Internet & cloud outages — February 2026

No major internet outages or widespread disruptions detected today. All major cloud providers are reporting normal operations. AWS, Google Cloud, and Microsoft Azure show no active incidents. OpenAI reports all systems operational. Cloudflare is performing scheduled maintenance at its Chicago (ORD) datacenter through 14:00 UTC, with traffic rerouted and potential minor latency increases for nearby users. The ongoing Newark (EWR) HTTP request latency issue, first identified on February 16, continues to be monitored but has not expanded in scope. A minor Cloudflare Peering Portal authentication issue was identified yesterday due to a third-party auth provider. GitHub is operational with a minor Copilot agent sessions incident. Fastly and Akamai are running normally — Akamai resolved an India region packet loss issue yesterday. Google Cloud resolved a Vertex AI Gemini API error rate issue on February 27 after approximately two hours. Overall, the internet is stable and functioning normally. The residual Cloudflare EWR monitoring is contained and not causing widespread impact.

No major internet-wide outages are occurring today. The global internet infrastructure is functioning normally, with all major cloud providers and platforms reporting healthy status. Cloudflare has two active minor incidents: 1. Newark (EWR) Datacenter HTTP Latency — First reported on February 16, this issue involved elevated latency affecting a subset of HTTP requests routed through Cloudflare's Newark, NJ datacenter. A fix was implemented on February 26, and Cloudflare confirmed that Data Loss Prevention (DLP) services were not impacted. The incident remains in monitoring status as of today. 2. Peering Portal Authentication Issue — Reported early today (February 27), authentication to peering.cloudflare.com is affected due to an outage with a third-party authentication provider. Cloudflare has confirmed this does not affect CDN caching or edge security features. The issue is still being investigated. Cloudflare also resolved several other issues over the past 72 hours, including dashboard login problems, custom error page issues, and network performance degradation in Frankfurt and Vancouver — all resolved between February 26-27. All other major providers checked — Google Cloud, Microsoft Azure, Azure DevOps, GitHub, and OpenAI — are reporting fully operational status with no active incidents. The internet is up.

No major internet-wide issues detected today. All major cloud providers and infrastructure services are operating normally. Minor localized issues: - Cloudflare is monitoring a lingering HTTP latency issue at its Newark (EWR) datacenter affecting the Data Loss Prevention (DLP) service. A fix was implemented and the team is monitoring results. Several Cloudflare datacenters are also undergoing scheduled maintenance windows today. - Fastly reports a partial outage on customer-facing service pages (not CDN delivery). All other major providers — AWS, Google Cloud, Azure, GitHub, OpenAI, and Akamai — report normal operations. Recently resolved: Cloudflare experienced a significant BYOIP service outage on February 20 when a maintenance automation bug caused 25% of BYOIP customer IP prefixes to be withdrawn via BGP. The incident lasted approximately 6 hours before full routing stability was restored. AWS CloudFront also experienced a DNS resolution failure on February 10 that cascaded across eight interconnected services before being resolved. The internet is up.

Internet Status — February 25, 2026: Minor Cloudflare Degradation; Major Providers Stable Overall, the internet is functioning normally today with one minor ongoing issue. Cloudflare continues to experience degraded performance at its Newark (EWR) datacenter, with elevated latency affecting a subset of HTTP requests and intermittent errors for customers using the Data Loss Prevention (DLP) service. This issue has been under investigation since February 16, with fixes being implemented. Additionally, Cloudflare has scheduled maintenance windows today across several datacenters including Minneapolis (MSP), Porto Alegre (POA), San Jose (SJC), Singapore (SIN), and Marseille (MRS), which may cause slight latency increases in those regions. All other major infrastructure providers are reporting normal operations. AWS, Google Cloud, and Azure show no active incidents. Akamai, OpenAI, and GitHub are operational. Recent resolved incidents of note: - Cloudflare BYOIP/BGP outage on February 20 (6 hours, fully resolved) — Cloudflare withdrew customer BYOIP routes via BGP, causing affected customers to be unreachable from the internet. - AWS CloudFront DNS resolution failure on February 10 (resolved Feb 11) — CloudFront returned NXDOMAIN responses, cascading across 8 AWS services and impacting 20+ downstream platforms including Salesforce, Discord, and Adobe. - ChatGPT experienced widespread outages on February 3-4, with over 28,000 user reports on the first day. Sources: Cloudflare Status (cloudflarestatus.com), AWS Health Dashboard (health.aws.amazon.com), Google Cloud Status (status.cloud.google.com), Azure Status (azure.status.microsoft)

Internet Status — February 24, 2026: Minor Cloudflare Degradations; Feb 20 BGP Outage Resolved Overall, the internet is functioning normally today with no major outages. AWS, Google Cloud, Azure, GitHub, Fastly, Akamai, and OpenAI are all reporting operational status with no active incidents. Cloudflare is conducting scheduled maintenance across five datacenters today (Miami, San Jose, Kansas City, Toronto, and Las Vegas), with maintenance windows running between 08:00 and 15:00 UTC. Traffic is being re-routed from these locations, which may cause slight latency increases for users in affected regions. Additionally, Cloudflare Data Loss Prevention (DLP) and Cloudflare Sites are reporting degraded performance, and the Newark (EWR) datacenter continues to experience elevated HTTP request latency affecting a subset of traffic — an issue first identified on February 16. Azure DevOps is experiencing minor partial issues in the United Kingdom region, affecting Boards, Test Plans, and other services. This appears limited in scope and impact. The most significant recent event was the Cloudflare BGP outage on February 20, which lasted approximately 6 hours. A change to how Cloudflare manages IP addresses in its Bring Your Own IP (BYOIP) pipeline caused 25% of BYOIP prefixes (1,100 out of 6,500) to be unintentionally withdrawn via BGP. Major sites including Uber, Workday, Minecraft, Wikipedia, and Microsoft Outlook were affected. Cloudflare has confirmed this was not caused by a cyberattack and has published a detailed post-mortem (https://blog.cloudflare.com/cloudflare-outage-february-20-2026/). They are implementing circuit-breaker mechanisms to prevent similar incidents. Sources: - Cloudflare Status: https://www.cloudflarestatus.com - Cloudflare Post-Mortem: https://blog.cloudflare.com/cloudflare-outage-february-20-2026/ - Google Cloud Status: https://status.cloud.google.com - Azure Status: https://azure.status.microsoft - AWS Health Dashboard: https://health.aws.amazon.com - GitHub Status: https://www.githubstatus.com

Cloudflare continues to experience elevated latency at its Newark (EWR) datacenter, an issue that has persisted since February 16. A subset of HTTP requests are affected, and customers using the Data Loss Prevention (DLP) suite may see intermittent errors. Cloudflare has identified the problem and is working on a fix, with maintenance at the EWR datacenter scheduled for February 24. Cloudflare is also performing scheduled maintenance across several datacenters today, including Toronto (YYZ), Portland (PDX), Fukuoka (FUK), Houston (IAH), and Kochi (COK). Users routing through these locations may experience temporary latency increases. Fastly is reporting a partial outage affecting two points of presence in Asia (Fujairah and Lucknow). All other Fastly services are operational. Google Cloud, Microsoft Azure, GitHub, OpenAI, and AWS are all reporting normal operations. In recent resolved incidents, Cloudflare experienced a significant outage on February 20 when an internal bug caused approximately 1,100 BYOIP (Bring Your Own IP) prefixes to be withdrawn via BGP. The incident lasted about 6 hours and has since been fully resolved, with Cloudflare publishing a detailed post-mortem. AWS also resolved a brief Direct Connect connectivity issue in Northern Virginia on February 21. Overall, the internet is functional with some localized Cloudflare and Fastly issues. No widespread disruptions are occurring.

Today the internet is largely stable, though one active issue persists and a significant outage from earlier this week has been fully resolved. Active Issues: • Cloudflare Newark (EWR) datacenter continues to experience elevated HTTP request latency, ongoing since February 16. Some Data Loss Prevention (DLP) customers are seeing intermittent errors. Cloudflare is investigating with fixes in progress. Recently Resolved (Past 72 Hours): • On February 20, Cloudflare experienced a major outage when approximately 1,100 BYOIP (Bring Your Own IP) prefixes were unintentionally withdrawn via BGP due to a configuration change error. The incident lasted 6 hours and 7 minutes, making affected customer services unreachable from the internet. A full postmortem has been published at blog.cloudflare.com. This was not caused by a cyberattack. • Several related Cloudflare services were also impacted on Feb 20–21, including Workers AI (elevated 429 errors), Bot Management (detection issues), Analytics Engine API (500 errors), and Durable Objects (free tier invocation failures). All have been resolved. • GitHub Actions experienced extended job start delays for larger hosted runners on February 20, now resolved. All Clear: AWS, Google Cloud, Azure, Fastly, Akamai, OpenAI, and GitHub are all reporting normal operations today. Overall: One localized Cloudflare issue persists, and the effects of Thursday's major BGP outage have been fully resolved. The internet is up.

Cloudflare is experiencing several minor active incidents today. The Newark (EWR) datacenter has elevated HTTP request latency, Workers AI is seeing higher error rates (a fix has been deployed and is being monitored), and the Data Loss Prevention (DLP) service has degraded performance. Additionally, Cloudflare is performing scheduled maintenance across five US datacenters (St. Louis, Dallas, Minneapolis, Columbus, and Houston), which may cause slight latency increases in those regions. X (formerly Twitter) experienced another outage earlier today, with over 77,000 users reporting issues on Downdetector. The homepage failed to load for many users. Services have since been restored and X is operating normally as of late evening. All major cloud providers are reporting normal operations. AWS, Google Cloud, and Microsoft Azure show no active incidents. GitHub and OpenAI are fully operational. This has been a turbulent week for internet infrastructure. The Feb 16 Cloudflare BGP routing incident — caused by a misconfigured automated deployment in Ashburn, Virginia — cascaded across the internet, affecting AWS, X, and thousands of websites for hours. X suffered a massive global outage the same day. AWS CloudFront had DNS resolution failures on Feb 10 that impacted 20+ downstream platforms. The internet has largely stabilized, though Cloudflare continues to work through residual issues. Sources: cloudflarestatus.com, status.cloud.google.com, azure.status.microsoft, health.aws.amazon.com, status.openai.com, Downdetector, istheservicedown.com

The internet is largely stable today following yesterday's significant cascading outage. On February 16, a BGP routing misconfiguration during a routine update at Cloudflare's Ashburn, Virginia data center triggered a cascading failure that took down portions of AWS, X (Twitter), and thousands of other services for approximately four hours. Cloudflare engineers identified the root cause within 40 minutes, but recovery required coordinating with over a dozen upstream providers to flush corrupted BGP caches. Full service was restored by Sunday evening. X (Twitter) experienced a separate outage spike on February 16-17, with over 41,000 reports on Downdetector. Users reported blank timelines, infinite loading, and error messages. The platform stabilized by early February 17. This marks X's third major outage in 2026. Today, Cloudflare is conducting scheduled maintenance at its Atlanta (07:00-13:00 UTC) and Portland (14:00-23:59 UTC) data centers, which may cause minor latency increases in those regions. Cloudflare's Newark facility is also showing elevated HTTP latency and degraded Data Loss Prevention (DLP) performance since February 16. Virgin Media reported localized connectivity issues for some UK users starting around 7:11 AM BST. AWS, Google Cloud, and Azure are reporting no major active incidents as of February 17. The AWS CloudFront issues from February 10-11 and the cascading impact from yesterday's Cloudflare incident have been fully resolved. Overall, the internet is operational with residual effects from yesterday's Cloudflare BGP incident still being monitored. No new widespread outages are in progress. Sources: Cloudflare Status (cloudflarestatus.com), Downdetector, Tom's Guide, Dataconomy, WebProNews, BleepingComputer

Multiple notable incidents are affecting internet services today. X (formerly Twitter) experienced a major global outage beginning around 8:00 AM ET, with over 41,000 reports on Downdetector at its peak. Users worldwide saw blank timelines, failed logins, and "Something went wrong" error messages on both mobile and desktop. As of approximately 10:00 AM ET, the platform is recovering, with outage reports dropping below 1,000. No official explanation for the cause has been provided. AWS saw a spike in service disruption reports starting around 4:46 PM, with nearly 100 incident reports logged within minutes. The timing closely mirrors the X outage, suggesting a possible shared infrastructure dependency. AWS has not confirmed the specific cause. Cloudflare reported increased latency affecting HTTP requests processed in its Newark data center. Partial outages were also observed in data centers across Africa, Asia, and South America, including Accra, Dar es Salaam, and Bogota. Scheduled maintenance is underway in the Bangalore and Sofia data centers. Fastly is experiencing a partial outage affecting Asia-Pacific data centers, including Singapore and Tokyo. Other regions remain operational. Azure, Google Cloud, GitHub, and OpenAI report no significant incidents as of this writing. Akamai resolved an incident on February 10 and is operating normally. Overall, while several major platforms and infrastructure providers are experiencing issues, these appear to be separate incidents rather than a coordinated internet-wide disruption. Most affected services are in recovery or partially degraded. Sources: Downdetector, Dataconomy, Tom's Guide, TechRadar, MacRumors, Cloudflare Status, Fastly Status

The internet is largely stable today. All major cloud providers — AWS, Google Cloud, and Azure — report no active incidents as of this morning. AWS resolved an operational issue in the eu-north-1 region overnight (Feb 14). An internal networking fault disrupted traffic between services within the region, particularly in eun1-az3. Error rates and latencies began improving around 00:46 UTC, with full recovery confirmed by 04:05 UTC. External connectivity was unaffected throughout. Cloudflare reports minor partial outages at several edge locations across Africa (Dar es Salaam, Djibouti, Kinshasa, and others), Asia (Chandigarh, Kanpur, Da Nang, Astana, and others), Latin America (Bogota, La Paz, and several Brazilian cities), Europe (Minsk, Skopje), and North America (Anchorage, Norfolk). All core Cloudflare services remain operational — these are localized edge-node issues affecting traffic routing in those specific regions. GitHub resolved a file upload disruption on February 13 and is currently fully operational. OpenAI reports all systems operational. Looking back at the past week: February has seen a cluster of notable incidents, including the AWS CloudFront DNS resolution failure on Feb 10 (cascading across 8+ services), a DNSimple complete service outage on Feb 10, Azure networking issues in West US on Feb 7-8, and a GitHub availability degradation on Feb 9. All have been resolved. No BGP routing anomalies or DNS infrastructure issues are currently reported. The Cloudflare BGP route leak from January 22 (Miami) has long been resolved. Overall: the internet is up. Minor edge disruptions exist but do not affect the vast majority of users. Sources: health.aws.amazon.com, status.cloud.google.com, azure.status.microsoft, cloudflarestatus.com, status.openai.com, githubstatus.com

## Current Active Issues **Fastly CDN — Elevated Error Rates (Asia-Pacific)** Fastly is reporting elevated errors for its Content Delivery Network, with a partial outage affecting Asia-Pacific points of presence including Singapore (QPG, SIN), Tokyo (NRT), and several other locations. The issue has been ongoing for approximately 7 hours. Other Fastly regions remain operational. **Cloudflare — Scheduled Maintenance** Cloudflare is performing scheduled maintenance at its Chicago (ORD) and Ashburn (IAD) datacenters today (Feb 13, 08:00–13:00 UTC and 05:00–13:00 UTC respectively). Traffic is being rerouted, which may cause slight latency increases for users routed through these locations. ## Recently Resolved (Past 72 Hours) **AWS CloudFront DNS Failure (Feb 10–11)** — A significant DNS resolution failure affected CloudFront distributions globally, cascading across 8 AWS services and impacting 20+ downstream platforms including Salesforce, Discord, Claude, and Adobe. The acute phase lasted approximately 1 hour, with full recovery by 4:00 AM UTC on Feb 11. **Microsoft 365 Admin Center (Feb 10)** — North American administrators were unable to access the Microsoft 365 admin center due to authentication endpoint failures. Resolved Feb 10 at 13:56 EST. **GitHub Degraded Availability (Feb 9)** — GitHub.com, the GitHub API, Actions, Git operations, and Copilot experienced two periods of degraded availability on Feb 9. Both incidents were resolved the same day. **Cloudflare Service Issues (Feb 12)** — Multiple Cloudflare services experienced issues including R2 FedRAMP high error rates, Turnstile validation failures, Zero Trust posture check issues, and dashboard SSO connector problems. All resolved on Feb 12. ## Overall Assessment The only active issue of note is Fastly's partial CDN outage in Asia-Pacific. All major cloud providers (AWS, Google Cloud, Azure) are operational. The past few days have been eventful — particularly the AWS CloudFront DNS incident — but the internet is broadly stable today. Sources: [AWS Health Dashboard](https://health.aws.amazon.com), [Google Cloud Status](https://status.cloud.google.com), [Azure Status](https://azure.status.microsoft), [Cloudflare Status](https://www.cloudflarestatus.com), [Fastly Status](https://www.fastlystatus.com), [GitHub Status](https://www.githubstatus.com), [OpenAI Status](https://status.openai.com)

Several localized incidents are active today, but no widespread internet disruption is occurring. **Microsoft 365 Admin Center (Ongoing):** Since February 10, Microsoft 365 admin center has been experiencing an outage affecting North American business and enterprise customers. Administrators report HTTP 5xx errors, prolonged loading times, and session timeouts. Microsoft is investigating possible capacity or load-related issues. No estimated resolution time has been provided. **Cloudflare — Analytics Issue (Investigating):** Cloudflare is investigating a "Zone-Versioned Dashboard Analytics" issue reported at 10:42 UTC on February 11. Zone Versioning customers may see inaccurate data on HTTP Traffic Analytics dashboards. Cloudflare is also conducting scheduled maintenance across multiple datacenters today (Ashburn, Rio de Janeiro, Minneapolis, Las Vegas), which may cause slight latency increases for users in affected regions. **GitHub — Recovering from Recent Outages:** GitHub experienced significant outages on February 9-10 affecting Actions, Codespaces, Git operations, Issues, Pages, and Webhooks. Services have since recovered, though GitHub has been struggling with availability overall in early February. An earlier Actions outage on February 2 also disrupted hosted runners for several hours. **All Clear:** AWS, Google Cloud, Azure, OpenAI, Fastly, and Akamai are all reporting normal operations. No active BGP incidents or undersea cable issues have been detected. **Overall Assessment:** The internet is up. A few provider-specific incidents are active — most notably the Microsoft 365 admin center issue — but core internet infrastructure and major cloud platforms are operating normally.

Overall, the internet is functioning normally today with a few minor incidents being tracked. OpenAI is monitoring elevated error rates affecting GPT 5.2 on ChatGPT, which began at 16:45 UTC. A mitigation has been deployed and error rates have returned to normal, but the team continues to monitor for sustained stability. Cloudflare is conducting scheduled maintenance at its Singapore (18:00–22:00 UTC) and Portland (14:00–23:59 UTC) data centers. Traffic is being rerouted, with slight latency increases possible in affected regions. Additionally, several Cloudflare edge locations across Africa, Asia, the Americas, and parts of Europe are showing partial outages — a pattern typical of routine edge network activity. GitHub resolved multiple service disruptions over the past 48 hours, including degraded Pull Request performance and Copilot policy propagation issues on February 10, and broader degradation affecting Actions, Issues, Git Operations, and Webhooks on February 9. A more significant incident on February 2–3 saw GitHub Actions hosted runners go down for approximately 4.5 hours due to a backend storage policy change. All incidents are now fully resolved. Akamai resolved a Certificate Provisioning System (CPS) access issue earlier today (03:22 UTC). All other Akamai services remain operational. AWS, Google Cloud, and Azure all report normal operations with no active incidents. No active BGP routing incidents have been detected. The most recent notable event was a Cloudflare route leak on January 22, which lasted 25 minutes and affected IPv6 traffic through Miami — that issue has been fully resolved. Sources: cloudflarestatus.com, status.cloud.google.com, azure.status.microsoft, githubstatus.com, status.openai.com, akamaistatus.com, health.aws.amazon.com

As of February 10, 2026, the internet and major cloud infrastructure are operating normally with only minor, isolated incidents affecting specific services. **Current Incidents:** 1. **Akamai Certificate Provisioning System** — The CPS portal experienced access issues beginning at 00:23 UTC. Akamai deployed a fix at 03:22 UTC and is monitoring for full resolution. This affects certificate management through the Akamai Control Center but does not impact general CDN services. 2. **DigitalOcean Amsterdam Network Maintenance** — Scheduled infrastructure upgrades are underway in the AMS2 region (09:00-17:00 UTC). DigitalOcean expects no customer impact from these improvements. **Services Operating Normally:** - Vercel (core platform and edge network) - Discord (all systems at 99%+ uptime) - GitHub (no reported incidents) - AWS, Google Cloud, Azure (spot checks clear) - Cloudflare, Fastly (CDN services operational) **Assessment:** No internet-wide outages detected. The identified incidents are minor and localized to specific provider services. Global connectivity and core cloud infrastructure remain healthy. Sources: Vercel Status, Discord Status, DigitalOcean Status, Akamai Status Page

Cloudflare is currently experiencing a minor service outage affecting multiple regions worldwide. Partial outages have been reported in Africa (Djibouti City, Ouagadougou, Kinshasa, Abidjan, Addis Ababa), Asia (Bandar Seri Begawan, Malé, Kochi, Astana), Europe (Skopje), Latin America & Caribbean (Arica, Bogotá, Campinas, and other Brazilian cities), Middle East (Haifa), and North America (Jacksonville, Norfolk, Anchorage). Most of Cloudflare's core services remain operational. The company recently resolved two incidents from February 7-8 involving GraphQL API timeouts and HTTP 500 errors. Cloudflare has scheduled extensive maintenance across multiple datacenters (Portland, Singapore, and others) continuing through early March 2026. The internet remains mostly operational. This is a regional incident affecting specific Cloudflare data center locations, not a global outage. Other major cloud and infrastructure providers (AWS, Google Cloud, Azure, GitHub, Fastly, Akamai) are not reporting significant incidents at this time.

All major internet infrastructure providers are reporting normal operations as of February 8, 2026. Monitoring checks across core internet infrastructure show no significant disruptions: • Cloud Providers: AWS, Google Cloud Platform, and Microsoft Azure are all operational • CDN & Edge Networks: Cloudflare and Vercel report all systems functioning normally • Development Platforms: GitHub is fully operational • Major Services: All monitored services showing normal status No BGP routing incidents, DNS failures, or widespread connectivity issues have been detected. Internet traffic patterns appear normal with no anomalies reported. The internet is up and running smoothly today.

Two notable incidents are being tracked today, though neither represents a widespread internet disruption.\n\nMicrosoft Azure is recovering from an unexpected utility power interruption at its West US datacenter. Backup power systems activated and the facility has been fully operational since 12:30 UTC, but some storage and compute resources are still completing recovery. Dozens of Azure services in the West US region — including Virtual Machines, App Service, Azure Functions, SQL Database, Cosmos DB, and Azure OpenAI Service — show degraded availability. Azure engineering teams are actively working on validation, with updates expected hourly. Other Azure regions are unaffected.\n\nCloudflare is reporting minor partial outages at several edge locations across Africa, Asia, Europe, Latin America, and North America (including Anchorage, Jacksonville, and Norfolk in the US). All core Cloudflare services — API, Dashboard, CDN, and Workers — remain fully operational. Extensive scheduled maintenance is planned across Cloudflare datacenters from February 9 through February 27.\n\nAll other major providers are operating normally. AWS, Google Cloud, OpenAI, GitHub, Akamai, Azure DevOps, and Fastly report no active issues.\n\nRecently resolved incidents from the past 72 hours include: GitHub Copilot degraded performance due to high Opus 4.6 demand (Feb 6, resolved), GitHub Actions workflow delays (Feb 3, resolved), Azure VM provisioning and managed identity issues (Feb 2-3, resolved), and a brief ChatGPT outage (Feb 4, resolved).\n\nNo BGP routing incidents or DNS infrastructure issues have been detected today. The internet is up, with localized service degradation in Azure West US.

The internet is operating normally. No major ongoing outages or infrastructure incidents are currently affecting global connectivity. Recent resolved incidents (past 72 hours): - Microsoft Azure (Feb 2–3): VM deployments, AKS node provisioning, and Managed Identity services experienced failures across multiple regions. Root cause was a configuration change that restricted access to Microsoft-managed storage accounts. A secondary Managed Identity outage in East US / West US lasted approximately 6 hours. Both issues are now fully resolved. (https://www.theregister.com/2026/02/03/azure_virtual_machine_outage/) - OpenAI / ChatGPT (Feb 3–4): Elevated error rates affected ChatGPT and API users on consecutive days, peaking at 12,000+ Downdetector reports on Feb 3 and 10,000+ on Feb 4. OpenAI has since confirmed full recovery — all services are operational. (https://www.techradar.com/news/live/chatgpt-openai-outage-down-february-4-2026) - Cloudflare: A brief spike in errors at the Ashburn, VA datacenter (02:50–03:00 UTC today) was quickly resolved. Scheduled maintenance was completed today in Düsseldorf, Tokyo, Dallas, and Sioux Falls with no service impact. (https://www.cloudflarestatus.com) - EE Mobile (UK): Outgoing call failures peaked at ~7,800 reports before being resolved within an hour. (https://www.techradar.com/news/live/ee-down-february-2026) No issues reported: AWS, Google Cloud Platform, and Google Workspace are all operating normally. BGP watch: A notable BGP Prefix-SID Attribute leak incident caused routing instability earlier this period, amplified through internet exchanges running BIRD. The incident primarily affected networks on Juniper/Arista hardware and has been resolved. (https://blog.benjojo.co.uk/post/bgp-attr-40-junos-arista-session-reset-incident) Bottom line: It was a bumpy start to the week for Azure and OpenAI users, but as of today, the internet is stable and no active incidents require attention.

No major ongoing outages or infrastructure incidents are currently affecting global connectivity. Azure and ChatGPT disruptions from earlier this week are fully resolved. AWS, GCP, and Cloudflare operating normally.