Google Cloud London Power Event: Recovery Underway in europe-west2-b

Google says most workloads in europe-west2 were recovering after a power event caused an outage in europe-west2-b. The incident remained an active service disruption.

Reader ratingRate this article

Loading reader rating…

Privacy-conscious voting uses a signed random browser identifier and a short-lived network hash; the rating tables contain no raw IP-address column.

Reporting cutoff: October 20, 2026 at 9:00 AM EDT / 2026-10-20 13:00:42 UTC.

Google Cloud says a power event caused an outage in the europe-west2-b zone of its London region.[1] The power event itself has been mitigated, but service recovery was still underway at our cutoff.[1]

ACTIVE SERVICE DISRUPTION

Current status at the reporting cutoff

Google said most workloads in europe-west2 were recovering as of 5:10 a.m. PDT (8:10 a.m. EDT / 12:10 UTC).[1] The incident remained active, and Google said it would provide another update by 7:00 a.m. PDT (10:00 a.m. EDT / 14:00 UTC).[1]

The incident remained classified as service disruption; Google still described services as unavailable in europe-west2-b, with possible elevated errors and intermittent network issues elsewhere in europe-west2.[1]

What happened

Google’s incident record says the disruption began at 6:08 UTC on October 20 (11:08 p.m. PDT on October 19).[1] The public incident object was created at 8:35:37 UTC, more than two hours after the stated start.[1]

Early updates described elevated errors and intermittent network failures across the London europe-west2 region.[1] At 11:12:42 UTC, Google attributed the outage to a power event in europe-west2-b; at 11:58:15 UTC, it said services were recovering but new-workload creation could still fail.[1]

  • 6:08 UTC: Google’s retrospective incident-start time.[1]
  • 8:35:37 UTC: The public incident object was created.[1]
  • 11:12:42 UTC: Google first identified a power event in europe-west2-b and said the power event was mitigated.[1]
  • 11:58:15 UTC: Google reported service recovery, prioritized existing workloads and warned that new workloads might not yet be creatable.[1]
  • 12:58:49 UTC: Google said most workloads in europe-west2 were recovering as of 12:10 UTC and scheduled its next update for 14:00 UTC.[1]

Scope and independent corroboration

Google attached 38 cloud products to the incident, including Compute Engine, Google Kubernetes Engine, Cloud SQL, Cloud Storage, BigQuery and Cloud Run.[1] That product list describes the incident’s declared scope, not proof that every product or every customer failed in the same way.[1]

Reuters separately reported disruptions in Google Cloud’s London region and cited a person familiar with the matter who linked Revolut’s outage to the Google Cloud disruption.[2] Reuters did not independently verify Google’s power-event explanation in the report we reviewed.[2]

What Google says is fixed

The power event itself was mitigated, and recovery of services had begun.[1]

What was not yet fixed

The incident remained active; Google still described services as unavailable in europe-west2-b, with possible regional errors and intermittent network issues.[1]

Practical guidance for operators

Google’s workaround: if feasible, affected customers can try moving an impacted workload to another available zone or region.[1]

Our practical interpretation: use that option only through an already tested and authorized failover plan.[1] A status-page recovery statement should not replace checks of your own workload, queued jobs and transactions before you declare normal operation.[1]

What remains unknown

The canonical Google incident record had not explained what triggered the power event, identified the exact failed infrastructure, quantified affected customers, confirmed data loss, or published a completed root-cause report by our cutoff.[1]

Nothing in the canonical incident record we reviewed alleged a cyberattack or sabotage.[1]

Local relevance

Google’s canonical location scope was London europe-west2; it did not list a Florida location.[1] Our analysis: the practical Florida connection can be indirect if a local business’s own systems—or a vendor it depends on—run in Google Cloud’s London region.[1]

Sources

  1. Google Cloud Service Health — incident Xs2XTHzAVC9YnafnPyzj. Live incident object; accessed 2026-10-20 13:00:42 UTC.
  2. Reuters — “Google Cloud services experience outage in Europe”. Published October 20, 2026; accessible syndication checked at 2026-10-20 12:19:22 UTC.

Image disclosure: The featured image is an original AI-generated editorial illustration. It is not a Google interface, a photograph of the affected facility or documentary evidence of the incident.