Google Cloud London Outage: 'Power Event' Blamed

Google Cloud says multiple services in its London region, europe-west2, have been disrupted since early on Tuesday 20 October, and has blamed a “power event” that caused an outage in one zone, europe-west2-b. The problem began at 07:08 BST, according to Google’s Cloud Service Health incident page, and the company has listed 38 affected products, from Compute Engine and Kubernetes Engine to BigQuery and Cloud Storage.

For businesses that run their systems in London, this is a regional disruption rather than a single-product glitch. In an update at 12:12 BST, Google said: “Due to a power event we had suffered an outage within europe-west2-b zone.” It added that the power event “is mitigated” and that partial mitigation of infrastructure services would be seen within the next 30 minutes. The incident was still open, with the next update promised by 13:00 BST.

In brief

Google Cloud says multiple services in europe-west2, its London region, have seen elevated errors and intermittent network failures since 07:08 BST.
Its status page lists 38 affected products, including Compute Engine, Cloud Run, Cloud SQL, BigQuery and Cloud Storage.
At 12:12 BST Google blamed a “power event” for an outage in the europe-west2-b zone, saying it is mitigated. The incident is not resolved, and the next update is due by 13:00 BST.

What Google has said so far

Google’s summary is brief: “Multiple services are experiencing issues in the europe-west2 region.” The description adds: “We are experiencing an issue with multiple services in europe-west2 beginning on Monday, 2026-10-19 23:08 PDT.” That is 07:08 BST on Tuesday in the UK.

Google’s latest update, posted at 12:12 BST, gives a cause for the first time. “Due to a power event we had suffered an outage within europe-west2-b zone. The power event is mitigated,” it says. “Partial mitigation to infrastructure services will be observed within the next 30 minutes and the remaining services will be mitigated there after.” Earlier updates, including one at 12:00 BST, had said only: “Our engineering team continues to investigate the issue.”

The description of the symptoms has also changed. Google now says: “Google Cloud services are unavailable in europe-west2-b zone; some services in europe-west2 region may experience elevated errors and intermittent network issues.” Earlier updates said: “Multiple Google Cloud services are experiencing elevated errors and intermittent network failures in europe-west2 region.” Google promises another update by 05:00 PDT, which is 13:00 BST, and adds: “We apologize to all who are affected by the disruption.”

The incident appeared on the status page at 09:35 BST, about two and a half hours after the start time Google gives. That first notice said there was no workaround “for now”. By 10:24 BST Google had added one, and noted that “Looker Studio is not impacted by this outage.” That note appeared in the 10:24 and 12:00 BST updates but was dropped from the 12:12 BST update.

Time (BST, 20 Oct)	What Google’s status page shows
07:08	Issue begins in europe-west2
09:35	First update; no workaround yet
10:24	Update; workaround added
12:00	Update; still under investigation
12:12	Cause given: “power event” in europe-west2-b zone; power event mitigated
By 13:00	Next update promised

Source: Google Cloud Service Health, incident Xs2XTHzAVC9YnafnPyzj, checked at 12:15 BST.

Dozens of products, one region

The list of affected services shows how much of Google’s platform sits in a single region. Alongside core computing products such as Compute Engine, Google Kubernetes Engine, Cloud Run and App Engine, the status page names databases including Cloud SQL, Cloud Spanner, AlloyDB for PostgreSQL, Firestore and Bigtable. It also lists BigQuery, Cloud Storage, Persistent Disk, Pub/Sub, Dataflow and several Memorystore products.

Networking is affected too. Cloud Load Balancing, Cloud NAT, Virtual Private Cloud, Private Service Connect and Network Connectivity Center all appear on the list, as do Apigee and Managed Service for Apache Kafka. Google’s status page records only one affected location: London (europe-west2).

Google has not said which customers are affected, or how many. Its advice is to move work elsewhere where possible. “If feasible, the impacted customer may try to migrate their workload impacted to another available location (zone or region),” the status page says. Google does not say how many customers can do that, and the advice is qualified by “if feasible”.

Cloud outages have very different causes. TSN has previously reported on drone strikes that left Yandex Cloud in “limited mode”, a very different incident. In Google’s case, the company has so far said only that a “power event” caused an outage in the europe-west2-b zone.

Our read

This is TSN’s analysis. A disruption that touches 38 products in one region is a reminder that “in the cloud” still means a physical place, and that resilience depends on how many places a business uses. The key questions now are when Google restores service, and whether its eventual incident report says more about the “power event” than the status page has so far.