Google Cloud outage, 20 October 2026: what happened and why apps went down

MacWindowsiPhoneAndroidUpdated

On 20 October 2026, a data centre in Google Cloud's London region lost power and one of the region's zones, europe-west2-b, went offline. Google's status page called it a "power event". Apps that run on Google Cloud were disrupted that morning, and Revolut and the Emma money app both said the Google Cloud outage was behind their problems.

Quick answer

On 20 October 2026 the facility hosting Google Cloud zone europe-west2-b in London lost its mains power at 06:53 UK time. The backup generators didn't start, and servers went down at 07:00 when the batteries ran out. Google's final report, published on 27 October, says the primary impact was mitigated after 6 hours 8 minutes, at 13:08.

What happened on 20 October 2026?

Google published updates on its Cloud Service Health page during the day, a preliminary incident report on 22 October and a final incident report on 27 October. All times below are UK time (BST), taken from those pages.

  • 06:53. The final report says the electricity supplier lost the facility's main feed because of an underground cable fault on the utility network outside the data centre. The facility was already running on a single feed, because the second had been switched off for repairs after a generator load test on 14 October found problems.
  • 07:00. The generators never got the signal to start. Google's report puts this down to a misconfiguration in the controllers that manage them. The batteries carried the load until they ran out, and then servers, storage and network hardware in europe-west2-b lost power.
  • 07:04. Google's monitoring alerted its engineers, according to the final report.
  • 07:54 to 08:20. Staff switched the facility to a standby supply, and power reached the data halls again at 08:20.
  • 09:35. Google's first public update said multiple services were having issues in the europe-west2 region.
  • 12:12. Google's update said the outage within europe-west2-b was due to a power event, and that the power event was mitigated.
  • 13:08. The final report says 99.5% of Compute Engine virtual machines were working again and new ones could be created.
  • 15:03. Google's last update said the problem was mitigated for most customers, with a very small number still affected.

Recovery took longer than the power restoration because, as Google explains, storage devices that had shut down suddenly needed time to recover before services could safely restart.

Why did one zone cause so much disruption?

A Google Cloud region is split into zones. Google's report names three in europe-west2, a, b and c, and says the power outage was isolated to the facility hosting zone b. Anything that ran only in zone b became unreachable.

Some customers in the other two zones were affected too. Google gives two reasons: a sudden surge of demand as work moved over to zones a and c, and control systems in those zones that still depended on parts of zone b. The final report lists dozens of affected products, including Compute Engine, Cloud SQL, BigQuery, Google Kubernetes Engine, Cloud Storage and Pub/Sub. It says Google found no impact outside the europe-west2 region.

Which banks and apps were affected?

This guide names a company only where its own statement links its problems to the Google Cloud outage.

  • Revolut. On the morning of 20 October, Revolut posted on X that it had a service disruption affecting multiple services, and said: "We can confirm that all our customers' money and data are safe." At 18:39 it posted again, saying customers may have had temporary disruption on the app because of a recent Google Cloud outage, and that services were restored.
  • Emma. At 09:29, the Emma money app posted on X that it was down because of a Google Cloud outage affecting services in London since 7am, and called it "a technical outage, not a security issue". At 13:42 it said the app was running again and that data and money were safe throughout.

We haven't named any other bank or service, because we found no statement from the company itself, or from an official body, linking its problems to Google. If your bank had problems on 20 October, its own service status page or official account is the place to check. Our guide on what to do when a banking app is down explains how.

Was it a cyber attack?

Google described the cause as a power event, and its final report traces it to the cable fault and the generator controls. No source we found describes it as a cyber attack. Our guide was the Google outage a cyber attack? covers this, and how to tell an outage from an attack. For attacks on power networks that agencies have documented, see can hackers cause a blackout?

Has Google Cloud had outages like this before?

Yes, with different causes. On 12 June 2025, a Google Cloud outage hit multiple Google Cloud products. Google's incident report says it was caused by an invalid automated quota update to its API management system, which was distributed globally. That report doesn't describe an attack either.

Outages have many causes, from power and software changes to planned work. For the planned kind, see website down for maintenance.

Who oversees the cloud providers that banks use?

On 10 July 2026, the FCA announced that the Treasury had designated Google Cloud EMEA Limited, along with Amazon Web Services, Microsoft and Oracle, as critical third parties to the UK financial sector. The FCA said the Bank of England, the Prudential Regulation Authority and the FCA would start overseeing them on 13 July 2026.

The FCA says this regime adds to, and doesn't replace, the rules for banks themselves. Banks remain responsible for managing their own third-party arrangements, including contingency planning. Under the FCA's operational resilience rules, firms had until 31 March 2025 to make sure they could keep their important business services within set limits for disruption.

What has Google said it will change?

The final report of 27 October lists repairs to the utility cable and a return to two power feeds, generator tests, and a fix to the controller configuration, which Google says it completed on 26 October. It also says Google will check controller settings at its sites worldwide with a similar design, with an estimated completion date of 15 November, and pause planned maintenance at those sites.

Guides in this series

Related: security update scam emails, and what to do in a power cut at home.

More questions

What is europe-west2?

It's Google Cloud's London region. Google's status page lists the affected location for the 20 October 2026 incident as London (europe-west2), and its report names three zones in the region, a, b and c. The power loss hit the facility hosting zone europe-west2-b.

How long did the Google Cloud outage last?

Google's final report says the impact ran from 07:00 to 13:08 UK time on 20 October 2026, a total of 6 hours 8 minutes. A small number of customers were still affected after that, and Google posted its closing update at 15:03.

Has Google published a report on the outage?

Yes. Google posted a preliminary incident report on 22 October 2026 and a final incident report on 27 October 2026, both on its Cloud Service Health page. The final report gives the root cause and a list of repairs and checks.

Was customers' money safe?

Revolut said on 20 October 2026 that all its customers' money and data were safe, and Emma said data and money were safe throughout. For how UK deposit protection works, see is my money safe when my bank's app is down?

Did the outage affect Google Cloud outside London?

Google's final report says it did not identify any impact of the outage outside the europe-west2 region.

Sources