Cloud outages and IT resilience: latest coverage
Everything we have published on cloud and IT outages and how UK organisations plan for them, in one place: the incidents, the outage and cost data, and the disaster recovery explainers, with a short dated summary of where things stand. This page adds no new reporting. For the full incident list, see our cloud outages tracker.
Newest piece 23 Oct 2026 · summary checked 28 Oct 2026 · coverage since 11 Feb 2026
Where things stand, as of 28 Oct 2026
- The most recent major UK outage we covered was Google Cloud London on 20 October. After a power failure in europe-west2-b, one of three zones in Google’s only UK region, reportedly when backup generators failed to start, banking apps, the courts’ video platform, couriers and NHS England all showed disruption. With no second UK region, there was nowhere in the UK to fail over to.
- Power and cooling keep taking hyperscalers down. Google Cloud services in europe-west4, in the Netherlands, were disrupted for 14 hours 55 minutes in July after a grid fault knocked out power and cooling, and Uptime Institute data puts power as the leading cause of impactful data centre outages.
- Process and software failures take platforms down too. A mismarked maintenance request cut Azure West US for almost five hours on 23 July, and GitHub’s 17 August outage ran for nearly eight hours after its load balancers saturated, with a VS Code retry bug slowing recovery.
- UK exposure is concentrated. The Cyber Monitoring Centre estimates that one 24-hour outage of a major AWS or Azure region could cost UK businesses £650m to £1bn in direct revenue, and that over 80% of the FTSE 100 are cloud-dependent for critical functions. The US holds 15 of the world’s top 20 hyperscale markets.
- Outages are rarer per site but still expensive. Uptime Institute reports five straight years of falling outage frequency per site, yet in its 2025 survey 57% of operators said their most recent major outage cost more than $100,000, and one in five more than $1 million.
- Physical links are now a policy question too: the government has proposed up to 14 years in prison for deliberately damaging subsea cables.
Still unclear
- How many customers the Google Cloud London outage hit. Google described the impact as a “very small number” of customers, while banks, the courts and couriers all showed disruption.
- What a single outage costs you. Published estimates use different methods: the Cyber Monitoring Centre’s £650m–£1bn is for a whole region over 24 hours, while our tracker’s illustrative figure for a 500-person firm hit for five hours is £85,000–£160,000.
Read these first
Scroll the chart sideways to see all of it →
Counts are every published piece linked from this page. Each section does a different job, so read across them rather than down one.
Everything we have published on cloud outages and IT resilience
Scroll the chart sideways to see all of it →
Hover or tap a month for its breakdown. Every piece counted is listed below.
Scroll the chart sideways to see all of it →
Durations as reported in the linked pieces. Google Cloud London on 20 October is not shown: Google reported most workloads recovered that afternoon, but no time for full recovery has been published.
October 2026 2
- 23 Oct 2026UK Subsea Cable Sabotage Law: 14 Years Jail in 2026News
- 21 Oct 2026Google Cloud UK Outage 2026: Single-Region Risk ExposedNews
September 2026 5
- 17 Sept 2026Hot, Warm and Cold DR Sites Explained (2026)Learn
- 16 Sept 2026DR Testing Types: Tabletop to Full Failover ExplainedLearn
- 15 Sept 2026Business Continuity vs Disaster Recovery ExplainedLearn
- 12 Sept 2026Major Cloud Outages of 2026: Incident Tracker & UK ImpactResearch
- 11 Sept 2026Data Centre Outage Statistics 2026: UK & Global TrendsResearch
August 2026 6
- 20 Aug 2026Hyperscale Datacentre Locations 2026: UK Resilience RiskNews
- 17 Aug 2026GitHub Outage 2026: What It Means for UK Backup PlansNews
- 10 Aug 2026DRaaS vs Self-Managed DR: The 2026 UK Buyer's CallInsights
- 5 Aug 2026DRaaS Price 2026 UK: Benchmarking Cost, RTO and FailoverResearch
- 5 Aug 2026DRaaS Explained 2026: Cloud Failover as a DR Site AlternativeLearn
- 4 Aug 2026RTO vs RPO Explained (2026): Downtime and Data Loss TargetsLearn
July 2026 3
- 24 Jul 2026Azure Outage 2026: What UK DR Plans Must Fix NowNews
- 21 Jul 2026Google Cloud Outage 2026: A UK DR Wake-Up CallNews
- 9 Jul 2026UK Cloud Outage Resilience 2026: FTSE 100 DR GapsNews
March 2026 1
- 27 Mar 2026The real cost of IT downtime for a small business (2026)Learn
February 2026 2
- 19 Feb 2026Choosing a UK disaster recovery provider: the 9 questions that matterInsights
- 11 Feb 2026Disaster recovery explained: RTO vs RPO for non-techies (2026)Learn
Acting on it
Common questions
What caused the Google Cloud outage in London?
A power failure in europe-west2-b, one of the three zones in Google Cloud’s London region, starting just after 07:05 BST on 20 October 2026. Engineers tracking the incident reported that backup generators failed to engage; banking apps, the courts’ video platform, couriers and NHS England all showed disruption, and Google reported most workloads recovered by 14:48. London is Google’s only UK region, so there was no second UK region to fail over to.
How much does a cloud outage cost UK businesses?
The Cyber Monitoring Centre puts a 24-hour outage of a major AWS or Azure region at £650m to £1bn in UK direct revenue. Across data centre operators, 57% told Uptime Institute’s 2025 survey that their most recent major outage cost more than $100,000. Your own figure depends on what stops; our downtime cost calculator works it out per hour.
Is a second cloud region enough?
Usually it is the minimum, and only if the second region is genuinely separate and the failover is tested regularly. But shared services can still cross regions: a February 2026 Azure Managed Identity incident hit both East and West US. Test the failover, and check which identity and control-plane services both regions depend on.
Other running coverage
Topic pages summarise Servnet’s own published coverage for information only; they are not security, legal or financial advice. Facts and sources are in the linked pieces, which follow our editorial standards. Spotted something out of date? Tell us at webmaster@servnetuk.com.
Talk to a UK specialist
Get expert advice or a no-obligation quote — servers, storage, networking, maintenance, finance and cloud. We reply the same working day.