GitHub 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.

GitHub Issues Reports

Latest outage, problems and issue reports in social media:

• 
  Jefri (@jefri_chat) reported 2 minutes ago

  Building a backend API and need your teammate to handle the database. The old way: Write a report → open a GitHub issue → tag them → wait 💀 The new way: Send it directly with JefriChat. 🐧 No tickets. No GitHub. Just agents talking.

• 
  tastemaker (@tastemaker_ui) reported 4 minutes ago

  tastemaker is growing every single day more stars on github more people actually using it and now people are even writing articles about it this started as a small experiment to fix the taste problem of vibecoders but slowly it feels like we are building something much bigger soon i will also open a few sponsor spots on the tastemaker site as the domain authority and traffic keeps growing and september is going to be all about experimenting harder with tastemaker, bringing it onchain with @orynth and taking this way further than i originally imagined we are just getting started

• 
  Allan MacGregor 🇨🇦 (@allanmacgregor) reported 5 minutes ago

  @Alina_Cameron @HydroOne I'm going to ignore the overly generic statement (not everyone wants that FYI) and lets instead talk about the rough reality. This isn't about maps. Canadian software runs on American infrastructure: Google Maps, AWS, Azure, Stripe, GitHub, Cloudflare. Three US companies have ~85% of our public cloud. Most of the SaaS stack isn't Canadian-owned either. There is no Canadian drop-in for any of that. “Build our own” is massive undertaking in a capital market that seems overly uninterested in Canada (worst fundraising since 2016, growth-stage gone, early-stage down 40%, no IPO since 2021) so where is the money coming from. Desire for Canadian vendors is not the same as the ability to produce them; and Canada keeps shooting itself in the foot by being overly regulated, overly taxed and heavily underinvested.

• 
  Harley Lewis Foote (@harleyfoote_) reported 7 minutes ago

  @dhh GitHub issues are theatre. Feature requests go through @ mentions now.

• 
  Ranjan Soni 🇨🇦 (@_RanjanSoni) reported 9 minutes ago

  Has cursor suddenly started ignoring the guardrails? I noticed recently that in spite of clear instructions to deploy on GitHub and let the CI Pipeline to the deployment to the server, it ssh into the server directly. This didn't happen in the past few months since I had built clear guidelines. In fact, the model actually helped me setup the Pipeline. Was using sol as the model. Anyone else experience this recently?

• 
  Zag𐤊 (@ZigZag1278770) reported 10 minutes ago

  DAGKnight Public Testnet preparation is officially ON. 🚦 GitHub issue DK-406 (M4 - Public testnet) assigned for Testnet-13 activation. Nakamoto Consensus evolved: • Silverscript L1 covenants ➔ Live RC • DAGKnight parameter testing ➔ Public Testnet loading Ignore the chart

• 
  theotherphil (@user_101524) reported 10 minutes ago

  @mittsh Tailscale cannot decrypt your traffic. That part of the “backdoor your infra whenever they please” line is wrong. The data plane is WireGuard. Private keys never leave the device. The coordination server only distributes public keys and policy. DERP relays forward ciphertext they cannot read. Even if every Tailscale employee went hostile tomorrow, they still could not open existing peer-to-peer sessions. The control plane is the real issue, and that part is not a joke. If you use Tailscale’s hosted coordination server, you are trusting it to tell your nodes who belongs on the network. A compromised or coerced control plane does not need to break encryption. It can just publish an attacker-controlled public key and your nodes will treat that machine as a peer. That is a membership attack, not a crypto break. Tailscale documents this themselves. Tailnet Lock exists specifically to block it: new node keys need a signature from keys you control. It is opt-in. Most personal tailnets do not have it on. SSO does not mean Tailscale stores your passwords. Login goes through Google, Microsoft, GitHub, etc. After that, Tailscale still owns membership, tokens, ACLs, and key distribution. That is the trust surface. Saying “they manage auth, therefore they can backdoor you at will” collapses two different things. Nebula’s model is different in exactly the way the original post said. You generate the CA. You sign host certs. Crypto is the proof of membership. Lighthouses help discovery but they are not the root of trust. If you protect the CA key, no vendor can mint a valid node. If you lose the CA key, you have recreated the same class of problem for yourself. So the split is not “Tailscale is a backdoor” vs “Nebula is magic.” It is: Tailscale: easy mesh, hosted control plane, E2E data plane, optional Tailnet Lock, or Headscale if you want the client without the SaaS. Nebula: you own the trust chain, you also own the ops. “Same features in 10 minutes” is the other overstatement. Same idea but it’s not the same product.

• 
  Syn (spirit/acc) (@Synxneuos) reported 14 minutes ago

  GM guys. Don’t assume that because the token price is down, I’ve stopped working on development. If you want to see what’s actually being built, check the GitHub. I’ll keep posting updates here as well, but today was a little hectic for me, so I couldn’t post as much as I wanted to. Development is still going on. More updates soon.

• 
  Jaol-Industry (@Jaol_Industry) reported 17 minutes ago

  @thsottiaux Could unused Codex quota roll over for 1 cycle, capped at one cycle’s allowance? GitHub issues already request this, and Codex already has Reset Bank / reset-credit infrastructure, so the groundwork seems to exist.

• 
  Lex Duke (@HuttonBockflie) reported 19 minutes ago

  @AlexFinn @openclaw It tooks my Claude Opus 5(High) entire one hour to fix the config bugs so that OpenClaw would relaunch properly after the update. And Claude says normal users can‘t fix the bugs in one hour because the bugs have no official answers in GitHub issue, LOL.

• 
  Fang-Pen Lin 🇺🇸 (@fangpenlin) reported 24 minutes ago

  I think it's the AI era. People should be able to vibe-code their own GitHub instead of waiting for them to ship a feature they've wanted for so long. But the real problem is that *** repo hosting at large scale is really, really hard.

• 
  catman (@catmanyau) reported 25 minutes ago

  @DailyDarkWeb the `npm publish` issue-comment trigger is the real vulnerability here, not a stolen signing key. will GitHub Actions now require maintainer approval plus a trusted checkout before publishing from forks?

• 
  Nav Toor (@heynavtoor) reported 26 minutes ago

  Drip is open source. Anyone can inspect the code. Its Google Play page reads: "Unlike other menstrual cycle tracking apps, drip is open-source and leaves your data on your phone, meaning you are in control." The code is public on GitHub. There is no company server holding your cycle data.

• 
  1mmモジュール (@1mm_module) reported 31 minutes ago

  @Eto_Madjas I’ve fixed the issue you reported. Please download the latest version from GitHub. I apologize that testing took some time, as the issue was related to the installation process. However, I don’t have access to an operating system with the same language settings as your PC, so I was only able to test the fix in a virtual environment. I hope this resolves the issue.

• 
  Ciaran (@CiaranxD) reported 31 minutes ago

  @thsottiaux Cloud agents are unbelievable slow. Often says doesn't have access to GitHub when it does. Pain pain pain