141件の追跡済みインシデント · 2015–2026
インシデントトラッカー
データセンターとクラウド全体の主要な停止と障害 — 事実に基づき、日付付き、情報源付き。重大な問題が発生したときにメール通知を受け取りますか？アラートを設定。
2026-10-20
Google Cloud power event takes down europe-west2-b zone in London
Google Cloud · europe-west2-b (London), United Kingdom
A power event caused an outage in the europe-west2-b zone starting around 06:08 UTC on October 20, 2026. Affected Google Cloud services included Kubernetes Engine, Cloud Run, App Engine, Cloud Filestore and BigQuery Data Transfer Service, and customers such as Lloyds Bank, HSBC, Royal Mail, BT and Revolut reported disruption. By 14:03 UTC Google said service had resumed for most customers, with a very small number still affected. Google has not published a root cause beyond 'power event'; a fuller analysis was promised.
The Register — Google Cloud says a power event caused today's outage ↗
2026-10-08
Google Cloud Storage write errors in us-central1 last about 12.5 hours
Google Cloud · us-central1 (Iowa), United States
From 11:00 UTC to 23:30 UTC on October 8, 2026, Cloud Storage upload and write operations in us-central1 saw elevated latency and error rates, and a smaller share of reads was slow or failing. Google gave a preliminary cause: a traffic sorting optimization meant to ease localized capacity shortfalls exhausted resources on specific backend storage nodes, with a separate database service issue adding metadata errors for part of the window. A full root cause analysis was still pending.
2026-10-07
Ransomware forces IDC Frontier to isolate its East Japan Region 1 data center
IDC Frontier (IDCF Cloud) · Shirakawa, Fukushima, Japan
Unauthorized access began around 03:40 JST on October 7, 2026. IDC Frontier confirmed a third-party ransomware attack, then isolated and halted the East Japan Region 1 data center. Contracts tied to that region cover 495 companies and local governments, and some customers were told data recovery could be difficult. The intrusion method and whether data was stolen had not been disclosed.
BleepingComputer: Ransomware attack disrupts Japan's IDCF Cloud used by govt clients ↗
2026-09-30
Azure gateway outage cuts ExpressRoute, VPN and Firewall connectivity across 18 regions
Microsoft Azure · Multi-region (18 Azure regions, incl. West US, North Europe, Southeast Asia, Japan West, South Africa North)
From 20:30 UTC on September 30 to about 02:15 UTC on October 1, 2026 (roughly 5 h 45 min), Azure ExpressRoute Gateway, VPN Gateway, Firewall, Application Gateway/WAF and Azure VMware Solution customers saw degraded or lost network connectivity. Microsoft's status history said the timing correlated with infrastructure operating system servicing activity, which was paused. No final post-incident review was available at the time of writing.
2026-09-30
West Java halts construction of BDx 640 MW CGK4 AI campus eight days after groundbreaking
BDx Data Centers · Jatiluhur, Purwakarta, West Java, Indonesia
On September 30, 2026, West Java Governor Dedi Mulyadi ordered work stopped at BDx's CGK4 campus, which had broken ground on September 22. The governor said two permits had not been issued — the environmental impact assessment (AMDAL) and the building approval (PBG) — and raised public concerns about water use from the Jatiluhur reservoir, noise and wastewater. CGK4 is planned as a 640 MW, six-building campus with 845 MVA of grid supply from PLN; Building 1 (120 MW) was targeted for early 2027. Reporting did not include a BDx statement, and the length of the stoppage is not known.
2026-09-29
Fire: STT Seoul 1 (Gasan-dong)
ST Telemedia Global Data Centres · Seoul, South Korea
A fire believed to have started in a transformer in a UPS room was reported at 11:55 and extinguished by 15:13. STT GDC said an electrical component failed while new equipment was being tested in an environment separate from customer systems. 105 people were evacuated; the operator said customer systems and data center operations were not affected.
DatacenterDynamics — Fire at data center in Seoul, South Korea, sees 100 people evacuated ↗
2026-09-21
Subsea cable faults between Australia and Singapore degrade Asia-Pacific capacity; Cloudflare reports congestion
Vocus (ASC); SUBCO consortium (Indigo West/Central); Cloudflare · Perth–Singapore corridor, Indonesian waters off Anyer, Java
Indigo West and Indigo Central suffered electrical shunt faults near Perth on August 8–9, 2026, and the Australia-Singapore Cable (operated by Vocus) was severed in Indonesian waters off Anyer on September 4. Per Submarine Networks, the faults removed all direct low-latency Perth–Singapore paths (about 100 Tbps), pushing round-trip times from roughly 60 ms to 300–400 ms as traffic rerouted via Guam, Japan or the transpacific. Cloudflare's status page reported multiple subsea outages causing congestion between its Tokyo and Singapore data centers from September 21 at 02:20 UTC and a regional performance degradation in Asia-Pacific from September 23. Repair of ASC was awaiting the cable ship CS Lodbrog, with delays attributed to Indonesian permitting and security approvals.
Submarine Networks — Perth–Singapore subsea outages deep-dive ↗
2026-09-19
Russian strikes hit Kyiv data centers — BeMobile/UCloud and De Novo facilities targeted
UCloud (BeMobile); De Novo · Kyiv, Ukraine
Russian forces struck at least two commercial data center facilities in Kyiv within a five-day span in mid-September 2026. On September 15, an attack hit the BeMobile facility operated by UCloud; the company did not disclose the extent of physical damage but confirmed some hosted websites, including Ukraine's National Paralympic Committee site, were temporarily knocked offline and expected to be restored within hours. On September 17, a broader wave of missile and drone strikes damaged energy and telecommunications infrastructure across the capital and injured more than 20 people, according to Ukrainian officials. On September 19, Russia's TASS state news agency said strikes had hit De Novo, one of Ukraine's largest commercial data centers — a Tier III-compliant, 360-rack facility that hosts major Ukrainian banks and the country's largest Infrastructure-as-a-Service cloud platform. Russian officials characterized De Novo as supporting Ukrainian military data processing; De Novo describes itself publicly as a sovereign-cloud and AI provider serving mission-critical civilian and commercial systems. Neither company has published a full damage assessment or restoration timeline, and the extent of physical destruction at either site has not been independently verified.
DataCenterDynamics — Russia claims strikes on De Novo data center in Kyiv, Ukraine ↗
2026-09-16
Salesforce global platform outage disrupts Hyperforce instances during Dreamforce
Salesforce · Global — Hyperforce instances across multiple regions, including the US, Japan, India, the UK, France and Germany
On September 16, 2026 — the second day of Salesforce's own Dreamforce conference in San Francisco — a global outage hit hundreds of customer instances from roughly 08:30 to 19:20 UTC (about 10.5 hours). Salesforce said requests were stalling while waiting on a response from an internal legacy login service, which consumed available server resources and limited the platform's capacity to process incoming requests; first-party GovCloud environments were largely unaffected. Customers reported severe delays, intermittent errors and inability to access some services, including support-case creation and scheduled job execution. Salesforce rolled out region-by-region fixes, and some customers needed manual session resets or cache clears after the fix before declaring the incident resolved at 19:20 UTC.
The Register — Salesforce staggers back to feet after global outage ↗
2026-09-16
Microsoft 365 sign-in failures lock users out worldwide
Microsoft (Microsoft 365 / Entra ID authentication) · Global, with concentrated user reports in the US and UK
Starting around 13:22 UTC on September 16, 2026, Microsoft tracked a spike in Microsoft 365 sign-in failures (incident MO1472904) accompanied by 5xx server errors on its authentication layer. Users with an already-active session could often keep working, but anyone forced through a fresh sign-in, token refresh, MFA prompt, portal session renewal or new-device enrollment could fail to authenticate. Microsoft said it was reviewing service telemetry and user reports to isolate the root cause and was checking whether upstream dependencies contributed; some affected customers saw service return as the review continued, but Microsoft had not published a confirmed root cause or final duration at last report. This is a separate, later incident from the larger Microsoft 365/Exchange Online authentication outage of August 31 – September 3, 2026 (IDs EX1464935/MO1465074).
2026-09-15
AWS confirms permanent customer data loss in Bahrain and UAE regions after Iran-linked strikes on facilities
Amazon Web Services (AWS) · Middle East (Bahrain) region and Middle East (UAE) region, mec1-az2 availability zone
In a September 15, 2026 update to its Health Dashboard, AWS said damage from drone/missile strikes earlier in the year on facilities supporting its Middle East (Bahrain) and Middle East (UAE) regions spanned multiple Availability Zones and "exceeded what our regional and multi-AZ services are designed to withstand." AWS concluded it cannot restore access to resources and data hosted exclusively in the Bahrain region, and reached the same conclusion for the mec1-az2 Availability Zone in the UAE region; work continues on remaining UAE zones (mec1-az1, mec1-az3). The permanent loss affects only customers whose sole data copy sat inside the damaged infrastructure — those with cross-region replication or backups elsewhere could rebuild — but AWS has not publicly quantified the number of affected customers or the volume of data lost. The episode is a rare confirmed case of physical destruction overwhelming cloud multi-AZ redundancy design assumptions.
Help Net Security — AWS Middle East outage: permanent data loss in Bahrain, UAE ↗
2026-09-03
xAI Memphis compute center outage cascades into simultaneous ChatGPT, Claude and Grok disruptions
xAI / OpenAI / Anthropic · Memphis, Tennessee, USA (xAI compute center), plus OpenAI and Anthropic serving infrastructure
Around 7:43am PT on September 3, 2026, three unrelated AI chat services — ChatGPT, Claude and Grok — went down within the same roughly 90-minute window, an anomaly given each is normally on a separate failure path. OpenAI told The Register that "a routing error" made ChatGPT and Codex unavailable for some users starting at 7:43am PT. Separately, xAI said an outage at its Memphis, Tennessee compute center that morning took down Grok and also disrupted service for unspecified "compute partners," which xAI/SpaceX identified as including Anthropic; Anthropic's status page confirmed a 3-hour-6-minute partial outage with elevated error rates on Claude.ai, Claude Code, Claude Cowork and the Claude API, restored by 16:16 UTC. Google's Gemini was unaffected throughout, since it runs on separate Google Cloud infrastructure. The episode illustrates how concentrated AI compute supply chains can turn one facility's outage into a multi-vendor incident.
The Register — ChatGPT, Claude, and Grok all had outages at the same time ↗
2026-09-01
Google Cloud: network degradation in us-central1-b takes down 15 products for up to 4h08m
Google Cloud · us-central1-b (Council Bluffs, Iowa)
On September 1, 2026, Google's own Service Health dashboard logged incident J5ia5t9p3g9Q5Wi7r8Ev: multiple products running in the us-central1-b zone experienced network service degradation. Affected services included Compute Engine, Google Kubernetes Engine, Cloud SQL, BigQuery, Cloud Bigtable, Cloud Spanner, Cloud Run, App Engine, AlloyDB for PostgreSQL, Apigee, Cloud Filestore, Cloud Dataflow, Hybrid Connectivity, Looker and Virtual Private Cloud. Impact duration varied by product, from about 14 minutes for VPC to 4 hours 8 minutes for Cloud Run and App Engine. Google did not publish a root-cause explanation alongside the status update.
2026-08-31
Microsoft 365: authentication-configuration fault disrupts Exchange Online, Teams, SharePoint and Defender XDR for multiple days
Microsoft · Global (Microsoft 365 cloud services)
Starting around 5:30pm UTC on August 31, 2026, Microsoft opened incident EX1464935 after users reported failed authentication, delayed mail delivery, broken mailbox search and admin-portal problems in Exchange Online. Microsoft's follow-up incident MO1465074 confirmed the disruption spread to Teams, SharePoint Online, OneDrive for Business, Microsoft Graph, Purview, the Microsoft 365 Admin Center, Copilot and Defender XDR because the services share an authentication component. Microsoft identified the root cause as a fault in a core authentication configuration and rolled out a phased remediation. Mailbox connectivity returned to expected thresholds within about 24 hours, but search functionality and full recovery of Exchange Online, Universal Print, OneDrive for Business and SharePoint Online remained impacted as of September 2, roughly 48 hours after the incident began.
2026-08-27
Proton: total cooling failure at Frankfurt datacenter takes down Mail, VPN, Drive and Pass for 2h18m
Proton AG · Frankfurt, Germany (Proton-operated datacenter)
Between 00:09 and 02:27 CEST on August 27, 2026, nearly all Proton services — Mail, VPN, Calendar, Drive, Pass, SimpleLogin, Wallet and the Lumo assistant — became unreachable. Proton engineers identified the cause at 00:38 CEST as what CEO Andy Yen publicly described as an "unprecedented total cooling failure" at the company's Frankfurt datacenter, which required manual intervention and a shift of traffic to backup sites rather than automated failover. Recovery began around 01:37 CEST, with full resolution by 02:27 CEST. Yen stated that no user data was exposed, deleted or compromised, since the failure was physical and facility-level rather than a breach or software defect, and Proton's end-to-end encryption was unaffected.
2026-08-20
Google Cloud us-west1 (Oregon) region: multi-service degradation for about 3h40m
Google Cloud · us-west1 region (The Dalles, Oregon)
From roughly 08:40 to 12:20 Pacific time on August 20, 2026, customers in Google Cloud's us-west1 region experienced timeouts, degraded service and elevated error rates across more than 20 products, including Compute Engine, GKE, Cloud SQL, BigQuery, Cloud Storage, Pub/Sub, Bigtable, Dataflow, Cloud Run and IAM. Google's Service Health dashboard tracked the incident through mitigation and recovery; services outside us-west1 were not affected. Google had not published a detailed root-cause explanation as of this writing.
2026-08-17
GitHub outage: retry storm and Central US datacenter network saturation cause 7h47m disruption to Issues, PRs, Actions and Copilot
GitHub (Microsoft) · Central US datacenter region, with failover traffic routed to Northern Virginia
From 13:28 to 21:15 UTC on August 17, 2026, GitHub.com suffered elevated errors and latency across Issues, Pull Requests, the API, Actions, Copilot and SAML/OIDC authentication. According to GitHub's own incident thread, a misconfigured autoscaling policy on an Istio sidecar pod caused four HAProxy nodes in GitHub's Central US datacenter to exhaust their connection flow limits, degrading the gateway authentication path. A retry bug in a client (identified as VS Code) then amplified Copilot Token Service traffic from a normal 7,000-9,000 requests per second to 70,000-100,000 requests per second, triggering a broader retry storm. At peak, web and API error rates reached about 20%, while archive and raw-content downloads saw roughly 50% error rates. GitHub mitigated the incident by throttling gateway retries, temporarily blocking inbound Copilot Token Service requests with HTTP 403s, and shifting failing traffic to its Northern Virginia site while the Central US network issue was debugged. Most services recovered by 16:36 UTC, GitHub Actions remained degraded until about 18:03 UTC, and the Copilot Token Service fully recovered at 21:02 UTC.
2026-08-14
Cloudflare logs 13 separate incidents in 8 days across R2, Durable Objects and Workers KV
Cloudflare · Global (regional impact concentrated in North America, Middle East and Southeast Asia)
Cloudflare's own status history recorded 13 distinct incidents between August 7 and August 14, 2026, rather than a single outage: the run opened with an R2 object-storage failure in Cloudflare's Eastern North America region on August 7, included a major-severity, Spamhaus-related email delivery disruption on August 12, and closed with a Durable Objects and Workflows availability drop resolved at 8:05 PM UTC on August 14. In between, Cloudflare logged elevated HTTP 503 errors on Magic Transit, Cloudflare WAN and the CF1 Appliance (resolved 9:16 PM on August 13), authentication failures on the Workers AI/MCP Server Portal, network congestion in the Eastern US and at Querétaro, Mexico, and a regional 5xx error spike affecting Kuwait, Bangkok, Jakarta and Dammam (resolved 5:37 AM on August 14). Twelve of the thirteen incidents were labeled minor severity by Cloudflare; only the August 12 email disruption was rated major. No single incident approached the scale of Cloudflare's November 18, 2025 global outage, but the density of the cluster renewed scrutiny of concentration risk given Cloudflare's share of web traffic.
2026-08-13
Data center cooling failure after utility power loss takes down Namecheap hosting, DNS and email
Namecheap (core infrastructure hall, PhoenixNAP data center) · Phoenix, Arizona, US
Overnight storms caused a utility power interruption that knocked out the chillers in Namecheap's core infrastructure hall at a Phoenix data center; to prevent hardware damage as the hall overheated, Namecheap deliberately shut down more than 5,000 servers. Because the facility also hosts Namecheap's authoritative DNS and account/control-plane systems, the outage took down namecheap.com, the customer dashboard, shared/VPS/dedicated hosting, EasyWP and Private Email (roughly 1.4 million inboxes), and broke DNS resolution for any third-party site still pointed at Namecheap's nameservers. The incident ran from about 6:28 a.m. ET, with service largely restored by mid-afternoon on 13 August and Namecheap reporting near-full restoration on 14 August; the company estimated more than 24 million registered domains were affected.
Engadget — Hosting provider Namecheap is down after data center cooling failure ↗
2026-08-06
Minneapolis carrier-hotel outage grounds Midwest flights, disrupts 911 and election systems
511 Building (Minneapolis carrier hotel/colocation facility; tenants include Lumen/CenturyLink) · Minneapolis, Minnesota, US (regional Midwest impact)
A telecommunications outage originated at the 511 Building in downtown Minneapolis, a carrier-hotel data center where most of Minnesota's telecom networks converge; a cybersecurity expert said it appeared to stem from a power or mechanical failure at the facility, though the exact technical cause was not publicly disclosed by the building's operator or tenants. The disruption knocked out links feeding a regional FAA air traffic control facility, prompting a roughly two-hour ground stop at Minneapolis–St. Paul International Airport around 2:00 p.m. CDT with average delays of 90 minutes, and knock-on disruption at airports across Michigan, Kansas, Missouri, North and South Dakota, Nebraska and Wisconsin. The same outage affected some 911 emergency call systems and, occurring the day before a Minnesota election, raised connectivity concerns for voting systems. The FAA said its investigation was continuing and reported no indication of a cyberattack.
MPR News — Outage at Minneapolis data center behind last week's telecommunications disruptions ↗
2026-07-24
AWS us-west-2 outage disrupts Reddit, Hulu, DoorDash, Apple Pay and PSN
Amazon Web Services · Oregon, US (us-west-2)
A networking hardware fault on the routing path between the us-west-2 region and the Seattle Metro cut connectivity for traffic crossing the region boundary, while traffic staying inside the region kept working. The core outage lasted about 80 minutes, with a longer recovery tail for some AWS Direct Connect customers routed through the Westin Building Exchange in Seattle (about 1 hour 17 minutes). It was the third distinct AWS reliability incident in about three months, following a data center thermal event in May and a network disruption in June, and knocked out or degraded Reddit, Hulu, DoorDash, Apple Pay, Snapchat, Fortnite and the PlayStation Network.
2026-07-23
Azure West US region loses datacenter-to-WAN connectivity after automated repair error
Microsoft Azure · West US region, US
An automated break-fix repair on optical network equipment inadvertently withdrew routes from all egress devices at a datacenter in the West US region, cutting datacenter-to-WAN connectivity. A subset of customers saw connectivity failures and increased latency on Azure and dependent Microsoft 365, Teams and Outlook services from 14:44 UTC; rollback began at 17:45 UTC, core connectivity was restored by 18:26 UTC, and full recovery was declared at 19:41 UTC, roughly 5 hours after onset.
2026-07-22
Transmission fault drops 3 GW of Northern Virginia data center load, causing wide-area grid flicker
PJM Interconnection / Dominion Energy (data center cluster) · Loudoun County ("Data Center Alley"), Virginia, US
A fault on a transmission line serving Northern Virginia's data center cluster triggered automatic protection systems at multiple facilities, which transferred to backup power and dropped about 3 GW of load — roughly 3% of PJM's total system demand — within seconds. The sudden load swing caused about 10 minutes of voltage flicker felt across a wide swath of the eastern US grid, from Chicago to Boston to Miami. No customer blackout resulted and normal operation was restored within minutes, but the event intensified scrutiny of data center load behavior on grid stability.
Data Center Knowledge — Fault in Data Center Alley triggered 3 GW load drop on PJM ↗
2026-07-16
AWS CloudFront serves 5xx errors, taking dependent websites offline
Amazon Web Services · Global (CloudFront CDN)
AWS's CloudFront content delivery network began serving 5xx errors instead of content, knocking offline or degrading multiple websites and services that depend on it for delivery.
The Register — AWS CloudFront outage serves errors instead of websites ↗
2026-07-15
Google Cloud europe-west4 outage after power and cooling failure
Google Cloud · Eemshaven, Netherlands (europe-west4-a)
A utility power disturbance triggered protective shutdowns of power and cooling systems in one zone, causing a cascading failure across multiple Google Cloud services. The incident lasted roughly 15 hours before full recovery.
2026-06-27
Cooling failure: University of Cambridge — West Cambridge Data Centre
University of Cambridge · Cambridge, United Kingdom
Systems went down during a UK heatwave in an incident attributed to the facility's cooling infrastructure; the cooling supplier said its units performed to specification. The university reported full cooling capacity restored, with access expected back on 6 July. The Dawn AI supercomputer and other services were offline for more than a week.
DatacenterDynamics — Data center housing UK's Dawn supercomputer suffers heatwave-related outage ↗
2026-06-24
Cooling failure: Portsmouth Hospitals University NHS Trust — Queen Alexandra Hospital on-site data center
Portsmouth Hospitals University NHS Trust · Portsmouth, United Kingdom
Chillers serving the on-site data center failed during a heatwave with temperatures above 34C. The loss of cooling capacity raised temperatures across parts of the estate while the trust worked to restore it. The hospital declared a critical incident; digital systems, theatres, cath labs and scanners were affected and some planned care was postponed.
2026-06-05
Fire at STT GDC/Tata Next-Gen Tower data center in Delhi
ST Telemedia Global Data Centres India / Tata Communications · Greater Kailash, New Delhi, India
A fire that started in a lithium-ion UPS battery room caused extensive damage to the STT Delhi 2 facility, injured two firefighters, and disrupted Google Cloud, Netflix and local ISPs across Delhi-NCR. Google Cloud reported degraded network capacity in the Delhi metro for about three weeks, and some colocation clients feared permanent data loss.
2026-05-07
Fire at NorthC data center takes IBM Cloud Amsterdam offline
NorthC Datacenters (tenant: IBM Cloud) · Almere, Netherlands
A fire broke out at the NorthC Almere facility at around 08:45 and took emergency services until the evening to bring under control, cutting power to the site. IBM Cloud's AMS03 data center went offline, and Utrecht University systems, a regional water board and Transdev bus/tram services were among those affected; restoring redundant power took about a week.
2026-05-07
Cooling failure: Amazon Web Services — Data center in zone use1-az4 (us-east-1), Northern Virginia
Amazon Web Services · Northern Virginia, United States
Temperatures rose inside a single data center and affected racks lost power during the thermal event. AWS brought additional cooling capacity online over several hours and recovered racks gradually, noting progress was slower than first expected. EC2 instances and EBS volumes on affected hardware in one availability zone were impaired.
2026-03-01
AWS me-central-1 outage after drones strike UAE data centers
Amazon Web Services · United Arab Emirates (me-central-1)
AWS reported that 'objects' — later confirmed as drones amid regional attacks — struck two availability-zone data centers, causing sparks and fire; fire crews cut power to one facility and its generators. With two AZs down, more than 100 services in the me-central-1 region were disrupted as the remaining zone became overloaded.
2026-02-20
Cloudflare BYOIP outage withdraws customer routes via BGP
Cloudflare · Global
A bug in Cloudflare's Addressing API during an automated cleanup task caused about 25% of Bring-Your-Own-IP prefixes worldwide to be withdrawn from BGP, making affected customers unreachable from the internet. The incident lasted 6 hours and 7 minutes; Cloudflare confirmed it was not an attack.
2026-02-07
Power failure: Microsoft Azure — West US data center
Microsoft Azure · United States
An electrical failure in an on-site transformer cut utility power to part of the facility. Recovery proceeded in phases, with some storage and compute infrastructure slow to return, which delayed dependent services. Multiple Azure services in West US saw unavailability, timeouts or latency from 07:58 UTC on 7 February to 04:24 UTC on 8 February; Windows Update and the Microsoft Store were also disrupted.
DatacenterDynamics — Microsoft suffered 'power interruption' at West US cloud region in February ↗
2026-01-25
Weather event: Oracle — Primary US data center site used by TikTok
Oracle · United States
A winter storm caused a power outage at an Oracle-operated site, leading to network and storage problems. TikTok's US joint venture said on 1 February that it had restored the platform together with Oracle. TikTok's US service was disrupted for about a week, with tens of thousands of servers affected.
DatacenterDynamics — TikTok says outage is resolved, confirms it was an Oracle data center ↗
2026-01-24
Fire: Global Switch — London Docklands data center (Nutmeg Lane)
Global Switch · London, United Kingdom
A lead-acid battery failed in a basement battery room run by a subcontractor, starting a small fire in the early hours. London Fire Brigade sent eight engines and about 60 firefighters because of the high-rise location, and the fire was out within two hours. No customer disruption was reported.
DatacenterDynamics — Global Switch suffers battery room fire at London data center ↗
2025-11-28
CyrusOne CHI1 cooling failure halts CME futures trading
CyrusOne (tenant: CME Group) · Aurora, Illinois, US
A chiller plant failure at CyrusOne's CHI1 data center on the evening of November 27 caused server rooms to overheat, forcing protective shutdowns. CME Group halted trading across its global futures and FX markets for roughly 10 hours on November 28, freezing about 90% of global listed derivatives trading.
2025-11-18
Cloudflare global outage from oversized bot management feature file
Cloudflare · Global
A database permissions change caused Cloudflare's Bot Management feature file to double in size, exceeding a hard limit and crashing the core proxy software across the network. Large parts of the web returned 5xx errors for several hours, affecting sites and apps including X, ChatGPT and Spotify.
2025-11-05
Cooling failure: Microsoft Azure — West Europe data center (single availability zone)
Microsoft Azure · Netherlands
Microsoft reported a thermal event affecting data center cooling systems, after which storage units in one availability zone shut down. The company estimated recovery at about 90 minutes and later cleared the incident. A subset of storage scale units went offline, degrading multiple Azure services in the West Europe region.
2025-10-29
Azure Front Door configuration error disrupts Microsoft services globally
Microsoft Azure · Global
An inadvertent configuration change deployed to the Azure Front Door platform produced DNS and routing failures from about 16:00 UTC, taking down or degrading the Azure Portal, Microsoft 365 web apps, Xbox/Minecraft authentication and thousands of customer sites. Microsoft rolled back to a last-known-good configuration, with recovery declared after roughly 8 hours.
2025-10-20
AWS us-east-1 outage after DynamoDB DNS automation failure
Amazon Web Services · Northern Virginia, US (us-east-1)
A latent race condition in DynamoDB's automated DNS management wiped the service's DNS records in us-east-1, cascading into roughly 15 hours of disruption across dozens of AWS services and major dependent platforms including Snapchat, Fortnite, Slack, Ring and Coinbase. It was one of the largest cloud outages on record, with over 17 million user reports on Downdetector.
2025-10-09
Fire: Centron — Hallstadt data center, Hallstadt (Bamberg)
Centron · Hallstadt (Bamberg), Germany
After maintenance work, a fault in the main low-voltage distribution and its UPS system led to a fire that the local fire service put out quickly. Centron attributed the fault to human error during the maintenance and added control and training measures. No service interruption; redundant infrastructure carried the load.
DatacenterDynamics — German MSP Centron suffers UPS fire at data center outside Nuremberg ↗
2025-10-04
Cooling failure: Lörrach District Hospitals — Hospital data center
Lörrach District Hospitals · Lörrach, Germany
The data center's air-conditioning failed overnight and servers overheated. The hospitals said the disruption took about five hours to resolve and found no sign of a cyberattack. Three hospitals lost most IT systems, MRI and X-ray were unusable and emergency cases were diverted.
DatacenterDynamics — Overheated server brings down hospitals in Lörrach district, Germany ↗
2025-09-26
Battery fire at South Korea's national government data center
National Information Resources Service (NIRS) · Daejeon, South Korea
A lithium-ion battery fire that broke out during UPS battery relocation work at the NIRS facility knocked more than 600 government digital services offline, including the Government24 portal and postal services. Full restoration of affected systems took weeks, prompting a national review of battery safety and government cloud resilience.
2025-09-08
Fire: Chirisa Technology Parks — CTP-01, Chesterfield County, Virginia
Chirisa Technology Parks · Chesterfield County, Virginia, United States
A battery room fire was reported at about 11:40 and Chesterfield County fire crews remained on site for an extended period to deal with hot spots. No injuries; the fire was confined to one room. It was the second fire at the campus within weeks.
DatacenterDynamics — Chirisa data center in Virginia suffers second fire in three weeks ↗
2025-06-12
Google Cloud global outage from Service Control crash
Google Cloud · Global
An invalid automated quota policy update triggered a null-pointer crash loop in Google's Service Control API management binaries worldwide, as the failing code path lacked error handling and feature flags. Dozens of Google Cloud and Workspace services failed for up to about 3 hours (with us-central1 recovery taking longer), also disrupting Cloudflare, Spotify and other dependent platforms.
2025-05-31
Power failure: Equinix — SG1, Singapore
Equinix · Singapore, Singapore
Equipment failed while SG1 was running with reduced power redundancy for scheduled maintenance, cutting supply to two data halls. Equinix described the event as minor and localised and restored power the same day. Two data halls lost power for more than an hour, disrupting Singtel, StarHub, ViewQwest and other providers.
DatacenterDynamics — Equinix's SG1 data center in Singapore hit by outage during maintenance ↗
2025-05-22
Battery room fire at Oregon data center leased by X
Digital Realty (tenant: X Corp) · Hillsboro, Oregon, US
A fire ignited in a lithium-ion backup battery room at the PDX11 facility in Hillsboro Technology Park, prompting a building evacuation and a lengthy firefighting response. X (Twitter) suffered a widespread outage, with degraded login, messaging and timeline functionality persisting into the following days.
2025-04-20
Cooling failure: University of Manchester — Kilburn Building (SpiNNaker machine room)
University of Manchester · Manchester, United Kingdom
The building's chilled-water supply failed over the Easter weekend, so the room's chillers could not remove heat and temperatures rose. Servers did not shut down automatically and staff powered the system off manually. The SpiNNaker neuromorphic computer overheated and had to be shut down.
2025-03-30
Fire: Equinix — SP4, São Paulo
Equinix · São Paulo, Brazil
A small fire in electrical equipment inside a customer's cage, reported to belong to a dark-fibre provider, was contained after water was discharged. Equinix rerouted the affected cloud circuits the same day. Equinix Fabric circuits and virtual connections to Oracle and Google clouds went down after a dark-fibre link was cut.
DatacenterDynamics — Equinix's SP4 data center in São Paulo suffers fire over weekend ↗
2025-03-29
Power failure: Google Cloud — Zone us-east5-c, Columbus, Ohio
Google Cloud · Columbus, Ohio, United States
Utility power was lost and the UPS, which should have bridged to generators, suffered a critical battery failure, leaving machines without power. Engineers bypassed the failed UPS to restore supply. More than 20 Google Cloud services were degraded or unavailable in us-east5-c for six hours and ten minutes.
DatacenterDynamics — UPS issue caused Google Cloud's March outage ↗
2025-02-17
Weather event: State of Connecticut (data center on Pfizer campus) — Groton data center, Groton, Connecticut
State of Connecticut (data center on Pfizer campus) · Groton, Connecticut, United States
High winds after a winter storm caused a power outage at the facility. When power returned, technicians found and repaired several hardware failures. State police computer-aided dispatch and law-enforcement administrative systems were affected.
DatacenterDynamics — Winter storm knocked out power at Pfizer's Groton data center in Connecticut ↗
2025-01-23
Weather event: Detran-DF (Federal District Department of Transit) — Detran data center, Brasilia
Detran-DF (Federal District Department of Transit) · Brasilia, Brazil
A lightning strike hit the data center. The department rescheduled citizen appointments, deployed traffic officers to direct traffic manually and had maintenance teams restore power. Detran's digital services went offline and traffic lights dependent on the data center malfunctioned.
DatacenterDynamics — Lightning strikes Brazilian data center, putting services out of action ↗
2024-12-17
Fire: 9Mobile — Main data center, Lagos
9Mobile · Lagos, Nigeria
The operator blamed a fire at its main data center, together with multiple fibre cuts, for the outage and apologised to subscribers while its technical team restored service. Voice and data services were disrupted for 9Mobile customers.
DatacenterDynamics — Nigeria's 9Mobile blames network outage on data center fire in Lagos ↗
2024-11-26
Water incident: Stanford University (Joint Science Operations Center) — JSOC server room, Stanford, California
Stanford University (Joint Science Operations Center) · Stanford, California, United States
A four-inch chilled-water pipe burst and left several inches of water in the server room. The water was drained by the next day, but equipment had to be dried and assessed before a restart. Processing and distribution of data from NASA's Solar Dynamics Observatory and IRIS instruments stopped.
The Register — NASA's Solar Dynamics Observatory datacenter flooded ↗
2024-10-24
Power failure: Google Cloud — Zone europe-west3-c, Frankfurt
Google Cloud · Frankfurt, Germany
Google's preliminary analysis blamed a power failure and a cooling issue that caused part of the zone to be powered down. The other two zones in the region saw only minor errors and the facility was returned to full operation the same afternoon. Customers lost access to VMs and disks in europe-west3-c from 02:30 to 15:09 local time.
DatacenterDynamics — Google Cloud region in Germany suffers 12 hour outage ↗
2024-09-10
Fire: Digital Realty — SIN11, Singapore
Digital Realty · Singapore, Singapore
A fire involving lithium-ion batteries in third-floor battery rooms triggered alarms at 07:45 and was reported under control that evening. Water used for firefighting pooled in a server room, so one building was powered down as an emergency measure; Alibaba spent the following week drying and migrating hardware. Alibaba Cloud's Singapore Availability Zone C was disrupted for days, with knock-on effects for ByteDance services.
2024-07-10
Grid event: Multiple operators (Dominion Energy / PJM grid) — About 60 data centers in Fairfax County, Fairfax County, Virginia
Multiple operators (Dominion Energy / PJM grid) · Fairfax County, Virginia, United States
A failed lightning arrestor on a 230 kV line produced a series of faults, and voltage-sensitive data centers switched to their backup systems simultaneously. NERC's incident review said the grid is not designed for the sudden loss of loads of this size. Roughly 1,500 MW of data center load left the grid at once, forcing PJM and Dominion to cut generation output to keep the system stable.
2024-07-08
Weather event: Lumen Technologies — Houston data center (Greenspoint), Houston, Texas
Lumen Technologies · Houston, Texas, United States
Hurricane Beryl knocked out commercial power to more than two million customers in the Houston area. Lumen kept partial service on generators while it waited for the utility to restore supply. Services from the facility were partially disrupted while it ran on backup generators.
DatacenterDynamics — Power outages impact Lumen's Houston data center after Hurricane Beryl ↗
2024-06-25
Water incident: Skyguide — Geneva air traffic control data center
Skyguide · Geneva, Switzerland
Thunderstorm flooding reached the basement housing the data center's cooling system. Skyguide brought in an external cooling unit and a mobile generator, restored partial capacity the next day and full capacity on 28 June. Geneva airspace was closed for two and a half hours; more than 50 flights were cancelled and about a dozen diverted.
DatacenterDynamics — Geneva airspace restored after flooding impacted data center ↗
2024-03-26
Power failure: Flexential — Portland-area facility (Cloudflare PDX01), Portland area, Oregon
Flexential · Portland area, Oregon, United States
Four Flexential-owned switchboards feeding Cloudflare's space failed at once, removing both primary and redundant power paths. Failover work done after the November 2023 outage kept customer impact small. Cloudflare's cages lost all power, but APIs and dashboards recovered automatically within about seven minutes; analytics took longer.
Cloudflare — Major data center power failure (again): Cloudflare Code Orange tested ↗
2023-11-02
Power failure: Flexential — PDX-04 (Hillsboro), Hillsboro, Oregon
Flexential · Hillsboro, Oregon, United States
After a utility feed was lost during unplanned maintenance, a ground fault on a transformer took the remaining supply and the generators offline, and UPS batteries drained within minutes. Cloudflare moved most control-plane functions to its disaster-recovery site the same day and launched a resilience programme. Cloudflare's control plane and analytics services were down or degraded from 2 November 11:44 UTC until 4 November 04:25 UTC.
Cloudflare — Post mortem on the Cloudflare Control Plane and Analytics Outage ↗
2023-10-26
Fire: Multiple operators (Khawaja Tower tenants) — Khawaja Tower, Dhaka
Multiple operators (Khawaja Tower tenants) · Dhaka, Bangladesh
Fire broke out on the 13th floor of a 14-storey building housing internet gateways, exchanges and data centers and was under control within about two and a half hours. Several international internet gateway operators' facilities were reported destroyed and the fire service criticised the building's fire precautions. Three people died; about 40 percent of Bangladesh's broadband users and 20 percent of mobile internet users had outages.
DatacenterDynamics — Data center fire in Dhaka, Bangladesh, kills three, takes millions offline ↗
2023-10-20
Power failure: Microsoft Azure — West Europe data center (zone AZ-01)
Microsoft Azure · Netherlands
Microsoft moved a data center onto generators after detecting voltage sags and swells from the grid, but a failure in part of the electrical distribution system stopped about 10 percent of generators from taking load. The affected racks lost power until the distribution fault was isolated. App Service, Cosmos DB, SQL DB, Storage and Virtual Machines were hit for some customers between 07:31 and 09:15 UTC.
DatacenterDynamics — Electrical distribution system failure causes outage at Microsoft data center ↗
2023-10-14
Cooling failure: Equinix — SG3, Singapore
Equinix · Singapore, Singapore
During a planned system upgrade a contractor wrongly sent a signal that closed valves on the chilled-water buffer tanks, and temperatures in several halls rose above the operating range. The Monetary Authority of Singapore ordered both banks to investigate why their backup data centers did not restore service quickly. DBS and Citibank online banking, payments and ATMs were disrupted for about two days; 2.5 million payment and ATM transactions could not be completed.
DatacenterDynamics — Equinix blames Singapore outage on third party contractor ↗
2023-10-09
Fire: PGIMER — Nehru Hospital data center UPS (C Block), Chandigarh
PGIMER · Chandigarh, India
The fire started in the battery bank of the centralised UPS serving the hospital's data center after a battery short-circuited. An inquiry found the institute had twice been advised by the fire department to move the 160-battery bank outside the building. A fire broke out inside a working hospital building.
2023-08-30
Cooling failure: Microsoft Azure — Australia East data center, Sydney
Microsoft Azure · Sydney, Australia
A utility voltage sag during a thunderstorm tripped all five running chillers in a data hall and only one standby unit came online. Microsoft powered down hardware to avoid heat damage; its review found the night shift of three people was too small to restart the chillers quickly and temporarily raised staffing to seven. Azure services in one Australia East availability zone were degraded after compute and storage units were shut down.
DatacenterDynamics — Microsoft's slow outage recovery in Sydney due to insufficient staff on site ↗
2023-08-30
Fire: Proximus — Evere data center, Brussels
Proximus · Brussels, Belgium
A fire detected around 09:00 in a Proximus data center prompted a brief staff evacuation. Operations were back to normal by about 10:30 and the national crisis centre activated alternative emergency numbers in the meantime. Belgian emergency numbers 112 and 101 were unreachable for less than half an hour.
2023-06-22
Cooling failure: Te Whatu Ora Health New Zealand — Hospital server facility, Wellington
Te Whatu Ora Health New Zealand · Wellington, New Zealand
An internal cooling system failed at 04:00 and several servers overheated. Applications were redirected automatically where possible and the fault was later resolved. Hospitals in Wellington, Kenepuru, Hutt and Wairarapa had delays to outpatient care and lost access to patient records, lab results and other systems.
2023-05-03
Fire: Zenith Bank — Primary data center
Zenith Bank · Nigeria
Local reporting said the backup power system caught fire and cut power to the data center. The disaster-recovery site was not running active-active, so services had to be switched over manually. Most of the bank's services went offline from about 07:00, affecting millions of customers.
DatacenterDynamics — Report: Nigeria's Zenith Bank hit by outage due to data center fire ↗
2023-04-26
Water incident: Global Switch — Paris campus
Global Switch · Paris, France
A cooling-system water leak reached a battery room and started a fire in one of the two buildings on the campus. The fire brigade contained it without injuries; Google reported water intrusion causing a multi-cluster failure and warned of an extended regional outage. Google Cloud region europe-west9 suffered a multi-zone outage, and French municipal and government websites hosted at the site went offline.
2023-04-19
Fire: Government of South Australia — Glenside Data Centre, Adelaide
Government of South Australia · Adelaide, Australia
An electrical fire started at 09:45 and was out before firefighters arrived, but it knocked out the data center's cooling system. Systems were still offline that evening while power to the cooling plant was being restored. SA Health's electronic medical records system and other government services failed, affecting several Adelaide hospitals.
DatacenterDynamics — Fire breaks out at Glenside Data Centre in Adelaide, Australia ↗
2023-03-29
Cooling failure: China Telecom (Guangzhou Telecom) — Guangzhou Telecom data center
China Telecom (Guangzhou Telecom) · Guangzhou, China
Chinese media reported that a cooling-system failure at the data center caused the outage, which was fixed later the same day. Tencent reportedly reprimanded, demoted or dismissed several managers over gaps in its disaster-recovery planning. WeChat and QQ functions including voice, payments, file transfer and mail were unavailable for several hours.
The Register — Tencent reportedly fires tech bosses after brief outages ↗
2023-03-13
Power failure: Vocus — VDC-MEL03 (530 Collins Street), Melbourne
Vocus · Melbourne, Australia
Faulty components in the A-feed UPS caused two losses of power, including to protected circuits, between 09:02 and 10:20. Emergency repairs restored dual-feed colocation power and Vocus scheduled a full UPS replacement. Power was lost twice in one morning, disrupting Vocus network, cloud, NBN and voice services and ISPs such as Aussie Broadband, Dodo and iPrimus.
DatacenterDynamics — Vocus data center in Melbourne suffers outage after UPS fault ↗
2023-02-14
Fire: Cyxtera — BOS1 (Waltham), Waltham, Massachusetts
Cyxtera · Waltham, Massachusetts, United States
An electrical arc flash blew the doors off a battery cabinet in a power room, producing smoke and a fire response. Firefighters shut down power and evacuated the building; one person was taken to hospital for possible smoke inhalation before Cyxtera restored supply. Oracle NetSuite customers across the US were offline from around midday on 14 February until late morning the next day.
2023-02
Cooling failure: Microsoft Azure — Southeast Asia data center (single availability zone), Singapore
Microsoft Azure · Singapore, Singapore
A utility power surge tripped cooling units offline in one data center. As temperatures climbed Microsoft shut down compute and storage scale units to protect hardware and reduce cooling load, then restored them once cooling recovered. Customers in the Southeast Asia region lost services; affected sites included Nanyang Technological University and Singapore's Central Provident Fund Board.
2022-12-18
Cooling failure: PCCW DC (Vantage Data Centers) — Site hosting Alibaba Cloud Hong Kong Zone C
PCCW DC (Vantage Data Centers) · Hong Kong, Hong Kong
Refrigeration equipment failed at a third-party site used by Alibaba Cloud. Engineers from both companies repaired the cooling plant and services were brought back gradually; Alibaba promised compensation under its service agreements. Alibaba Cloud ECS, database, storage and network products in Hong Kong Zone C failed, affecting customers including the crypto exchange OKX and Macau government websites.
2022-11-23
Fire: QTS — Piscataway data center (101 Possumtown Road), Piscataway, New Jersey
QTS · Piscataway, New Jersey, United States
A heavy fire on the roof was reported at 02:45 and extinguished by 05:00 using high-volume water streams. Firefighters took extra care because of flammable construction materials stored for an extension to the facility. No casualties and no outages were reported.
DatacenterDynamics — QTS New Jersey data center roof catches fire ↗
2022-11
Cooling failure: States of Guernsey — Government data centre
States of Guernsey · Guernsey
One of two air-conditioning systems had been broken since June and the second failed in November, leaving IT equipment without cooling. A PwC review found outdated equipment, unclear ownership and lapsed maintenance contracts for the UPS and generator. A series of four outages between November and January took the States' website and internal systems offline and disrupted schools, airport boarding cards and benefit payments.
DatacenterDynamics — Guernsey gov't IT outages caused by data center power and cooling failures ↗
2022-10-15
Fire: SK C&C (SK Group) — Pangyo data center
SK C&C (SK Group) · Pangyo, South Korea
A fire that reportedly started in a basement battery room at the SK Group facility housing Kakao and Naver infrastructure was extinguished the same night, but Kakao services stayed down for days. The government summoned the heads of Kakao, Naver and SK, and Kakao was criticised for lacking a workable failover; it later paid compensation to users and small businesses. KakaoTalk and most other Kakao services, along with some Naver services, went down across South Korea; DCD later put the disruption at 127 hours.
2022-09-05
Weather event: Twitter — Sacramento (SMF) data center region, Sacramento, California
Twitter · Sacramento, California, United States
Record heat in California led to a total shutdown of physical equipment in the SMF region, according to an internal memo. Twitter froze non-essential production changes while it ran on its two remaining regions. Twitter lost its entire Sacramento region; the service stayed online by shifting load to Atlanta and Portland.
DatacenterDynamics — Twitter data center brought down by California's extreme heat wave ↗
2022-08-08
Power failure: Google — Council Bluffs data center, Council Bluffs, Iowa
Google · Council Bluffs, Iowa, United States
An arc flash occurred while electricians were working on a substation next to the data center buildings. The injured workers were taken to the Nebraska Medical Center and Google said it was investigating with local authorities. Three electricians were critically injured.
2022-07-19
Cooling failure: Google Cloud — Data center hosting zone europe-west2-a, London
Google Cloud · London, United Kingdom
On the UK's hottest recorded day, multiple redundant cooling systems failed at the same time in one building serving europe-west2-a. Google shut down part of the zone to protect hardware; cooling was restored around midnight local time and Google said it would audit cooling equipment across its fleet. A portion of zone europe-west2-a was powered down, causing VM terminations and Persistent Disk errors for a subset of Google Cloud customers.
2022-07-19
Cooling failure: Oracle Cloud Infrastructure — UK South (London) data centre
Oracle Cloud Infrastructure · London, United Kingdom
During the record UK heatwave part of the cooling infrastructure in Oracle's London data centre developed a fault. Oracle powered off equipment to prevent hardware damage and restored services after the cooling plant was repaired. Networking, storage and compute resources in Oracle's UK South region became unavailable to some customers.
The Register — Google, Oracle cloud servers suffer outage in UK heatwave ↗
2022-07-19
Cooling failure: Guy's and St Thomas' NHS Foundation Trust — Two on-site hospital data centres, London
Guy's and St Thomas' NHS Foundation Trust · London, United Kingdom
Air-conditioning serving both of the trust's data centres failed during the July 2022 heatwave, bringing down patient-record and diagnostic systems. An NHS review put the cost at GBP 1.4 million and judged the incident likely preventable, citing a broken hose connector at St Thomas' and difficulty locating a water supply at Guy's. Clinical IT systems at two London hospitals failed; staff reverted to paper, appointments were cancelled and full recovery took about six weeks.
2022-06-29
Cooling failure: MTN South Africa — Gallo Manor data center, Johannesburg
MTN South Africa · Johannesburg, South Africa
Afrihost said extended load shedding caused a cooling-system failure at the MTN facility. MTN confirmed that one cooler was affected after a power outage and said the redundant cooling system was fully restored within 45 minutes. Afrihost hosting servers were damaged and customer websites and email were offline for more than 30 hours.
2022-06-04
Fire: University of Bretagne-Sud — Tohannic campus data center, Vannes
University of Bretagne-Sud · Vannes, France
A system in the data center caught fire on a Saturday night. The automatic nitrogen-based protection system stopped the fire quickly and prevented serious damage, and the university restarted services one by one. All university servers and most applications were shut down for an extended period.
DatacenterDynamics — Fire at University of Bretagne-Sud data center causes outage ↗
2022-05-26
Fire: Sarawak state government — Wisma Bapa Malaysia server room
Sarawak state government · Malaysia
A third-floor server room in the state secretariat building caught fire at about 19:00. Power to the building was cut and firefighters extinguished the fire by 20:45; the cause was under investigation. State government services including the SPay app and the government portal went offline.
DatacenterDynamics — Malaysian state government services brought offline by server room fire ↗
2022-05
Power failure: Equinix — MA1, Manchester
Equinix · Manchester, United Kingdom
Equinix said power load was transferred to generators during the incident and that it was working with a vendor on a battery replacement. Users reported outages of around ten minutes. Internet Exchange, Metro Connect, Fabric and Connect customers dropped for several minutes.
2022-04-12
Fire: Supreme Court of the Philippines — Centennial Building data center, Manila
Supreme Court of the Philippines · Manila, Philippines
A UPS in the data center reportedly exploded and caught fire at around 06:00. The Bureau of Fire Protection brought it under control quickly and the court went ahead with the results announcement. The Supreme Court's website went offline on the day Bar examination results were due.
DatacenterDynamics — Data center fire hits Manila's Supreme Court ↗
2022-03
Fire: Telecommunication Infrastructure Company (TIC) — TIC telecom data center, Tehran
Telecommunication Infrastructure Company (TIC) · Tehran, Iran
A fire was reported at a key telecom building in Tehran and NetBlocks recorded a sharp national drop in connectivity. A deputy telecommunications minister attributed the fire to an electrical connection and said it had been repaired. Internet connectivity collapsed in Tehran and other Iranian cities across multiple providers.
DatacenterDynamics — Reported fire at Iranian telco data center causes widespread Internet outages ↗
2022-03
Power failure: Cybernet — Karachi data center
Cybernet · Karachi, Pakistan
A UPS battery malfunction led the operator to halt the whole system to avoid further damage, according to local reporting. Services were restored in phases. Dozens of companies went offline, including the Pakistan Stock Exchange website and digital services of several banks.
2022-01-30
Fire: Equinix — MD2, Madrid
Equinix · Madrid, Spain
Smoke was detected in a power room at 04:15 on a Sunday. Madrid emergency services contained and extinguished a small fire with no injuries, and Equinix opened an investigation. Equinix reported only a brief interruption in power supply; no customer outages were reported.
DatacenterDynamics — Small fire at Equinix Madrid data center, no injuries ↗
2022-01-10
Power failure: Interxion (Digital Realty) — LON1 (Brick Lane), London
Interxion (Digital Realty) · London, United Kingdom
Power was lost at LON1 around 18:00 UTC; the fault hit multiple feeds and the switchgear that should have transferred load to the on-site generator also failed. Customers complained that Interxion's own support systems were down during the incident. Customers on the affected floor lost power for about four hours, and London Metal Exchange electronic trading was down for around five hours overnight.
DatacenterDynamics — Outage at Interxion facility in London takes down Metals Exchange ↗
2021-12-22
Power failure: Amazon Web Services — Data center in zone USE1-AZ4 (us-east-1), Northern Virginia
Amazon Web Services · Northern Virginia, United States
A single data center in USE1-AZ4 lost power early in the morning Pacific time. AWS restored power about an hour after first reporting problems, though recovery of affected instances took longer. EC2 instances and EBS volumes in one availability zone failed, disrupting services such as Slack, Hulu, Coinbase and Epic Games.
2021-12-02
Fire: Cyber 1, Jakarta
Cyber Data Center (Cyber 1 building) · Jakarta, Indonesia
Fire broke out in a second-floor server room around midday. The Jakarta fire service sent 22 trucks and about 100 firefighters and had the fire under control within half an hour. Two technicians died of smoke inhalation, and several Indonesian digital services hosted in the building were disrupted.
DatacenterDynamics — Two killed in fire at Cyber 1 data center in Jakarta, Indonesia ↗
2021-06-10
Cooling failure: Amazon Web Services — Data center in one eu-central-1 availability zone, Frankfurt
Amazon Web Services · Frankfurt, Germany
A failure took out air handlers and the hall overheated. A fire-suppression system then triggered and removed oxygen from the room, so staff had to evacuate and could not enter for about an hour to fix the fault. One availability zone was lost for about three hours as servers and network gear powered off at unsafe temperatures.
DatacenterDynamics — AWS Frankfurt shuts down when safety systems evacuate staff, block access ↗
2021-04-04
Fire: WebNX — Ogden data center, Ogden, Utah
WebNX · Ogden, Utah, United States
When city power dipped, a recently tested backup generator failed catastrophically and caught fire, triggering fire suppression. Firefighters cut power to the site as a precaution, which prevented a return to utility supply even after the grid recovered. Hosting customers, including some Ogden city IT services, suffered a prolonged outage and some servers needed rebuilding after water damage.
DatacenterDynamics — Generator fire brings down WebNX data center in Utah ↗
2021-03-19
Fire: OVHcloud — SBG1 battery room (Strasbourg campus)
OVHcloud · Strasbourg, France
Nine days after the SBG2 fire, staff found smoke in a room of disconnected batteries in SBG1 and called firefighters. OVHcloud cut the supply to SBG1 and SBG4 and sent most personnel home; nobody was injured and the source was dealt with quickly. Recovery work after the SBG2 fire was halted and power to SBG1 and SBG4 was cut again as a precaution.
DatacenterDynamics — OVHcloud: firefighters return to a second incident in Strasbourg data center ↗
2021-03-10
Fire: OVHcloud — SBG2 (Strasbourg campus)
OVHcloud · Strasbourg, France
A fire broke out at OVHcloud's Strasbourg campus and consumed the entire SBG2 building plus part of SBG1; power to all four data centers on the site was cut. A later report by the Bas-Rhin fire service noted that SBG2 had no automatic extinguishing system and no general electrical cut-off. OVHcloud restarted the surviving halls in stages over the following weeks. The SBG2 data center was destroyed and four of twelve rooms in neighbouring SBG1 were lost, taking customer servers and hosted websites offline.
DatacenterDynamics — OVH fire update: four halls of SBG1 destroyed, as well as all of SBG2 ↗
2021-03-09
Water incident: NOAA / National Weather Service — Data center at NWS headquarters, Silver Spring, Maryland
NOAA / National Weather Service · Silver Spring, Maryland, United States
A burst pipe flooded the NWS headquarters building and knocked out the data center housed there. NOAA worked with the building owner on drying, damage assessment and system restoration. The National Data Buoy Center's ocean and marine buoy data feed went offline.
DatacenterDynamics — NOAA data center floods, sinks buoy data ↗
2021-02-15
Weather event: City data center, Austin, Texas
City of Austin · Austin, Texas, United States
Rolling power cuts during Winter Storm Uri took the city's data center offline. It was brought back on generators and continued to run without grid power afterwards. City websites and the 311 non-emergency hotline were down from early morning until about 15:00.
DatacenterDynamics — Texas power outages bring Austin city data center offline ↗
2020-11-07
Fire-suppression discharge: Global Switch — Ultimo data center, Sydney
Global Switch · Sydney, Australia
A false alarm set off the fire-suppression system in the Ultimo facility, damaging servers belonging to one tenant. Global Switch said there was no fire and that power stayed up; Tabcorp restored services over the following days. Tabcorp's TAB wagering, Keno and gaming services were offline through a major racing weekend after its servers were damaged.
DatacenterDynamics — Global Switch's fire safety 'malfunction' damages Sydney data center servers ↗
2020-09-14
Cooling failure: Microsoft Azure — UK South data center
Microsoft Azure · United Kingdom
Several chilled-water pumps shut down and temperatures rose past operating limits, so automation powered off network, compute and storage. Engineers put the cooling plant in manual mode, reset the pumps and restored infrastructure once temperatures were safe. The facility was unavailable from 13:54 UTC to 00:41 UTC, taking down services including the UK's Covid-19 tracking site.
DatacenterDynamics — Cooling loss causes outage at Microsoft Azure's UK South region ↗
2020-08-27
Fire: Telstra — London hosting centre (Greenwich View Place, Isle of Dogs)
Telstra · London, United Kingdom
A fire damaged part of a supply room and tripped bus-bar breakers, cutting one of the site's power systems; unconfirmed reports pointed to a faulty UPS. London Fire Brigade sent four engines, nobody was hurt and engineers worked to restore power by generator. The lower half of the facility went offline.
DatacenterDynamics — Report: faulty UPS triggers fire at Telstra's London data center ↗
2020-08-18
Power failure: Equinix — LD8 (Harbour Exchange, Docklands), London
Equinix · London, United Kingdom
A faulty UPS system caused a power outage that was followed by a fire alarm, although there was no fire. Equinix restored supply during the day and relaxed access rules so customers could recover equipment. Hosting, cloud and telecom customers in LD8 lost power from about 04:20, and many London Internet Exchange members were affected.
DatacenterDynamics — Equinix LD8 data center experiences major outage ↗
2020-05-18
Cooling failure: Microsoft Azure — Central India data center, Pune
Microsoft Azure · Pune, India
A utility power problem moved the data center to generators, but a number of air handling units in two internal zones did not make the transition. Temperatures rose and automation shut down network and storage; technicians isolated the faulty units and restored equipment in stages. Azure services in Central India were disrupted from 12:41 UTC on 18 May to 08:30 UTC on 19 May; payment processors were among those affected.
DatacenterDynamics — Microsoft Azure suffers lengthy Indian cloud outage ↗
2020-03-15
Power failure: Microsoft Azure — West Central US region
Microsoft Azure · United States
Microsoft said an isolated power event hit a limited set of infrastructure in the region. Engineers isolated and recovered the affected equipment. A subset of West Central US customers had trouble connecting to resources between about 20:30 and 00:30 UTC.
2019-10-09
Grid event: Lawrence Berkeley National Laboratory (NERSC) — NERSC supercomputing center, Berkeley, California
Lawrence Berkeley National Laboratory (NERSC) · Berkeley, California, United States
PG&E cut power to the Berkeley Lab campus as part of a wildfire-prevention shutoff affecting hundreds of thousands of customers. NERSC and Cori were returned to service after power came back. NERSC supercomputers including Cori were taken offline and the laboratory closed.
DatacenterDynamics — NERSC shuts down supercomputers amid PG&E blackout ↗
2019-08-31
Power failure: Amazon Web Services — Data center in us-east-1, Northern Virginia
Amazon Web Services · Northern Virginia, United States
A data center lost utility power during the US Labor Day weekend and its backup generators then failed. AWS recovered most systems, but hardware damage made some storage volumes unrecoverable. EC2 instances were lost and some EBS volumes were destroyed; affected customers had to restore from backups.
The Register — AWS celebrates Labor Day weekend by roasting customer data in US-East-1 BBQ ↗
2019-08-23
Cooling failure: Amazon Web Services — Data center in one ap-northeast-1 availability zone, Tokyo
Amazon Web Services · Tokyo, Japan
A failure of the data center control system caused multiple redundant cooling systems to stop in part of the zone from 12:36 JST. Cooling was restored at 15:21 and most instances and volumes had recovered by 18:30. A share of EC2 servers in one availability zone shut down from overheating, impairing instances and degrading EBS volumes.
AWS — Summary of the Amazon EC2 Issues in the Asia Pacific (Tokyo) Region ↗
2019-02-07
Fire-suppression discharge: Wells Fargo — Shoreview data center, Shoreview, Minnesota
Wells Fargo · Shoreview, Minnesota, United States
Wells Fargo said power was shut down at a facility after smoke was detected following routine maintenance. The local fire department stated there was no fire and that a fire-suppression system had been set off by construction dust. Customers nationwide could not use ATMs, online banking or mobile banking.
DatacenterDynamics — Widespread Wells Fargo issues blamed on data center outage, cause disputed ↗
2018-10-15
Fire: AT&T — 1666 Firman Drive data center, Richardson, Texas
AT&T · Richardson, Texas, United States
An electrical fire that began at a power switch burned both primary and backup electrical systems, according to the City of Richardson. An AT&T spokesperson initially attributed it to a lightning strike, which the company later declined to confirm. AT&T U-verse broadband and TV customers in the Dallas-Fort Worth area lost service for much of the day.
2018-09-04
Cooling failure: Microsoft Azure — South Central US data center, Texas (South Central US region)
Microsoft Azure · Texas (South Central US region), United States
Lightning caused voltage swells on the utility supply; one data center moved to generators and its mechanical cooling shut down despite surge suppression. Once the thermal buffer was used up, hardware powered down automatically and some equipment was damaged, lengthening the recovery. Azure services in South Central US, plus dependent services such as Office 365 and Skype, were disrupted for several days.
DatacenterDynamics — Microsoft publishes preliminary analysis of major cloud outage ↗
2018-06-19
Cooling failure: Microsoft Azure — North Europe data center, Dublin
Microsoft Azure · Dublin, Ireland
As outside air temperature rose, environmental control systems failed in part of the data center and humidity climbed in the operational areas. Network devices and hard drives supporting storage units suffered hardware failures and had to be recovered. A range of Azure services in North Europe were down from 17:44 UTC on 19 June to 04:30 UTC on 20 June.
The Register — Azure North Europe downed by the curse of the Irish – sunshine ↗
2018-06-02
Fire-suppression discharge: State data center
State of New Jersey (Office of Information Technology) · United States
A fire-suppression system was triggered in the state's data center on a Saturday and brought down government systems. Most were restored over the following days. State agencies were offline for more than 24 hours, including the Election Law Enforcement Commission days before a primary election.
DatacenterDynamics — Fire suppression system causes State of New Jersey data center outage ↗
2018-05-31
Power failure: Amazon Web Services — Data center in one us-east-1 availability zone, Northern Virginia
Amazon Web Services · Northern Virginia, United States
A power event hit one data center at 15:13 PDT. AWS warned that resulting hardware failures meant some instances and volumes in that zone might not be recoverable. RDS, Redshift, WorkSpaces, EC2 and EBS in one availability zone had connectivity problems for about 30 minutes.
DatacenterDynamics — 'Power event' at AWS data center disrupts US-EAST-1 ↗
2018-05-26
Power failure: National Australia Bank — Melbourne mainframe data centre
National Australia Bank · Melbourne, Australia
NAB said a power issue isolated its mainframe and that technicians needed several hours to bring systems back. The bank offered compensation to affected business customers. ATMs, EFTPOS, internet and mobile banking and call centres were down for about seven hours on a Saturday.
The Register — NAB mainframe turns its TOESUP after power outage, offline 7 hours ↗
2018-04-18
Fire-suppression discharge: DigiPlex — Väsby data centre (Nasdaq module), Upplands Väsby (Stockholm)
DigiPlex · Upplands Väsby (Stockholm), Sweden
An errant fire-extinguisher system released inert gas in the module rented by Nasdaq, and the noise of the discharge destroyed server disk drives. Nasdaq switched to a backup system and brought in replacement servers. Nasdaq Nordic trading was offline until 14:00 local time.
The Register — Noise from blast of gas destroys Digiplex data depot disk drives ↗
2018-02-13
Power failure: Vocus — Alexandria data centre, Sydney
Vocus · Sydney, Australia
UPS maintenance went wrong shortly before 07:00 and cut power to the whole site. Services were restored through the morning, although some equipment failed to recover. All A and B power feeds were lost, taking down colocation customers such as Servers Australia and some Vocus network services.
DatacenterDynamics — Botched UPS maintenance causes outage at Vocus data center ↗
2017-12-12
Power failure: UKFast — MANOC 5, 6 and 7, Manchester
UKFast · Manchester, United Kingdom
A contractor working about 0.7 km away drove a spike through the main feed to UKFast's transformer at 10:28. The damaged supply was unstable, the UPS route was disturbed and the generator sets failed to synchronise; power was restored at 11:42. eCloud platforms and customers' physical servers went down; power returned after about 75 minutes but some recovery ran into the evening.
DatacenterDynamics — UKFast suffers outage after workman takes pick axe to cable ↗
2017-11-21
Fire: Cairns and Hinterland Hospital and Health Service — Health service server room
Cairns and Hinterland Hospital and Health Service · Cairns, Australia
A short circuit in a battery cabinet in the server room is believed to have started a small fire around 10:00. Emergency services attended and the health service deployed extra staff to cover the disruption. Admission and discharge systems and the hospital switchboard failed across two hospitals and the wider health service.
DatacenterDynamics — Data center battery incident causes fire in Australian hospital ↗
2017-11-09
Power failure: OVH — SBG1, SBG2 and SBG4, Strasbourg
OVH · Strasbourg, France
An electrical failure left SBG1, SBG2 and SBG4 without power; OVH's founder said the site's power design lacked dual feeds and normal redundancy because two of the halls were built quickly from shipping containers. A separate optical-network software bug hit Roubaix the same morning. OVH announced a plan to rework the Strasbourg electrical infrastructure. All Strasbourg customers were offline from about 07:15 to 11:15 while three data centers had no power.
2017-09-29
Fire-suppression discharge: Microsoft Azure — North Europe data center
Microsoft Azure · Ireland
Inert fire-suppression gas was released unexpectedly during routine maintenance, which shut down the air handling units by design. Temperatures rose in the affected zone and some systems shut down or rebooted before the air handlers were restarted. Some North Europe customers could not reach or manage resources between 13:27 and 20:15 UTC.
DatacenterDynamics — Fire suppression accident causes Microsoft Azure outage ↗
2017-07-10
Power failure: Global Switch 2 (GS2), 10th floor, London
Global Switch · London, United Kingdom
The tenth floor of GS2 lost power for a short period in the afternoon and supply returned within about 20 minutes. On 17 July the same floor lost power again for under an hour. EX Networks' customers briefly lost service; a second power loss hit the same floor a week later.
DatacenterDynamics — Global Switch suffers short power outage in London ↗
2017-06-29
Water incident: OVH — P19, Paris
OVH · Paris, France
Coolant from OVH's rack water-cooling system leaked onto a Dell EMC VNX storage array at around 19:00 and crashed it. OVH restored the hosted databases from backups. More than 50,000 websites were unavailable for about 24 hours.
The Register — What did OVH learn from 24-hour outage? Water and servers do not mix ↗
2017-05-27
Power failure: British Airways (facility managed by CBRE) — Boadicea House, London (Heathrow)
British Airways (facility managed by CBRE) · London (Heathrow), United Kingdom
Power to the Boadicea House data centre was interrupted at about 09:30 and the site was down for roughly 15 minutes. BA said the uncontrolled way power was restored damaged systems and brought down its failover as well. BA's IT systems collapsed over a holiday weekend, grounding flights and stranding tens of thousands of passengers.
The Register — BA IT systems failure: Uninterruptible Power Supply was interrupted ↗
2017-05-10
Fire: Undisclosed operator (tenant: UniSuper) — Port Melbourne data centre
Undisclosed operator (tenant: UniSuper) · Melbourne, Australia
A fire at one of the fund's data centres early on a Wednesday triggered a partial system shutdown. UniSuper said no data was lost and restored member services after working through the following night. UniSuper's member portal and telephony were down for more than a day.
2017-03-31
Cooling failure: Microsoft Azure (third-party managed facility) — Japan East data center
Microsoft Azure (third-party managed facility) · Japan
A rotary UPS failed in a way that cut the power distribution feeding all air handling units, so the temperature rose throughout the data center. Resources shut down automatically to avoid overheating and were restored once cooling returned. Virtual machines and services depending on Azure Storage were unavailable for several hours across Japan East.
DatacenterDynamics — Azure Cloud in Japan suffers after UPS failure, cooling outage ↗
2016-09-10
Fire-suppression discharge: ING Bank Romania — Regional data center
ING Bank Romania · Romania
A planned test of the Inergen inert-gas extinguishing system produced noise loud enough to damage dozens of hard drives in servers and storage. ING restored service by moving operations to its backup data center. ATMs, card payments and online banking were unavailable for about 10 hours, affecting more than one million customers.
DatacenterDynamics — Noise from fire drill breaks ING Bank data center ↗
2016-09-10
Power failure: Global Switch 2 (GS2), London
Global Switch · London, United Kingdom
A high-voltage fault occurred in the cable end box of the output circuit breaker for one of the site's diesel rotary UPS units. Damaged components were replaced and the remaining DRUPS terminations were checked; the site ran on mains without backup for two days. A 222 ms power break disrupted customers including Claranet, EX Networks and Tagadab; Claranet took about two days to restore all services.
DatacenterDynamics — Details emerge of Global Switch outage in London ↗
2016-08-08
Power failure: Delta Air Lines — Atlanta data center, Atlanta, Georgia
Delta Air Lines · Atlanta, Georgia, United States
Delta attributed the failure, which began around 02:30 EDT, to a loss of power at its Atlanta data center; Georgia Power said the fault was in Delta's own equipment rather than the grid. Check-in, reservations and flight-status systems came back in stages. A roughly six-hour systems outage grounded more than 450 flights on the first day, with hundreds more cancelled the next day.
DatacenterDynamics — Tens of thousands of Delta passengers grounded by outage ↗
2016-07-21
Power failure: Telehouse North, London
Telehouse · London, United Kingdom
A power failure on the third floor of Telehouse North, which Telehouse linked to a tripped circuit breaker, cut supply to carrier equipment. It came one day after a UPS fault at the nearby Telecity LD8 site. BT broadband customers and ISPs relying on BT Openreach lost or had degraded service from about 07:45.
2016-07-20
Power failure: Equinix (Telecity) — LD8 (8/9 Harbour Exchange), London
Equinix (Telecity) · London, United Kingdom
A fault in one UPS system at the former Telecity site caused a brief power interruption. Equinix said service to the limited number of affected customers was restored within minutes and opened an investigation. About 10 percent of BT customers' internet usage was affected, along with other ISPs present in the building.
DatacenterDynamics — Faulty UPS at Telecity data center causes Internet outages across the UK ↗
2016-06-05
Power failure: Amazon Web Services — Facility in one ap-southeast-2 availability zone, Sydney
Amazon Web Services · Sydney, Australia
Severe weather knocked out a regional substation and several AWS facilities lost utility power. In one of them the diesel rotary UPS line-ups failed to transfer load to their generators because of an unusually long voltage sag, and a latent software bug then slowed recovery of some instances. EC2 instances and EBS volumes in one availability zone lost power, taking down Australian sites including Stan, Foxtel Play, Menulog and Nine.
AWS — Summary of the AWS Service Event in the Sydney Region ↗
2016-05-09
Power failure: Salesforce — Primary data center for the NA14 instance
Salesforce · United States
Power supply problems at the instance's primary data center caused the first outage and prompted a site switch to another facility. The instance then failed again at the new site, and chief executive Marc Benioff apologised publicly. The NA14 instance was down for nearly two hours after the power fault and then for more than 11 hours after a follow-on failure.
The Register — Marc Benioff apologizes as Salesforce NA14 instance goes TITSUP ↗
2016-01-28
Power failure: GitHub — Primary data center
GitHub · United States
A brief power disruption caused a large part of GitHub's servers and network gear to reboot, and some machines failed to come back because firmware no longer recognised their drives. Engineers rebuilt the affected clusters to restore service. GitHub.com was unavailable for about two hours.
The Register — How a power blip briefly broke GitHub's boxes and tripped it offline ↗
2016-01-14
Power failure: Verizon — Data center hosting JetBlue systems
Verizon · United States
A maintenance issue caused a power outage at the Verizon facility at 11:37. Reservations and check-in were back by about 14:30 and all services from the site were restored by around 20:00. JetBlue booking, check-in and flight-status systems failed, delaying flights for several hours.
DatacenterDynamics — Verizon data center failure causes JetBlue air travel delays ↗
2015-12
Water incident: Vodafone UK — Kirkstall Road data center, Leeds
Vodafone UK · Leeds, United Kingdom
Severe flooding from the River Aire reached the facility and it temporarily lost power. Repairs took longer than usual because engineering teams were delayed by the weather; all issues were reported resolved by early January. Vodafone customers experienced disruption to voice and data services over the Christmas period.
DatacenterDynamics — Vodafone UK data center suffers outage due to floods ↗
2015-08-20
Incident: Equinix; CoreSite; Internap; Level 3 — Equinix LA1, CoreSite LA1 (One Wilshire) and nearby sites, Los Angeles, California
Equinix; CoreSite; Internap; Level 3 · Los Angeles, California, United States
An explosion and fire in the basement of the 811 Wilshire office building, believed to have started in its generator, took out an on-site power station and briefly left twelve buildings without electricity. Equinix LA1 and CoreSite LA1 failed over to generators until utility power returned. At least two data centers lost utility power and ran on generators; some customers saw network and power disruption.
DatacenterDynamics — Los Angeles explosion disrupts data centers ↗
2015-08-13
Weather event: Google Cloud — Data center hosting zone europe-west1-b
Google Cloud · Belgium
Four successive lightning strikes caused brief power loss to storage systems; Google staff later said the local grid rather than the building was hit. Backup power took over quickly, but some older storage hardware could not retain recent writes after repeated battery drain. Some Persistent Disks in europe-west1-b had I/O errors for hours and a very small share of recently written data was permanently lost.
The Register — Act of God damaged data on Google cloud disks ↗
2015-02
Cooling failure: Rideout Health — Off-site data center, Marysville, California
Rideout Health · Marysville, California, United States
One of two HVAC units in the data center burned out and the backup overheated and failed shortly after. Clinicians lost electronic access to parts of patient records until systems were restored. The hospital's electronic health record system was down for about a week and some treatments were postponed.
DatacenterDynamics — Aircon failure paralyzes California hospital ↗