This tracker logs 11 major cloud and internet-infrastructure outages in 2026 (latest: October 27, 2026), each taken from the operator's own status page or incident review. Across the 6 incidents whose operator published a root-cause time, the mean Diagnostic Lag Ratio™ is 0.27 (median 0.10): the share of the outage spent before the cause was identified.
Incidents (reverse chronological)
| Start (UTC) | Operator | Incident | Duration | DLR™ | Tier | Operator document |
|---|---|---|---|---|---|---|
| Cloudflare | ERR_TIMED_OUT and ERR_EMPTY_RESPONSE being observed by Cloudflare CustomersFrom the operator's status page; no incident review read yet | 2 h 44 min | —"Identified" after 0 min | — | Cloudflare status page incident (major)cloudflarestatus.com · dk2obcyw2aew | |
| Cloudflare | Network connectivity issues affecting services in North AmericaNetworkFrom the operator's status page; no incident review read yet | 2 h 56 min | —"Identified" after 21 min | — | Cloudflare status page incident (major)cloudflarestatus.com · sufgcbe8rrwt | |
| Microsoft AzureMulti-region | Network Connectivity – Gateway service degradation in multiple regionsFrom the operator's status page; no incident review read yet | 5 h 45 min | — | 1 | Post Incident Review (PIR)azure.status.microsoft · 7Q30-010 | |
| Google Cloudglobal, us-west1 | We are investigating an issue where customers may experience timeouts, service degradations, errors, and elevated latencies across multiple products in the us-west1 region.AlloyDB for PostgreSQL / Apigee Edge Public Cloud / Apigee X / Artifact Registry / BigQuery Data Transfer Service / Cloud BuildFrom the operator's status page; no incident review read yet | 3 h 40 min | — | 1 | Google Cloud Service Health incidentstatus.cloud.google.com · utF3FMFdQfwBzJcGG6vf | |
| Microsoft AzureWest US | Break-fix maintenance withdrew IP routes from all optical devices egressing a datacenterMulti-service regional network | 4 h 57 min | 0.52 | 2 | Post Incident Review (PIR) - Network connectivity - Issues accessing resources in West USazure.status.microsoft · ZJV6-SGG | |
| Google Cloudeurope-west4-a | Utility voltage transient and DRUPS failure caused cooling loss and thermal shutdownGCVE / Bare Metal Solution / NetApp Volumes | 14 h 55 min | 0.00 | 3 | Incident Report - GCVE BMS and GCNV service outage in europe-west4-astatus.cloud.google.com · europe-west4-a cooling failure | |
| Microsoft AzureMulti-region (Australia East / Sweden Central / Europe) | Upstream change reclassified capacity failures as retriable and triggered retry amplificationAzure OpenAI Service | 7 h 26 min | 0.90 | 1 | Post Incident Review (PIR) - Azure OpenAI Service - Latency and intermittent failures in multiple regionsazure.status.microsoft · LYXT-C1Z | |
| Microsoft AzureWest US 2 | Thunderstorm voltage sag placed cooling systems in protective lockout across two availability zonesCompute / Storage / Networking / Telemetry | 22 h 06 min | 0.00 | 2 | Post Incident Review (PIR) - Multiple services - Power/cooling issues in West US 2azure.status.microsoft · GHRP-84G | |
| DENICGlobal | Invalid DNSSEC signatures published during a key rollover made .de domains bogus to validating resolvers.de TLD authoritative DNSFigures pending: operator document not yet read end to end | 3 h 18 min | — | 1 | Final Report: DNS Outage of 5 May 2026blog.denic.de · DENIC final report 5 May 2026 | |
| CloudflareGlobal | Cleanup sub-task sent an empty pending_delete parameter and deleted all BYOIP prefixesBYOIP prefix advertisement / Magic Transit / Spectrum / CDNThe operator's document contains figures that do not reconcile | 5 h 07 min | 0.16 | 1 | Cloudflare outage on February 20 2026blog.cloudflare.com · Cloudflare outage on February 20 2026 | |
| Microsoft AzureMulti-region | Control plane failures affecting resource creation and management for managed identitiesVirtual Machines / managed identities and dependent servicesFigures pending: operator document not yet read end to end | 6 h 27 min | — | 1 | Post Incident Review (PIR) - Virtual Machines Managed identities for Azure resources and dependent servicesazure.status.microsoft · FNJ8-VQZ | |
| CloudflareGlobal | Bug in generation logic for a Bot Management feature file degraded core servicesBot Management and dependent edge servicesFigures pending: operator document not yet read end to end | — | — | 1 | Cloudflare outage on November 18 2025blog.cloudflare.com · Cloudflare outage on November 18 2025 | |
| Amazon Web Servicesus-east-1 | Race condition in DynamoDB DNS automation emptied the regional endpoint recordDynamoDB / EC2 / Network Load Balancer and dependents | 14 h 32 min | 0.04 | 1 | Summary of the Amazon DynamoDB Service Disruption in the Northern Virginia (US-EAST-1) Regionaws.amazon.com |
What counts. A major outage of AWS, Microsoft Azure, Google Cloud, Cloudflare or a top-level-domain registry, documented by the operator itself: a post-incident review, a post-event summary, an engineering postmortem, or a status-page incident the operator rated major (Google Cloud "high" severity; Cloudflare "major" or "critical" impact). Third-party monitors and outage-report counts are excluded.
Diagnostic Lag Ratio™ (DLR, v1.0). DLR = minutes from customer impact start to the operator's published root-cause identification ÷ total customer impact minutes. Both terms come from the operator's own timeline; nothing is estimated. It is computed only for incidents whose review was read end to end by Axis Intelligence Research. For status-page incidents, the time to the operator's first "identified" update is shown separately and is not a DLR. Tier: 1 = multi-region or global, 2 = one region, 3 = one zone or service; status-page rows carry a tier only when the operator lists more than one region.
Sources. Operator documents only: Google Cloud Service Health, Cloudflare status page and blog, Azure Status History (post-incident reviews), AWS Post-Event Summaries, registry reports. Times in UTC. Operator pages are named, not linked, under our outbound-link policy.