Google Cloud europe-west2-b outage: banks and mail hit in London

A power event in one availability zone of Google Cloud's London region disrupted services at some of Britain's best-known organisations on 20 October 2026, and it puts a spotlight on how workloads are designed to cope with a single-zone failure.

The incident was in europe-west2-b, one of three local availability zones in the London region. Those zones are intended to help protect against hardware or regional failures. A fault in one zone reaching so many well-known names is a reminder that resilience depends on how each customer builds on top of the platform.

What changed

The disruption began at 0608 UTC. Google Cloud's status page said multiple services were "experiencing elevated errors and intermittent network failures in europe-west2 region". Google confirmed that BigQuery Data Transfer Service, Cloud Filestore, Cloud Run, Google App Engine, Google Cloud NetApp Volumes, Google Kubernetes Engine and Managed Service for Apache Kafka were all impacted.

Early on there was no estimated time of recovery, and Google's answer on workarounds was blunt: "none for now". It later updated its status page to say Studio was not impacted by the outage and told customers looking for a workaround: "If feasible, the impacted customer may try to migrate their workload impacted to another available location." The company planned a next update at around 0930 UTC, and said its engineering team was continuing to investigate and apologised to everyone affected.

According to ABC News, the Google Cloud website attributed the failure to "a power event" and said the incident was isolated to a data centre in London. The Google Cloud website said: "Due to a power event we had suffered an outage within europe-west2-b zone." Google said the incident began shortly after 6am UK time, with disruption reported well into the afternoon. It later said the server had been restored and that the issue had been mitigated for most customers, with a very small number still affected while its engineering team worked on it. Google said it would carry out an internal investigation and publish an analysis once that has concluded.

Impact on businesses

According to Down Detector, the outage-tracking site, customers caught up included Lloyds Bank, HSBC, Royal Mail, Rightmove, DPD, BT, Screwfix and Revolut. Halifax wrote on X that customers were "having issues with our app and online". British media reports also said that Microsoft was among the companies experiencing issues, alongside Royal Mail and DPD.

Revolut said: "We are currently experiencing a service disruption affecting multiple services. We can confirm that all our customers' money and data are safe." It added that it had been working to fix the issue since it was identified in the morning, and apologised to anyone affected. Reassurance about money and data is a different matter from availability, and for a bank a login screen that will not load is still an operational incident.

The pattern also reaches beyond finance. Techies on Reddit said the interruption was having a cascading effect on multiple suppliers. One commenter summed up the engineering reality: "oh man the classic one zone takes a nap and your non-HA service is in it." HA, or high availability, means designing a service to keep running when a component fails, typically by running copies in more than one zone. Another Reddit user mused: "Loads of stuff seems to be down, DNS somewhere."

Impact on consumers and users reported struggling to log into their mobile banking, and customers of major UK banks reported issues accessing their accounts. British media reported that the outage affected everything from e-bike rental services to budgeting apps.

The public sector appears to have felt it too. The UK Courts and Tribunals Service said on X that it was aware of an issue affecting Cloud Video Platform (CVP), used for some remote hearings. Court live streams, including that of the ongoing trial of comedian Russell Brand, were also reportedly down, and the outage appears to have disrupted live streams of court cases across Britain.

For ordinary users the practical advice is modest: Revolut said money and data were safe, but anyone with a time-sensitive payment or hearing is sensible to have a fallback, such as a phone line or branch, for when an app is down.

Impact on the wider industry

Google Cloud has said it will publish an analysis of the event once its internal investigation has concluded. That is where the detail on how a power event in one facility led to "elevated errors and intermittent network issues" in parts of Europe should emerge.

The outage also illustrates a standing concern in cloud computing: when many firms rely on one provider or one location, a single fault can reach a wide range of services at once. Banks, a postal service and the courts were all named in the reporting on this incident.

The architectural lesson applies to any firm running workloads in a single zone.

What comes next

Based on what Google has said, the sequence looks like this:

• Immediately: Google said the server had been restored, with a very small number of customers possibly still affected while engineers worked on it.
• In the coming days and weeks: Google's internal investigation concludes and it publishes its analysis of the event.
• After that: affected firms may review their own architecture and resilience arrangements in light of what the analysis shows.

What IT teams should check

The outage gives practitioners a ready-made checklist:

• Confirm which workloads run in a single zone, including managed services such as Cloud Run, Google Kubernetes Engine and Managed Service for Apache Kafka.
• Test failover between zones and, where the business case justifies it, between regions. A plan never exercised is a hypothesis.
• Be ready to migrate a workload to another location, since that was the workaround Google pointed customers towards.
• Map hidden dependencies, such as DNS and logins, that sit in one zone even when the main application does not. One Reddit user mused about DNS "somewhere".
• Agree recovery time and recovery point targets with the business, and check they match reality.

Key takeaways

• One zone, europe-west2-b, failing was enough to disrupt banks, a postal service and court live streams.
• Google's suggested workaround was to migrate affected workloads to another available location.
• Reassurances that money and data are safe do not equal availability.
• Much of the fix sits with customers: multi-zone design and tested failover.

Verdict

On balance this is a negative for customers and a useful, if costly, prompt for IT teams. The hours of disruption left some people struggling to log into their banking, and court live streams were reportedly affected. The benefit goes to organisations that treat it as a free resilience test and act before the next power event, not after. Watch for Google's post-incident analysis, and for what it says about how one zone's trouble spread as far as it did.