customers their money and data were safe [4]. Reuters also reported that Revolut’s status page showed card payments and ATM withdrawals as affected, and cited a person familiar with the matter saying the Revolut outage was related to the Google Cloud disruption [3].

Traditional banks were hit too, though the public links between each bank’s issue and Google Cloud were not identical. Lloyds said it expected services to return to normal soon after the root issues were resolved, and HSBC said some customers were having difficulties accessing services [3]. The Guardian reported problems at Lloyds, HSBC and Halifax, while also noting disruption to Revolut’s sign-up, sign-in, card payments, transfers, ATM withdrawals and cryptocurrency services [2]. Euronews reported that Lloyds and HSBC confirmed customers were facing banking-service issues, while Revolut’s disruptions included ATM withdrawals, transfers and card top-ups [5].

For customers, the distinction between a bank outage and a cloud outage may feel academic. A declined card, a frozen app or an unreachable account is experienced at the point of need, not in an architecture diagram. That is why cloud resilience is no longer just a back-office technology topic. When payments, identity checks and digital customer service are cloud-dependent, a zone-level power problem can become a public confidence problem within minutes.

Courts and public services

The outage also affected the public sector. HM Courts & Tribunals Service experienced issues with the cloud video platform used for secure video links to court hearings in England and Wales [4]. The Guardian reported that the failure delayed several hearings and affected defendants, lawyers and other parties who use the system to attend proceedings remotely [2]. ITV News said journalists using the court video platform reported it had started working again shortly after Google’s noon update [4].

Other services appeared in outage reports as the morning unfolded. Reuters said Downdetector showed reports affecting Lloyds, HSBC, Rightmove, DPD, NHS England and Royal Mail, while noting that Downdetector aggregates user reports and actual impact can vary [3]. ITV also cited reports involving Royal Mail, DPD and Rightmove, and said the budgeting app Emma told users it was down because of the Google Cloud outage in London [4]. Those reports do not prove the same technical dependency for every affected name, but they show how a single infrastructure event can create a crowded field of simultaneous service complaints.

The bigger risk: concentration

The outage lands at an awkward moment for regulators. Reuters noted that the UK brought Google Cloud, Microsoft, Amazon Web Services and Oracle under direct oversight as critical third-party suppliers to the financial sector in July 2026, citing the risk that a major provider’s outage could disrupt multiple firms at once [3]. Reuters also reported that Google Cloud was among technology companies designated by EU regulators as critical third-party computing providers for the bloc’s finance industry [3].

That is the core lesson of the day: cloud can be resilient, but resilience is not automatic. A hyperscale provider may offer multiple zones, regions and failover tools, yet customers still have to design, pay for and rehearse architectures that survive the loss of a zone or region. Google’s own workaround during the incident was conditional: if feasible, impacted customers could try to migrate workloads to another available location [1]. “If feasible” is doing a lot of work. In a real outage, feasibility depends on decisions made months or years earlier.

For financial firms and public bodies, the question after this incident is not simply whether Google fixed the power event. It is whether critical services were too tightly coupled to one cloud region, one zone or one operational path. Redundancy that exists only on a vendor slide will not help a defendant waiting for a video hearing, a customer at a checkout or a small business trying to move money.

What happens next

Google says it will publish an incident analysis after completing its internal investigation [1]. Regulators and affected firms will want a timeline, a root-cause account, a description of why the event in europe-west2-b caused wider symptoms, and evidence that similar power events will be contained more effectively next time. Customers, meanwhile, will mostly remember whether they could pay, log in or attend court.

The cloud did not disappear; it stumbled. But the stumble was enough to show that the digital economy’s plumbing is now shared by banks, courts, delivery firms, property platforms and personal finance apps. Even the cloud, it turns out, sometimes needs an uninterruptible power-up mushroom.

Sources from the last 72 hours

1. [1]Incident details | Google Cloud Service HealthOct 20, 2026, 4:03 PM
2. [2]Google cloud failure hits millions as UK courts and banks knocked offlineOct 20, 2026, 11:45 AM
3. [3]Banks and courts affected by disruption as Google apologises for cloud outageOct 20, 2026, 1:40 PM
4. [4]Power outage hits Google Cloud services disrupting UK banksOct 20, 2026, 2:56 PM
5. [5]Google Cloud outage in UK eases after impacting digital banks, public sector websitesOct 20, 2026, 3:57 PM

AI-generated article based on recent web research, then preserved as a dated editorial snapshot.