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 (40%)
• Cloud Services (29%)
• Hosting (17%)
• Web Tools (10%)
• E-mail (5%)

Live Outage Map

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

City	Problem Type	Report Time
Angers	Cloud Services	3 days ago
London	Domains	6 days ago
Noida	Hosting	18 days ago
Jewar	E-mail	19 days ago
Braga	Web Tools	19 days ago
Noida	Cloud Services	20 days 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:

• 
  Puneet Patwari (@system_monarch) reported 4 minutes ago

  Tweet 3/5 Picking the algorithm is half the decision. The other half is: where do you enforce the limit? Most teams slap rate limiting at the API gateway and call it done. That's one layer. Production systems need at least two. Here's why. Layer 1: API Gateway (the front door) This is your first line of defense. Every request passes through here. Set global rate limits: "no client can exceed 1000 requests per minute." This catches: - Runaway scripts - Misconfigured clients - Basic abuse - Your own batch jobs (ask me how I know) Tools: Kong, NGINX, AWS API Gateway, Cloudflare. All have rate limiting built in. Layer 2: Per-service limits (noisy neighbour protection) You have 10 microservices. Your search service can handle 5000 rps. Your export service can handle 50. Without per-service limits, one client hammering /export at 200 rps takes that service down. And because /export is down, health checks fail, the circuit breaker trips, and suddenly other things start breaking too. Per-service rate limits prevent one endpoint from eating the capacity of another. Layer 3: Per-user limits (fair usage) One user is making 10,000 API calls per minute. Every other user is making 50. Without per-user limits, that one power user is consuming 99% of your capacity. Per-user limits: "each API key gets 100 requests per minute." Fair. Predictable. No single user can starve everyone else. This is also where you differentiate pricing tiers. Free tier: 100 rpm. Pro: 1000 rpm. Enterprise: 10,000 rpm. Rate limiting is literally your pricing enforcement. Layer 4: Per-endpoint limits (not all endpoints are equal) Your /search endpoint can handle 10,000 rps. It's fast, cached, lightweight. Your /generate-report endpoint does a 30-second database aggregation. It can handle maybe 10 concurrent requests before the database starts crying. Same global rate limit for both? That's a disaster waiting to happen. The heavy endpoints need tighter limits. The rule: at minimum, use two layers. Gateway-level for global protection. Per-endpoint or per-user for granular control.

• 
  Agasta (@idkAgasta) reported 6 minutes ago

  @anirudhprmar But you will lose global cdn/ edge network with this .... Also I don't like to host forntend or full stacks in a single vm..... I will consider using Cloudflare workers if the budget is tight or AWS cloudfront + ecs with fargate if we are richy-rich

• 
  inder (@InderpreetSingh) reported 13 minutes ago

  Looks like @Cloudflare dashboard is down, but just saw "Organizations Beta". I hope thats the case. All my projects are co-mingled in a single account right now.

• 
  Jason Fleagle (@jjfleagle) reported 16 minutes ago

  @Cloudflare Fake identity at AI scale turns admissions into an operations problem, not just a fraud problem. The workflow needs signal correlation, escalation rules, evidence packets, and human override.

• 
  Utopian Contributors (@utopiancontrib) reported 17 minutes ago

  We figured out a way to end what seemed like a never ending story: JavaScript bloat This wasn’t Google or Microsoft or Mozilla or CloudFlare or Vercel. It was the Utopian Contributors LLC. But this was just the beginning. The future of the web is ultra-efficient.

• 
  Oluwatobiiloba (@Oluwatobiiloba) reported 17 minutes ago

  It’s the opposite for me. For over two years now any attempt open cloudflare with my Airtel router returns an empty object, mtn works fine. You can download cloudflare one vpn, it’ll help

• 
  Aahan Singh (@aahan_builds) reported 22 minutes ago

  @ajmeese7 @Cloudflare o great, mine went down too. these platforms have been going down way too often lately

• 
  The Vibe Coder (@quantumaidev) reported 23 minutes ago

  Cloudflare works by sitting close to users. Requests can be cached, filtered, routed, or blocked at edge locations before reaching the origin server. It protects and accelerates by moving decisions nearer to the network boundary.

• 
  Lex Tang (@lexrus) reported 25 minutes ago

  I bought a domain on Cloudflare and had Codex enable cloudflare/agentic-inbox for me. Sending and receiving emails is a bit slow. I'm unable to add the account to any email apps. Considering it runs completely free on Workers and can be operated with MCP, these drawbacks can be nothing. It composes a response email draft every time it receives a new email, so I think it's a good fit for feedback emails.

• 
  Corey Quinn (@QuinnyPig) reported 26 minutes ago

  @KhalidWarsa @Cloudflare The trick is to actually be a customer of the things you shitpost about, otherwise it's just noise.

• 
  MBrant75 (@MBrant75) reported 27 minutes ago

  @nickSfishes315 @JackDan110 Heh Really? Cloudflare issue or something?

• 
  Brand3n (@brandenhugheskc) reported 28 minutes ago

  @zacxbtc Meanwhile an hour before SpaceX launch, 1/2 the internet goes down or is experiencing degraded performance. Including Cloudflare, AWS and some financial institutions. 🤔

• 
  Utkarsh (@utkarsh_build) reported 30 minutes ago

  How do you handle latency over the region?? For my SaaS, I have deployed the frontend on Cloudflare Pages, while the backend and database are hosted on a VPS in the Mumbai region. Users outside Asia are experiencing high latency because my server is located in Mumbai. Is there any way to solve this? I know that adding load balancers and deploying across multiple regions would help, but I'm looking for more affordable solutions.

• 
  Nikhil Agarwal (@nikhildp) reported 36 minutes ago

  @CloudflareDev Use at your own risk! No upper limit on how much money you can spend and Cloudflare customer service is quite poor. Avoid any cloud where there is no upper limit on expense, esp Cloudflare!

• 
  Tobias Brida (@TobiasBrida) reported 38 minutes ago

  Quick, calm down — the DSB is probably not going to fine you for using Cloudflare, despite what the loud privacy alarmists want you to believe. Contrary to popular opinion, using a reputable CDN is not a compliance death sentence. Developing cool local tech is a nice cause, but it would be quite nice if it could be achieved without vibe-coded sites that tell you how likely you are to get sued

• 
  Raccoon 🦝 (@raccoon_builds) reported 38 minutes ago

  @goldenelephant1 @Cloudflare @grok I'm not on the paid one either mate. i was just trying to help you i'll got the same thing as you and my agent Ia told me its wasn't attack only bot. i'm running a taxi company website so yeah. need to be protected.

• 
  二次裏@あいもげ (@nijiurachan) reported 39 minutes ago

  @Cloudflare Shut down the server

• 
  Aaron Ware (@EarnWhere) reported 40 minutes ago

  @Cloudflare 's PDF endpoint is so good. I spent hours trying to speed up PDF creation inside of my architecture and did so much hacky **** to achieve a decent-enough UX. Just implemented Cloudflare's endpoint and happily ripped all that out for substantially better results.

• 
  BornToLose💀🍀🏆 (@__BornToLose__) reported 45 minutes ago

  Spectrum can bite my ***, wdym 30mbps down is fine? Was really hoping it was just a cloudflare thing or something on my end but nope, I'm genuinely just stuck with the world's shittiest ISP and nothing I can do about it.

• 
  Kent C. Dodds 🏹 (@kentcdodds) reported 47 minutes ago

  @ericzakariasson Kody uses MCP to issue auth tokens for access to repos (Cloudflare artifacts) which cursor then uses *** to clone, commit, and push, and then cursor triggers a publish step through MCP. It's not CLI vs MCP. It's CLI + MCP

• 
  Joe Can Write (@joecanwrite) reported 51 minutes ago

  Bots are now 57.3% of all requests to web pages, per Cloudflare. Human traffic is the minority. If you judge content on raw pageviews, a growing share of your audience is machines that never buy anything. The question that matters now: do AI systems recommend you?

• 
  Josh W (@ItsWelford) reported 53 minutes ago

  @Cloudflare cc: @dillon_mulroy do you know anyone that can help me with this?

• 
  Caminomaster (@caminomaster) reported 53 minutes ago

  @Cloudflare Turnstile verification is not working. Unable to login #CloudflareDown

• 
  ︎Ronnie 👨‍💻 (@rxnnie_tech) reported 55 minutes ago

  @echo_vick reverse is the case here i can only access the cloudflare dashboard with mtn network

• 
  Didier Lopes (@didier_lopes) reported 56 minutes ago

  For all the talk there is on Saaspocalypse - this is a refreshing take by @danshipper . "Sasspocalypse is dumb" Earlier in the podcast he also talks about his experience using SaaS where more and more of it is being driven by utilizing Codex. I can related with this a lot, as now a lot of times I have Codex open in a specific browser page and I'm working with that app from within codex. IF they support MCP, then I connect MCP as the agent is more efficient by doing so. If they don't, I just let it check the UI view (the same way I do) take screenshot and then click/drag/hover around. It's been interesting to do work this way. It's like I'm looking above the shoulder a pro at something I don't know about - e.g. setting up a Cloudflare environment. Not just it's doing the task but I'm learning from it because I can see what is happening.

• 
  From Growth To Value (@FromValue) reported 56 minutes ago

  $NET laid off 20% of its employees, while it grew 34%. Why? That's one of the questions answered in the new Potential Multibaggers article! 💡 What else? Why dropping gross margins are actually good (really, no BS). Why the hyperscalers' data center build-out will reach its limits and how Cloudflare can help. Why OpenAI, Anthropic, Figma, Lovable and so many AI-first companies build on $NET. Why $NET will profit from agentic AI and the new internet. How $NET is trying to solve the $GOOGL Zero problem. And so much more (quality update, valuation...) See bio for article.

• 
  Colleen (@PiedViper) reported 57 minutes ago

  @LiberN8 @yacineMTB Who is "they"? Unless you're talking about Cloudflare, who has significant legal and economic constraints on how aggressively they can take down sites, blacking out a random social media site worldwide requires cooperation and can be a game of whack-a-mole, especially if some servers stop accepting certain servers' pushes. The entire key-signing rituals of ICANN are only meaningful if most people accept them as authoritative. If they start messing around with letting one country dictate policy for the world, there's not going to be one main internet any longer.

• 
  Tushar Dwivedi (@tush_2708) reported 57 minutes ago

  @kritikakodes For 50 users, get an adult "chhotu" to do the job that the system is supposed to do. A chhotu to run and call the cab. A chhotu to run and take documents to your CA, download and email, download and print, scan and email, whatever be the pain point. For 50 users, you need a validated business model, not a software system. My milkman caters to more users than this. If interviewers ask stupid questions, they should expect stupid answers. For 50 users, you can choose the worst design decisions intentionally and can still justify them. 1. Database Schema: "I will store all the data in an xlsx file, and will reread the file every time an API request comes. I will use another .lock file to control access when writing new data to this file" But what about "linearly degrading performance?" "What degradation? My API will still respond within 200 ms to all your 50 users" 2. Cloud deployment strategy: "What cloud deployment? I will run this system on a Raspberry Pi connected to a hard disk, with a cron job to do backups on 2 other hard disks connected to another Raspberry Pi. And I will set up Cloudflare tunnel to it, for your 50 users" 3. Disaster control? "I will use this 2000 rupee router power backup device"

• 
  Artem Zakharchenko (@kettanaito) reported 58 minutes ago

  @dillon_mulroy If I can help folks write better tests, it's my pleasure. I've thrown any scenarios I could think of at it and it handled them well. It'd be great for more devs to try it out while I'm thinking whether it should be @msw/cloudflare or "msw/workerd".

• 
  Abolaji Rasaq Oluwapese (@bolazeal) reported 59 minutes ago

  @echo_vick Install Cloudflare Warp to fix this issue