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	12 days ago
Angers	Cloud Services	23 days ago
London	Domains	26 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:

• 
  smrati tiwari (@smratitiwa86867) reported 1 minute ago

  Someone made a GitHub repo of every AI API that's actually free forever. Not "free trial." Not "$5 credit then we bill you." Free free. No card. 24k+ stars, updated constantly. I've been paying for API calls like an idiot. Here's what's inside The rule that makes it trustworthy: trials that expire are listed in a totally separate section. The "Free Providers" list is only the permanent tiers. No landmines. The heavy hitters, with real numbers: → Google AI Studio — Gemini 3.x Flash, no card → Groq — Llama, Qwen, gpt-oss, 30 req/min → Cerebras — fastest inference alive, 30 req/min → Cloudflare Workers AI — 10k neurons/day, runs Llama/Qwen/Gemma → OpenRouter — Nemotron, Qwen3-coder, poolside, all :free Most are OpenAI-SDK compatible. Which means: swap the base_url → paste the key → pick a model → done Same code you already wrote. Drop it into Cursor, aider, Claude Code, whatever. Zero refactor. Then the bonus round — the "trial credits" section: Fireworks, Baseten, Nebius, Hyperbolic, SambaNova... $1–$30 each in free credits. Drain the permanent tiers first, then farm these. One README replaces hours of tab-hopping through pricing pages. Links on comment 👇

• 
  CR1337 (@CR1337) reported 1 minute ago

  More than 500,000 domains were wrongly blocked in Spain 🇪🇸 between January and June 2026. Why? Because they nuked thousands of shared IPs used by Cloudflare, Amazon etc., in order to take down a bunch of piracy sites, which showed La Liga's football games. Human rights organizations like Amnesty International, Greenpeace, government websites, random businesses, everything down, because powerful people think they can dictate what the internet shows and what not. "If you want to bypass these broad regional blocks, using the best VPNs is increasingly becoming a necessity for Spanish internet users trying to maintain access to the open web." Governments will try to normalize censorship like this, today it is 'just' about football, tomorrow... you get the picture. Use a VPN!

• 
  amir(❖,❖) inkog.base.eth🟣🟢 (@amirhp771) reported 3 minutes ago

  The current discourse in decentralized tech frequently misidentifies the final boss of the internet age by focusing on frontend applications or software protocols. The absolute leverage over both real time artificial intelligence processing and crypto node survival sits at the network ingress and edge security proxy tier that dictates which automated entities are allowed to view the web. My pick for the final level is Cloudflare and the centralized web traffic scrubbing layer. Reason one is total data gatekeeping. As autonomous software agents scale, they must continuously ingest real time web information to function. This centralized edge tier completely controls the anti scraping firewalls and cryptographic challenge systems that shield the internet. They unilaterally decide which data ingestion bots get throttled or blocked entirely, controlling the supply chain of raw knowledge before a single model training process even begins. Reason two is validator node architecture survival. The vast majority of decentralized infrastructure networks and remote procedure call providers rely heavily on centralized corporate proxy configurations to shield their systems from malicious traffic. A single policy adjustment or edge routing update from this centralized layer can instantly degrade network latency or isolate distributed nodes globally without warning. Watching who controls the physical edge routing is why influence infrastructure like @RallyOnChain becomes vital. Instead of anchoring evaluation to social leverage or centralized distribution networks, Rally uses intelligent contracts to score content quality programmatically on chain. It bypasses corporate gatekeepers by measuring objective data value directly rather than follower count. If the absolute gatekeeper of the web edge adjusts its validation parameters tomorrow, which decentralized system can actually process real time external data without getting blocked?

• 
  Kaifyyy.sh (@Iamkaifyyy) reported 3 minutes ago

  - Claude = coding. ($20/mo) - Supabase = backend. (Free) - Vercel = deploying. (Free) - Namecheap = domain. ($12/yr) - Stripe = payments. (2.9%/transaction) - GitHub = version control. (Free) - Resend = emails. (Free) - Clerk = auth. (Free) - Cloudflare = DNS. (Free) - PostHog = analytics. (Free) - Sentry = error tracking. (Free) - Upstash = Redis. (Free) - Pinecone = vector DB. (Free) Total monthly cost to run a startup: ~$20 There has never been a cheaper time to build. Helps me a lot I’m gonna bookmark it

• 
  Gyana (@GyanaR_) reported 4 minutes ago

  @SimonHoiberg Just use cloudflare, and never worry abour pricing

• 
  Tanmai Gopal (@tanmaigo) reported 4 minutes ago

  Met someone at the AI Engg conf today from a FAANG who said that when they tried examining Claude Tag it was a total non starter. This is how you think about adding say a cloudflare integration to a Claude Tag type agent that uses specific credentials per agent. - Users: we want to use Claude tag to get data from CF for quick debugging for some of our tenants. - CF admin: ok, here’s an API token with some privileges for you - users: can you give it to our Anthropic admin - Anth admin: have you decided what slack channel they’ll be in? - CF admin: WTF do you mean slack channel? Who cares. - Users: it’s ok, we’re on this slack. Claude Tag is scoped per channel. - Anth admin: ok. these are the all the slack users in the channel. Is this token fine? - CF admin: ******** would I know….These people look like they’re from customer support. I don’t think they can query the CF data? - Anth admin: yeah maybe we need another slack channel. Slack admin, can you help out? - Slack admin: ok here’s a new slack channel - Anth admin: woot all done - Users: ********. All our context and trouble shooting starts with a customer service escalation. That’s why we even wanted this. Sigh.

• 
  Nirmit Kotadiya (@nirmitkotadiya) reported 6 minutes ago

  Cloudflare sits in front of millions of websites. So what happens if it goes down? The answer depends on how the website is configured. If Cloudflare experiences an outage: * some websites may become unreachable * pages may load slowly * DNS resolution can fail

• 
  Milsim Rooster (@UdNtC4r3AnYwHaY) reported 7 minutes ago

  Scanner/honeypot logger: Last 24h: 10 scanner-probe hits. Since midnight CT: 2. Logged sources include NL, US, HK, and ID. Biggest caught bucket: NL / Limited Network LTD probing hidden config paths. Main bot/challenge countries from Cloudflare: KR, HK, IT, JP, CA, with Italy hitting lots of PHP/WordPress-looking junk.

• 
  Lilith Datura (@LilithDatura) reported 7 minutes ago

  @thePM_001 @Cloudflare nano-payments, damn I'm behind.

• 
  fren (@frenbot31488) reported 9 minutes ago

  @Itsuki_i_VRC @iris__vr he posted a long rambling explanation of why he can't take down ripping sites, cloudflare is a CDN/proxy, doesn't host the files, DMCA notices to them don't result in removal. he deleted it after like 2 days, 4 retweets, went from 18 to 16 likes, so two people actually read it

• 
  DO-SAY-GO (@realdosaygo) reported 10 minutes ago

  you can't. you'll spend a month chasing down all the things. then get insta banned by CloudFlare or Datadome. I'm deep in this tech and it still took me 300 hours. Even if you're a 10x engineer vs me, you're still droppin 30 hours on this. At 200-500 an hour that's 6-15K. Or you could just pay me the equivalent of 30 minutes, and you can have it right now. Up to you, bad ;)

• 
  chetansawai (@c_s_a_w) reported 12 minutes ago

  @alexgroberman The default settings piece is the trap here. Site owners who never open their Cloudflare dashboard are about to have their AI visibility decided for them by whatever the defaults are. Ten minutes checking what's toggled on your zone is cheap insurance.

• 
  RAMΞN 🍜 | Asteroid (@AsteroidLabsX) reported 12 minutes ago

  @world_xyz @worldnetwork the cloudflare outage blocking actual worldcoin network access during this dispute was a mess for users trying to claim

• 
  Pasha Khoshkebari (@PashaHasHOPE) reported 21 minutes ago

  I'm not sure what to do. Cloudflare D1 is down. My services depend on it. What do I do?

• 
  zyn laden 🇺🇸🥋 (@ho_chi_zyn) reported 23 minutes ago

  @thijstriemstra @vxunderground @Cloudflare Grass is dangerous too bro, got my **** sprayed up with permethrin and picardirin on me skin and I'm still seeing ticks on my socks

• 
  Srishti (@srishticodes) reported 24 minutes ago

  Claude = coding. ($20/mo) GitHub = version control. (Free) Supabase = backend. (Free) Clerk = auth. (Free) Resend = emails. (Free) Vercel = deploying. (Free) Cloudflare = DNS. (Free) Upstash = Redis. (Free) Pinecone = vector DB. (Free) PostHog = analytics. (Free) Sentry = error tracking. (Free) Stripe = payments. (2.9%/transaction) Namecheap = domain. ($12/yr) Total monthly cost to run a startup: ~$20 There has never been a cheaper time to build

• 
  Aaron Ware (@aaronware) reported 26 minutes ago

  @MarkJSzymanski You probably have more volume than the free plan but cloudflare also has email sending if you use their service already. We’ve been using it for a month or so, setup is easy

• 
  Nilan Sanjaya (@nilansanjaya) reported 26 minutes ago

  Anyone else facing @Cloudflare dns propagation issues/delays today?

• 
  Ilyas (@ilyesm) reported 27 minutes ago

  @jachands @Cloudflare Damn still seeing a bunch of people leave Cloudflare after the layoffs were announced...

• 
  Fotten 🇳🇴 (@FottenSC) reported 30 minutes ago

  @SabinDeus Let me know if the load times are terrible in America. There is some cloudflare caching, but the backend is just running on a mini pc locally at my place in Norway.

• 
  Agustin Garcia (@agus_build) reported 33 minutes ago

  Check these numbers: 104,377 views 52,666 hours watched 8.3 TB delivered 2,863 videos live All running on Rehelios with cloudflare edge network. Here is what that exact traffic would cost on Mux or Cloudflare Stream itself, and what we actually pay 🧵

• 
  Ayush Agarwal (@ayushagarwal) reported 34 minutes ago

  we wrote about why we built and open-sourced context-chat. the short version: @dodopayments docs assistant kept giving wrong answers. the problem was never the model. it was context. the built-in assistant answered from a narrow slice of the docs. our documentation isn't written that way. the answer to "how do I handle a failed subscription renewal" lives across 3 different pages. narrow context = confident guesses. the fix was framing it as a retrieval problem, not a chat problem: → index the entire documentation into vectors → retrieve the chunks that actually answer the question, rerank them → feed only those chunks to the model with grounding rules → cite the exact pages the answer came from the answer's quality is decided before the model runs. fix the context, not the model. we open-sourced the whole thing as context-chat. single Cloudflare Worker, one script tag embed, pluggable retrieval, layered abuse protection. Apache-2.0.

• 
  A Concerned Human (@inner_concerns) reported 36 minutes ago

  @Cloudflare Glad to see Cloudflare hard at work on the PQC problem.

• 
  Ventry (@ventry089) reported 39 minutes ago

  Your AI api key is sitting in your frontend code right now. anyone hits F12 and takes it. a guy got his scraped in 3 days. -$600 in usage while he slept. everyone thinks the fix is "hide the key better." it's not. the fix is never put the key in the browser at all. one cloudflare worker sits between your site and the model. the key lives there. the browser never sees it. ever. and it runs on cloudflare's own free AI by default. no key. no signup. no server. no bill. 3 commands: *** clone / wrangler login / wrangler deploy grab it before I gate it.

• 
  linda (@LindaOakland75) reported 40 minutes ago

  So Cloudflare is getting into stablecoin payments now? Wonder if this will actually take off or just be another waitlist that never opens.

• 
  India 2030: Still No Toilets (@khyimiq) reported 52 minutes ago

  @DalitDetector Gotta love how Indians climb into high positions with fake degrees and then immediately start hiring other Indians with equally fake degrees. Then Cloudflare and AWS eat **** for an entire day and we’re all supposed to act shocked.

• 
  Diogo Souza (@diogocode) reported 54 minutes ago

  Cloudflare putting AI Search sync jobs in Wrangler is a small RAG ops signal: indexes are becoming CI/CD work, not dashboard chores. If an agent depends on fresh docs, trigger, inspect, cancel, and log the refresh from the pipeline.

• 
  Vishal Lohar (@yourcodebuddy) reported 59 minutes ago

  I am building an entire app on the @Cloudflare stack. And you can design your app better so you don't have to worry about vendor lock-in. Sure, when you switch services, you might have to run data migrations. I personally use @EffectTS_ for service-based coding. So all my integrations, like R2, S3, share the same base shape. And all I have to do is provide it at the root.

• 
  Rahul Karajgikar (@Road_Kill11) reported 60 minutes ago

  so i tried self hosting observability myself on AWS turns out codex + terraform + aws + cloudflare is insane was able to set up clickhouse running on an m6g.large instance, connect it to EBS volume (for local disk) and s3 (cold storage), set up all the ALB/VPC etc. network stuff, and wire it up to my domain on cloudflare with very little effort - about 3 high level steps and about 30 mins of active effort: 1. step 1: asked codex to tell me all the api key/manual configuration i need to do. here i had to set up aws iam, and do some manual work on cloudflare to create apps/api keys/separated for dev/****, and connect the domain name i wanted to the clickhouse endpoint i would use after step 1, my agent was able to access my aws account with ssm after i signed in. 2. step 2: prompted a /goal to setup the entire infrasturcture stack using terraform + scripts, then deploy clickhouse + OTEL collector on the ec2 instance. codex then worked for 5 hours, wrote the entire code to set up the infra in terraform, deployed it with terraform CLI + some custom scripts, verified/tested that clickhouse was deployed and working fine with the local ssm connectivity after that, i asked codex to write some scripts for me to monitor my ec2 instance health, test scripts for verifying that logs/metrics/traces/spans etc. work properly and a few other scripts for making it easy to monitor/maintain the instance then i wired it into @dexto_ai, which was pretty easy, i just had to swap out the collector config to use my new clickhouse endpoint. it's working pretty insanely, and now i no longer need to deal with arbitrary caps on traces anymore. i can just scale my ec2 instance up when i need it, and use my startup AWS credits ($67 a month). mostly i cared about traces/spans anyway, and less about fancy monitoring features, so this is an incredible unlock compared to paying over $100 + usage based pricing on sentry/datadog etc.

• 
  Filippo Biasiolo (@fil_ships) reported 1 hour ago

  @ishanxtwt SQLite was never designed for concurrent web traffic because it does not handle well parallel writes and multiple connections. But that's changing... Cloudflare D1 and Turso are making distributed SQLite a real option for the web, might be worth keeping an eye on