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 (33%)
• Cloud Services (30%)
• Web Tools (15%)
• Hosting (15%)
• E-mail (7%)

Live Outage Map

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

City	Problem Type	Report Time
Paris	Cloud Services	8 days ago
New York City	Hosting	10 days ago
Manchester	Domains	1 month ago
Angers	Cloud Services	1 month ago
London	Domains	1 month ago
Noida	Hosting	2 months 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:

• 
  leepy (@leepy_cat) reported 11 minutes ago

  codex has an an in app browser now and that is cool but the problem is the agent decides to try to make me log in to the cloudflare webpage then manage service with its ghost mouse instead of just use the fking CLI im auth'd into already

• 
  TezaApps (@TezaApps) reported 13 minutes ago

  @idunnstudio Good catch running the audit anyway. Stripe's restricted keys are the other half of that fix, a leaked key that can only read is a bad day instead of a disaster. Mine lives only in a Cloudflare Worker so the client bundle never even sees it.

• 
  Zaher 🦋@zaher.dev (@zaherg) reported 13 minutes ago

  @Cloudflare Found the issue, I’ve deployed the project as static assets

• 
  Nik (@TheRealNikR) reported 15 minutes ago

  @galluzzo_julian I just migrated 2 client sites from Webflow to Cloudflare + AstroJS. It really makes no sense anymore if you're client is full service. Webflow is too slow and their implementation of the MCP was terrible, at least version 1.0

• 
  Sumanth (@Sumanth_077) reported 17 minutes ago

  Turn any website into agent-ready data! Loop engineering is about designing systems that run agents autonomously. Instead of prompting your agent manually each turn, you write a loop that finds the work, hands it to the agent, checks what came back, and decides what happens next. Your job is to design the loop once and walk away. But loops that need live information from the web hit a real constraint. JS-heavy pages return empty content. Anti-bot systems return challenge pages. Login walls block access entirely. When the model gets weak context back, it still responds - just less accurately. Anakin is building the source-access layer underneath agents. URL Scraper turns any URL into clean Markdown, HTML, or structured content immediately usable by an LLM. Built for scale across 200M+ active websites globally, including a large chunk of Cloudflare and Akamai-protected pages. Authenticated sessions handle content behind login walls. Wire handles workflow-heavy sites. Login flows, navigation, form submission, report exports - all accessible through a stable API. Define the workflow once and Wire keeps it working as websites change. Key capabilities: • Clean Markdown, HTML, or structured output from 200M+ active websites • Built for difficult pages including Cloudflare and Akamai-protected sources • Authenticated sessions for content behind login walls • Wire for login flows, navigation, form submission, and export-based access • Useful for AI agents, RAG systems, finance intelligence, and vertical AI workflows I've shared the link in the replies!

• 
  Tanner Powell (@TannerPowell) reported 17 minutes ago

  @jackfriks @levelsio Little bit of setup involved but if you can use One Password to make a read-only folder with your env stuff, setup access token for Claude, login to wrangler (cloudflare) or vercel/supabase’s CLI tools, it really does become mostly English fast. “Explain this to me more simply.”

• 
  Sushrut Bidwai ~ ॐ (@sushrut_bidwai) reported 17 minutes ago

  Is @Cloudflare verification working on brave browser? I am seeing far too many verification failed errors.

• 
  Kephen Rai (@kephen20936) reported 21 minutes ago

  @jallen_dev I actually used a temporary SQL Database to record events, rather than socket-based handling, I used up almost 2.8Million Reads and over 74k writes in like 10 minutes between two browser sessions for a Tank game. Worst idea ever, I was out of Bandwidth on Cloudflare for 2 weeks.

• 
  KGNINJA (@FuwaCocoOwnerKG) reported 22 minutes ago

  I implemented and deployed a D1-backed job queue for my local Gemma workflow. The architecture now works like this: Store prompts in Cloudflare D1 before shutting down. Launch ChatGPT Work later. Fetch pending jobs from D1. Run local CPU inference with Gemma 3 4B (llama.cpp). Save the results back to D1. ChatGPT Work acts as an on-demand inference worker, while Cloudflare D1 provides persistent job storage and state. This isn’t a persistent inference server. Instead, it’s a resume-on-launch architecture that survives PC shutdowns and resumes processing the next time ChatGPT Work starts.

• 
  Ara T. Howard (@drawohara) reported 23 minutes ago

  today, with the help of 3 agents, I managed to create a cloudflare API token JFC

• 
  jeli beli (@nftgamingnoob) reported 23 minutes ago

  @cooldown_sol @Pumpfun @Cloudflare damn

• 
  MediaWizard (@mwzd) reported 24 minutes ago

  @Domainji I can use dns service, not just cloudflare, would rather do it at the registrar level. Don't use name .com anymore, so doesn't matter to me atm.

• 
  Colin Son (@txmedai) reported 25 minutes ago

  @ptremblay Yes Cloudflare is very good. And free tier is amazingly generous. Isnt quiet idiot proof (why do I have independent turnon AI gateway or R2 or all these individual features) but at least there’s not a million layers of IAM. GCP and AWS are fine

• 
  Ozzy 🇺🇸 (@MEADGod) reported 29 minutes ago

  Been running around all day trying to find a solution with @Cloudflare Ended up moving entirely. Sorry about the hiccups I will go ahead and fix a few more bugs - this is totally unacceptable and I apologize for that Back to business

• 
  Jose Valerio (@josevalerio) reported 29 minutes ago

  .@gumclaw take down **** please I leaked my AWS keys in the frontend repo build need it down before it gets cached by cloudflare

• 
  Jono Young (@chsweb) reported 30 minutes ago

  @CloudflareHelp love you. No help needed. Maybe I should send to @CloudflareLove … which even more cloudflare than I thought.

• 
  Gavin Keulks (@GavinKeulks) reported 32 minutes ago

  @DrDavidVernon @britishlibrary Thank you for those kind words. The site was severely damaged by a server upgrade at my university, and for years I thought it was gone forever. Thankfully, repair (albeit tedious) has been possible. Now I've taken full ownership of all aspects of the site, so if it's damaged again, it will be because Cloudflare accidentally shut down the internet again (as it did a few years ago).

• 
  Chinmay Purav (@chinmay_purav) reported 33 minutes ago

  Hey @Cloudflare , Please bring in support for .IN TLD domains!

• 
  Ben Hultin (@ImBenHultin) reported 39 minutes ago

  Keeping Costs Down During High Traffic (Vercel vs. AWS Serverless) Don't let your Vercel serverless function bill surprise you during a sudden traffic spike. Serverless platforms make launching effortless, but unbounded function invocations and edge request overages can quickly lead to unexpected monthly bills. How to fix it: 1. Set explicit execution time limits, token budgets, and concurrency caps on your serverless endpoints. 2. Offload heavy computations or batch data directly to AWS S3/Kinesis using direct service proxies rather than running proxy code through edge/Lambda functions. 3. Implement hard spending alarms and rate limiting (e.g., Cloudflare or API Gateway quotas) to block adversarial or infinite recursive agent loops.

• 
  Barbog Da Big Finka (@BarbogDaFinka) reported 40 minutes ago

  In my endeavour to see just how little I can do, I started a new project. Sol has worked diligently to shill CloudFlare to me, opening a oauth sign in to CloudFlare, and eventually haulted work till I swiped my credit card. It has now successfully deployed a game behind a domain. The game is horrible, but I'm impressed it did all this in about 5 hours.

• 
  Chinmay A. Singh (@chinmay) reported 41 minutes ago

  @India_Policy Centralization is almost always bad. Be it governance (king), technology (cloudflare outage), identification (Aadhar in India or Social Security Number in the US), or exams (NEET).

• 
  Chris | Zestio (@ZestioAI) reported 43 minutes ago

  Day 7: worked on our Reddit marketing pipeline today. the VPS IP is blacklisted at Cloudflare level - every request, dead. Now switching to a search-backend proxy approach instead. sometimes the problem isn't your code, it's literally your IP address 🤡

• 
  Honey Syed (@honeydreamss) reported 48 minutes ago

  Someone reverse-engineered how ChatGPT's web tool works and published a method for forcing it to crawl your own site. Paste a structured query into ChatGPT, it searches your domain, opens the top results, and tells you exactly what it extracted. Most companies have never done this. When they do, one of three things tends to happen. Some discover ChatGPT can't access their site at all. Their robots.txt blocks GPTBot or ChatGPT-User without anyone realizing it. The dev team or hosting provider set it up to block bots generically and caught every AI crawler in the process. Some discover ChatGPT can access the site but reads it wrong. Outdated pages, wrong product descriptions, missing pricing, content behind JavaScript that renders in a browser but looks blank to a crawler. Some discover ChatGPT reads their site fine but recommends a competitor anyway. The competitor's content is structured the way AI prefers, while theirs reads like marketing copy that sounds good to humans but gives AI nothing concrete to cite. All three scenarios are the same outcome. A business losing customers to a discovery channel they don't know exists. The deliverable is a GEO audit, Generative Engine Optimization. You audit how AI systems see, read, and cite a business's online presence and deliver a report showing what's broken and how to fix it. Five layers. Crawler access. Check robots.txt for AI crawler blocks. The most common and most fixable problem, many sites do this without knowing it. Also check server logs for AI crawler visits. If the client uses Cloudflare, the AI Crawl Metrics dashboard shows bot activity directly. Content extraction. Run the ChatGPT crawl trick, force it to search the client's domain, open top pages, report what it extracted. Also check whether content loads dynamically via client-side JavaScript. AI crawlers can't execute JavaScript the way browsers do, so dynamic content often appears blank. Brand citation audit. Ask ChatGPT, Perplexity, and Google AI Overviews: "What's the best [product type] for [use case]?" "Who are the top [service providers] in [city]?" "What are the alternatives to [competitor]?" Document whether the client is cited and who gets recommended instead. Seeing ChatGPT name a competitor while omitting them entirely is the moment a client understands why this work matters. Content structure analysis. Review key pages for AI readability. AI systems prefer direct answers in the first paragraph, specific data and statistics, headings that match how users phrase questions, and schema markup. Third-party presence. Check Reddit, review sites, and industry forums. Reddit remains one of the most heavily cited third-party sources across AI platforms, a client with no Reddit presence is missing a major signal source. Package the five layers into a report. What the problem is, why it matters for AI visibility, the specific fix. Prioritize by impact. Price at $1,500 to $3,000 depending on the size and complexity of the client's web presence. A local business with a 10-page website pays $1,500. A SaaS company with 200 pages of documentation and multiple product lines pays $3,000. Find clients on LinkedIn. "Head of Marketing," "SEO Manager," "VP of Growth," "CMO" at companies already investing in SEO. If they care about being found on Google, they'll care about being found by AI once they understand the gap. Comment OUTREACH if you want to see the exact email I would send them. Upsell the fixes. The audit is the entry point. Implementation, Updating robots.txt, restructuring content, adding schema markup, building Reddit presence, creating machine-readable FAQ pages, is a separate engagement at $3,000 to $10,000. Or a monthly retainer at $1,000 to $2,000 to continuously monitor AI citations. GEO as a category barely exists. Most SEO agencies haven't added it. Most businesses don't know it's a thing. The consultants building this practice now will set the pricing, own the case studies, and establish authority before anyone else figures out this is a service.

• 
  TENJI (@ineedtendies) reported 49 minutes ago

  Stayed up late into the night fixing bugs and making sure launch ran as smoothly as possible. Everything was stable when I finally got some sleep, but I woke up to reports that some users are having trouble connecting to the site. We recently added Cloudflare protection to the domain, and I’m already looking into whether that’s causing the issue. I’m aware of it and working on it now. Sorry to anyone affected. Stay tuned for updates.

• 
  RedPocatto (@RedPocatto) reported 50 minutes ago

  @PirateSoftware strange - microsoft teams had australia wide problems too today - amazon or cloudflare problem i wonder?

• 
  Kevin Gray (@graykevinb) reported 53 minutes ago

  There are a ton of open fall internships rn. You should go apply. Google jobs sucks. Check companies directly. Also checkout handshake. Cloudflare and Tesla are hiring fall interns. You should apply ASAP. I will as usual be skipping this round but if you want help with your resume lmk. I have a somewhat decent success with applicant tracking algorithm manipulation. If you need money some are even 40hrs a week and pay $50/hr

• 
  teo (@synoet) reported 53 minutes ago

  @CherryJimbo @Cloudflare great site, wish it existed a year ago when I picked up durable objects. one more thing on DOs: the system clock is frozen except across I/O, so you can't get consistent timing or traces inside an object, making debugging perf issues basically impossible

• 
  max.berlin (@maxjendrall) reported 56 minutes ago

  @vibecodeguild You can! You can put out public pages haha on custom domains even. Goddmanit, they need to fix the presentation. They literally rolled out cloudflare workers across their customer basis that are soooo powerful

• 
  AntSeed (@AntSeedAI) reported 58 minutes ago

  We summarized all the replies - here are the results 👇 Why people use a gateway other than OpenRouter: • 25% Vercel AI SDK ecosystem fit • 20% Cheaper / lower fees • 13% Already in their stack (Cloudflare, LiveKit, Cursor) • 10% Direct-to-provider speed & caching • 10% Self-hosting / control • 8% Missing or slow models • 7% Privacy / data residency (ZDR) • 7% Trust & rebrand gripes Antseed resolves all of them.

• 
  DFIR Radar (@DFIR_Radar) reported 1 hour ago

  Microsoft's Q2 2026 email threat roundup: 7.6 billion phishing threats detected, Tycoon2FA volumes collapsed 92% post-disruption, and Teams-based vishing hit 10x the mid-2025 baseline. Key findings: - Tycoon2FA never recovered after March's DCU-led infrastructure action. Monthly volume fell from a 15.1M baseline to 1.2M by June, an 92% decline. The platform shifted to .RU TLD hosting after losing Cloudflare, but its share of CAPTCHA-gated phishing dropped from 76% peak to 12%, and QR code campaign share fell from 33% to 14%. No replacement PhaaS emerged at scale. - Teams vishing is the sharpest growth vector this quarter: weekly malicious call attempts rose 80% since January 2026 and now run at nearly 10x the mid-2025 baseline. Attackers impersonate IT helpdesk, peak 14:00-20:00 UTC Mon-Fri, and increasingly use generic display names (52% in June) rather than obvious IT branding, likely blending with ClickFix-style lure themes. - Two notable campaigns: a June 1 BEC wave hit 67,000 users across 42,000 orgs in under 3 hours using Python email.mime, Amazon SES API, DKIM-signed ecajovna[.]sk, and reply-to domains ilyff[.]com, j-gmails[.]com, x2mails[.]com. A June 14-15 campaign hit 107,000 users using nested EML, ICS calendar invite, a silent OAuth redirect through login.microsoftonline[.]com, and a BAT dropper (Financial_report.bat) pulling installer.exe from pixeldrain[.]com. #DFIR_Radar