Cloudflare status: hosting issues and outage reports

Problems in the last 24 hours

The graph below depicts the number of Cloudflare reports received over the last 24 hours by time of day. When the number of reports exceeds the baseline, represented by the red line, an outage is determined.

Most Reported Problems

The following are the most recent problems reported by Cloudflare users through our website.

• Domains (41%)
• Cloud Services (25%)
• Hosting (16%)
• Web Tools (13%)
• E-mail (6%)

Live Outage Map

The most recent Cloudflare outage reports came from the following cities:

City	Problem Type	Report Time
Manchester	Domains	8 days ago
Angers	Cloud Services	19 days ago
London	Domains	21 days ago
Noida	Hosting	1 month ago
Jewar	E-mail	1 month ago
Braga	Web Tools	1 month ago

Full Outage Map

Community Discussion

Tips? Frustrations? Share them here. Useful comments include a description of the problem, city and postal code.

Beware of "support numbers" or "recovery" accounts that might be posted below. Make sure to report and downvote those comments. Avoid posting your personal information.

Cloudflare Issues Reports

Latest outage, problems and issue reports in social media:

• 
  Michael Ramos (@backnotprop) reported 1 minute ago

  Every "ADE" is going to be pushed into one of either of (some might try to do all): - Linear competitor - Notion competitor - diffxyz/Ai-review competitor and away from "a harness for harnesses" - and/or misstep into remote execution (this requires all in customer bets. Like you either all go all in into the linear model or you do not - I can't imagine this scaling). But there's a much stronger durable layer nobody's really hitting at other than the infrastructure providers - context/artifacts has a lot of exciting potential. You can see it with cursor origin, cloudflare artifacts, code[dot]storage are pointing at. A lot of innovation to be had here & on top of - beyond "hey, share your HTML with me" There's still room for middle layer execution innovation, and it might smell like memory, but nobody's doing memory right.

• 
  𝐀𝐤𝐫𝐚𝐦 (@beingakramraja) reported 4 minutes ago

  Akash is processing 1.7 billion tokens every single day on openrouter right now outpacing cloudflare venice, elizaos, morpheus, gensyn all paying customers running real ai workloads on akash the narrative isn't that akash could become the decentralized aws. it's that it already is for a growing list of ai companies who need cheaper compute akash just launched homenode beta people with rtx 4090s and 5090s sitting at home can now connect their gpu to the network and earn from ai inference demand this changes the supply side completely instead of relying on 58 enterprise providers, the network starts pulling in consumer hardware globally more supply means more competitive pricing which means more demand which means more akt burned the things akash is building that most ct hasn't priced in yet virtual machines launching this quarter enterprise workloads that couldn't run on containers now can starcluster acquiring 7,200 nvidia gb200 gpus protocol-owned compute at hyperscale confidential computing via tee the feature enterprises require before migrating serious workloads $akt is at $0.62 the roadmap reads like a company that's two quarters away from being unignorable

• 
  63green (@63green) reported 5 minutes ago

  So @Cloudflare works overtime to destroy thei reputation by sending emails despite every notification turned off, and I’m willing to grant their request to never, ever trust this criminal company, and never use them. Any company willing to **** you by email will **** your data.

• 
  IJav (@javmung) reported 10 minutes ago

  @MSU_NW_FANG @MaplestoryU @nexpacetime Also need to test WARP From CLoudflare.. that helps a lot. But most likely is his/her ISP. mine was disconencting a lot, they did reset my NAT, and assigned me a public IP address, and problems are gone.. so routing probably the problem.

• 
  Headmaster Duck (@HeadmasterDuck) reported 12 minutes ago

  @specialkdelslay First thing, put a free cloudflare account in front of this, see how much their basic bot mitigation helps. Next, if you don't mind throwing $20/mo at the CF pro plan, this is a mostly solved problem between their super bot fighter and ability to issue challenge requests from the predictable regions of the globe. If $20/mo isn't in the cards, you can keep blocking IPs and also look into blocking by certain headers and user agents.

• 
  John Doe (@StanleyMasinde_) reported 17 minutes ago

  @SamProgramiz I’m glad you’re now seeing that Rust is not babysitting programmers on line one. It just makes you acknowledge your mistake. A panic caused a Cloudflare outage earlier this year. On paragraph 2, Rust will not even compile. Yes memory safety goes beyond nulls. If the language doesn’t handle an illegal reference, the hardware will send an interrupt and the kernel will kill the entire program. Now imagine you have. // processing some stuff Foo() passes bar() fails with an interrupt or NPE fizz() is completely skipped. Your program just corrupted data because null possibility was ignored.

• 
  Midnightpulls JP (@midnightpullsjp) reported 18 minutes ago

  Always wild when you are using X, suddenly Cloudflare kicks in, X locks your account until you verify it via email, wtf is going on with this platform at times :>

• 
  Ayushman Mallick (@AyushmanMallick) reported 20 minutes ago

  5/ You hand over a paid API key, so it's security-reviewed. The key goes only over HTTPS, only in a header, to one stateless @Cloudflare proxy that never stores it. XSS and SSRF hardened. Templates use a strongly-consistent Durable Object. The proxy is fully open-source.

• 
  tcpdump (@dump_tcp) reported 20 minutes ago

  @EddCoates if need any help with cloudflare I can help with some rules also it seems you're webserver code is bottlenecking you causing that error usually due to not being able to handle that many connections or to much cpu usage the webserver process dies i suggest using #golang best lang

• 
  Shantanu Landore (@ShantanuVL) reported 22 minutes ago

  @itsasmolsush Oof well my non tech tech company has everything set up over cloudflare so we need to log in with MFA once a day... and the prompt to login comes at the worst point of the day everyday so

• 
  DownWithBigBrother (@DWBB1984) reported 24 minutes ago

  @ultrasxiv Fair on bandwidth being a real cost, but the 2GB figure is a long way out. Cheapest DO droplet includes 500GB+ outbound, Hetzner 20TB+. At 600-700GB household use you’re a pound or two over on DO, zero on Hetzner. Stays around the base £4-5 for most, not £300. And “un-bannable” was the precise word, not hyperbole. A commercial VPN is bannable because it’s a named brand with known IPs, a company that can be pressured or blocked. That’s the weakness. Self-hosting removes the target entirely. There’s no technical category called “a VPS used as a VPN.” It’s a rented server running standard encryption (WireGuard, IPsec), the same protocols carrying every bank settlement, ATM link and corporate tunnel on earth. To ban it you’d have to block those protocols (killing Visa, every corporate VPN, all remote work) or blacklist the datacentre IP ranges (AWS, Hetzner, OVH) that host the actual internet: payment gateways, banking backends, Stripe, Cloudflare, gov services. You can’t separate “server someone might tunnel through” from “server running the shop you’re buying from.” The second and third-order effects would cripple e-commerce, open banking and logistics, all riding the same cloud backbone. That’s the sovereignty point. You can ban a brand. You can’t ban the capability of renting a server and encrypting your own traffic, not without taking modern commerce down with it.

• 
  Command Code (@CommandCodeAI) reported 25 minutes ago

  We're aware of an ongoing incident. Partial outages at Cloudflare and Supabase are causing Command Code CLI to experience intermittent connection issues. We're actively investigating and working on a fix. Thanks for bearing with us.

• 
  Kolar😎 (@Kolar_Dev) reported 26 minutes ago

  As a product builder, avoid putting your entire infrastructure under a single provider. If your database is running on an EC2 instance, keep backups somewhere independent, such as Cloudflare R2. The goal isn't just redundancy, it's leverage. No single provider should be able to take your product offline, lock you out, or put your business at risk with a single outage or account issue.

• 
  Sulabh Puri (@sulabhpuri) reported 29 minutes ago

  A lot of problems with @Cloudflare today.

• 
  Adam Rush 🛫 (@Adam9Rush) reported 34 minutes ago

  @JamesSherlouk Hosted on R2 @Cloudflare, zipballs each dependency and mints it so it can never change. More control, mainly, we can just manage it pretty cheaply, but ultimately do some other stuff with it. I would quite like to build a little internal dashboard that shows our dependency graph, etc.

• 
  Ryan K 🌥 (@Yank) reported 37 minutes ago

  @tebayoso @Cloudflare Sorry to hear that. Do you have a case or ticket number from support?

• 
  starmex (@starmexxx) reported 39 minutes ago

  WHY WASTE 16 MINUTES OF YOUR TIME ON THIS AI ENGINEER EUROPE TALK WHEN I CUT THE 5 BEST MOMENTS INTO 4 MINUTES FOR YOU bright data engineer exposed why your ai agent lies about searching the web. cloudflare blocks 20% of web from ai. 60% of chatgpt citations are broken. agents hallucinate instead of saying "i can't" 00:00 - llms are programmed to please. they make things up instead of saying "i can't" 00:42 - cloudflare blocks 20% of web. 60% of chatgpt citations are broken 02:03 - gpt-5 fails all 5 web tasks without proper tools. zero out of five 02:42 - cloudflare labyrinth feeds ai fake data. bigger hallucinations 03:13 - don't parse with llm. build a parser. saves 99% of tokens bookmark this and watch the supercut below

• 
  marcelo mezquia (@IntentSim) reported 40 minutes ago

  Stopping the bad guys with Cloudflare: 15,448 malicious requests blocked or challenged in the last month #cloudflare

• 
  Tyler Mayberry (@tyler_mayberry) reported 41 minutes ago

  Agents like Codex really unlock things that I never would have done myself. I've been thinking about migrating all kinds of things to Cloudflare in the past, but it would take all day of just tedious admin work to take care of it. Today I just handed it over to Codex and it was all done in a couple of hours without me touching anything.

• 
  Agrit Tiwari (@agrit_tiwari) reported 42 minutes ago

  Worlds are a good concept but we need a bridge built by both sides of @vercel and @Cloudflare world. Even workflows has a long PR pending just to support the wasm runtime for running workflow SDK on Cloudflare. But they are continuously innovating with frontier stances on agent stack.

• 
  Nithish Rajan (@thenithishrajan) reported 44 minutes ago

  @Cloudflare @CloudflareHelp I urgently need: 1. Reason for the freeze 2. Exact verification or action required 3. Renewal restored, or expiry protected during review Any further delay will directly impact our business operations. Please treat this as utmost priority.

• 
  UWillC (@uwillc) reported 47 minutes ago

  Half the internet blinked last week. The cause was a backhoe, not a model. June 22. A fiber cut on Zayo routes rippled into Cloudflare. X, Reddit, Zoom, Teams. Down. X alone passed 30,000 outage reports before most services recovered in about 20 minutes. Every AIOps dashboard in those companies watched a problem none of them could fix. You cannot reroute around a cut you do not own. You cannot ask an agent to splice glass three states away. We keep automating the control plane. The physical plane stays one excavator from an outage. Your multi-cloud is a logical diagram. Underneath it is often a single carrier. An AI can monitor the fiber. It still cannot splice it. Your redundancy on paper: single-carrier underneath, yes or no?

• 
  world Evolution (ALT) (@EvolutionALT) reported 51 minutes ago

  4/8 Popular Repos (2026): yonggekkk/Cloudflare-vless-trojan (very active, has obfuscated versions) Surfboardv2ray/Trojan-worker & v2ray-refiner vfarid/v2ray-worker derivatives These support Trojan over WS+TLS on 443.

• 
  Special Situations 🌐 Research Newsletter (Jay) (@SpecialSitsNews) reported 52 minutes ago

  New Activist Name: Shares of $MTN are trading up 13% at $141.65 on Thursday, rebounding sharply from their 52-week low of $118.51 hit earlier this year, as the Semafor scoop circulates across trading desks. The intraday move lifts the company's market cap to roughly $5.05 billion. According to Semafor, Vail's bankers are tasked with assessing vulnerabilities across a broad front: labor unrest, weather-related demand swings, and the specific pressure campaign being waged by Prince, who co-founded Cloudflare (NET). Prince told a local Colorado publication in June 2026 that he is willing to invest $500 million in Park City Mountain Resort and admitted he has already fielded calls from activist investors probing Vail's weaknesses. His preferred blueprint would see Vail pivot to an asset-light model, acting as a partnership facilitator rather than a direct mountain owner, a structure that would almost certainly require carving up the company's core real-estate holdings. The timing is awkward for management. Vail reported fiscal Q3 2026 earnings per share of $8.81, missing the consensus estimate of $9.09 by 3.1%, while revenue of $1.21 billion came in roughly $10 million below forecasts. The company subsequently cut its fiscal 2026 net income guidance to a range of $128 million to $162 million and trimmed Resort Reported EBITDA guidance to $735$755 million, down from the prior range of $745$775 million. Net debt has climbed to $2.65 billion from $2.24 billion a year earlier, pushing net leverage to 3.5x trailing twelve-month EBITDA as of April 30, 2026, while cash on hand stood at $371.4 million. Into that environment, the board moved in May 2025 to recall Rob Katz, the executive who originally built Vail into a multi-mountain empire, ousting his hand-picked successor in the process. Katz has since focused on the operational grievances that drove customer dissatisfaction, particularly lift-line congestion and chronic labor shortages, introducing products like Epic Friend Tickets and discounted super-advanced lift tickets that are showing early traction. The move signals that Vail's board views operational credibility as its first line of defense against any activist pitch centered on mismanagement. Management also has a financial lever to highlight in any proxy fight. The company pays a quarterly cash dividend of $2.22 per share, with the next payment scheduled for July 9, 2026, equating to an annualized yield of roughly 6.6% at current prices. That yield argument, steady cash returns while the turnaround plays out, is a standard defensive talking point, though it carries less weight when leverage is rising and guidance is being cut. Investors will get a clearer read on whether Katz's operational fixes are gaining traction when Vail reports fiscal Q4 2026 results, tentatively scheduled for September 24, 2026. The setup is challenging: consensus EPS for that quarter stands at -$5.05, with eight analyst downward revisions in the past 90 days and no upward revisions, reflecting the structural headwinds Prince and any allied activist would likely exploit.

• 
  max guy 😐 (@GolerGkA) reported 52 minutes ago

  @artillain @ThePrimeagen Ok I’m stupid bear with me. Usually with cloudflare on front of static public site, users don’t hit my web service most of time anyway, they hit cloudflare cache. Does it still work? I assume that information that anybody would want to scrape would be on static public endpoints.

• 
  JustAnotherUser_4 (@JAU_4) reported 56 minutes ago

  @EddCoates Cloudflare firewall, thank me later. I blocked entire countries. Solved so many problems.

• 
  seanvfacer (@seanvfacer) reported 56 minutes ago

  Bots just beat humans on the internet. For the first time in history. Not coming. Already happened. Cloudflare — the company running 1 in 5 websites on earth — watched the moment it tipped. The old internet was built for people. The new one's built for agents that don't browse, don't linger, don't even see your sign. So if you're building anything in 2026 — your customer might not be human anymore.

• 
  Brandon Bedford (@wealthpotion) reported 57 minutes ago

  "Claude is down" is the new "Cloudflare is down" except way worse.

• 
  kocer (@kocer_eth) reported 58 minutes ago

  7 FREE AI API/TOOL TIERS YOU CAN USE TODAY BEFORE BUYING ANOTHER AI SUBSCRIPTION If you build agents, bots, research tools or small automations, start with this stack. 1. OpenRouter Use it as the router. It exposes free-priced models in the model list, so you can test routing before paying per token. 2. Google AI Studio / Gemini API Good for prototypes, evals, long-context tests, and agent experiments. Check the free tier before you burn paid credits elsewhere. 3. Cloudflare Workers AI Best when you want inference close to your app. The useful part is not just “free AI” — it sits inside the same place you can deploy Workers. 4. GroqCloud Use it when speed matters. Great for bots, voice loops, extraction, and any workflow where slow responses kill the demo. 5. GitHub Models Best for prototyping inside the GitHub flow. If your code, prompts, and tests already live there, this removes friction. 6. Tavily Research/search API for agents. Free plan shows 1,000 API credits/month, useful for browsing agents and research bots. 7. ElevenLabs Voice layer. Free plan shows 10k credits/month, enough to test narration, agents with voice, and demo content. > My rule: never build production on a free tier first. > Use free access to test: - latency - rate limits - output quality - tool calling fit - billing behavior - whether your agent actually needs the premium model Then pay only for the part that survives real usage. Most people skip this and buy 3 subscriptions before they even know which API call matters.

• 
  Wallstreet Dragon (@longwashere) reported 58 minutes ago

  DD: Long term holdings. $NET cloudflare and why it's important in the age of agents ELI5: The world is moving towards agent. Big industries need better cloud bot protection, developers need LLM computing on the edge. Cloudflare provides the most afforadable option for both, even heavy aws users are using cloudflare for these purposes. What is Cloudflare? For the technically challenged or pre-med professionals, Cloudflare is a web infrastructure and security company that acts as a protective, performance-enhancing shield between a website and its visitors by providing services like content delivery networks (CDNs), DDoS mitigation, and secure domain routing. TL;DR: For the simple folks, it's that **** that pops up with the CAPTCHA to make sure you're not a bot. For developers, it's that **** that makes your sites fast and secure from bots. What is Cloudflare's growing revenue? Application Security and Content Delivery Network (CDN). What is a Content Delivery Network? A CDN is basically a network of servers used to store files closer to its users for faster retrieval. Imagine an app creates a backend database storing all its images on AWS based in US-East. A CDN will then copy the most commonly used images in that S3 database and duplicate them across multiple regions (Asia, Europe, US-East). When an app makes a service request, it will make the request to Cloudflare first. Cloudflare then uses its internal logic to determine if the data needed is in a nearby Cloudflare edge server (on the edge) or if it needs to get it from the main database in US-East. This is called storage on the edge. This CDN mechanism is a relic of Web 2.0, but it will become significantly more important in the age of AI. Now, instead of storing images, large AI providers will be storing entire LLM contexts on the edge. So instead of training specialized ML models to do a specific task, app companies can use a general-usage LLM with a stored context for that specific task, and it will be fast, too. This mechanism is called Prefix Caching or Prompt Caching. By doing this, it makes the LLM responses almost instantaneous. So all your consumer apps that use LLMs—like CALai, Duolingo, Grok, etc.—are most likely already using this process. Beyond simply storing data on the edge, the industry is shifting toward deploying entire servers and specialized AI models locally. A major component of this architecture relies on LLM routing. Instead of hosting massive, resource-heavy models on every single edge device or regional server, companies are deploying highly optimized, lightweight router models at the edge. These local routers analyze incoming user prompts to determine the most efficient way to handle them. If a task is simple, the edge model processes it instantly to minimize latency and eliminate cloud compute costs. If the task requires deep reasoning or a massive knowledge base, the router intelligently forwards the request to a larger cloud-hosted model. Additionally, these edge routers leverage tool calling, which allows them to execute local APIs, query regional databases, or trigger specific code workflows without needing to round-trip back to a centralized data center. Moving from simple edge storage to localized edge intelligent compute represents a massive paradigm shift. It allows enterprises to scale AI applications efficiently, safeguard data privacy, and drastically slash infrastructure costs. Cloudflare Security in the Age of Agents This one is simple. You know that Cloudflare CAPTCHA that pops up when you're entering a website or checking out with a credit card? Websites PAY for that CAPTCHA. And they pay a lot. These features block spam, bots, and DDoS attacks. When you move your mouse to click the CAPTCHA, Cloudflare uses proprietary logic that determines you're human by calculating how fast your mouse moved, the angle you moved it, how long you waited, and any other actions you took. Sometime in the 2010s, every website figured out that paying for this small puzzle CAPTCHA was more cost-effective than getting DDoS'd by bots, so almost every single site adopted it. The CAPTCHA is only one of Cloudflare's products in its security suite to block bots from websites, but the overarching theme is the same for all its features: blocking bots. Well, it's 2026 now, and web traffic across the board has increased, mostly driven by AI and AI agents. Automated web traffic has increased by 600% in 2026 alone. Guess who is positioned perfectly for this? Cloudflare. Not only is Cloudflare blocking bot traffic, but it's also getting paid by them. Cloudflare is releasing a new product (Pay Per Crawl) that allows website owners and Cloudflare to get paid for LLMs crawling their content. Cloudflare is simply winning by creating the gates for web traffic and now charging a toll fee for bots to use them. Cloudflare is direct play on internet traffic, which is a correlating play on ai agents and LLM adoption and usage. If you think people will continue to use ai agents and LLM, then cloudflare is your guy. Cloudflare valuation has dropped recently because of the layoffs due to ai, even though revenue has sped up. This drop was more of emotional sell off than a fundamental one. It's valuation has already bounced back. (Cloudflare is trading at 235 as of this post, I bought in earlier in the 190s for a swing trade after the bogus layoff dips, wish i bought in more)