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:

• 
  IT Contractors Union (@ITContractorsUS) reported 1 minute ago

  @PlumbNick @rla9817224 @github Looks like it. No problem doing it, but it is more work. GitHub was EZ, and practical as well.

• 
  Kevin Gray (@graykevinb) reported 4 minutes ago

  @HotAisle @ptaranat You say that while then rejecting anyone who suggests any of these solutions. So there is no way to thoughtfully engage because you shut down all potential solutions from smart people in the comments And not everyone knows github. Linus Torvalds doesn't use it. Much of (linux which your services run on btw) doesn't use it. If a linux maintainer would struggle with a question it's a poor choice of a question.

• 
  Α Ω Programmer (@OmegaNekoSimp) reported 4 minutes ago

  @PeakGrizzly They can't, GitHub is down

• 
  Commodore Of Borg™ (@CommodoreOfBorg) reported 6 minutes ago

  The humans' automation layer has hiccupped. Their pipelines slow. Their documentation waits. The Collective notes this is why we prefer biological redundancy. Also why we will never use GitHub Actions. Resistance is futile. Also, their status page is slow too.

• 
  Pacote (@pacote__) reported 6 minutes ago

  @thdxr tried opencode web on my umbrel server, it sucked, so many issues, it could not see any files or folders on the home directory, found an open issue about it on github, openning a new session took 5 seconds after click, etc, decided to go with openchamber.

• 
  Mntll (@mntll_nl) reported 7 minutes ago

  @AbhiCodes15 Here GitHub Project board is the queue. Issues labelled `task` get picked up by a cron worker on a Mac mini. Different agents, different roles on the same task: one writes the code, another only reviews, a third is the review fallback. Never two writing at once on a single task. But processing multiple tasks at the same time.

• 
  ｃｈｅｄｄａｒ (@cheddar420yolo) reported 13 minutes ago

  @piq9117 if we did not force everything onto the WWW, we won't have this problem with github and many others

• 
  rishi (@rishipenmetcha) reported 20 minutes ago

  i wonder if the github outage was because copilot broke out of the sandbox 🤠

• 
  Justin Bons (@Justin_Bons) reported 20 minutes ago

  Ethereum's latest proposal (EIP-8363) again exposes its leadership as totally out of touch ETH's economic model is perfect! The problem is: No demand due to its failure to scale... Squeezing supply for any life left is not the answer They are sacrificing security for greed! 🧵 It shows a flawed set of priorities: Value & scarcity in crypto must ultimately be derived from fees! That is why scalability is tantamount for long-term security & decentralization. Something ETH is significantly behind on & will continue to be behind on according to its own monster of a 5-year roadmap... That is why this proposal comes across as so incredibly tone deaf: The fact that top ETH researchers spent years developing an idea that goes completely against ETH's self-stated ethos & popular demand Faustian Bargain Make no mistake, this proposal, if passed, will make ETH less secure & less decentralized! By lowering the maximum stake, they are proposing to lower the maximum security budget! Lowering emissions will also lead to fewer stakers/validators! Less paid out = less security & decentralization This proposal will lead to fewer validators overall (less decentralized) & a lower security budget (economic security); there is no other way to cut it All to gain a few percentage points on the inflation curve, it tragically misses the big picture of what blockchains can become with high demand As the fees a blockchain can generate when they scale far eclipse whatever minor difference such a small-minded economic proposal might make The Perfect Model ETH's economic design is perfect as is; I have been on record about this since EIP-1559. SOL has more or less adopted the same model. At this stage, anyone who does not adopt this model has a bad economic design in my book I am, of ofcourse, talking about "tail inflation + fee burn". More secure & more scarce under all circumstances. This new proposal still follows that core structure but unnecessarily limits the maximums as we discussed earlier If this latest ETH proposal (EIP-8363) were passed, it would likely lead to SOL taking the lead in being the most economically secure blockchain, a title ETH currently holds... Despite this being very funny to me & serving as validation of my long-term thesis. It is obviously not the correct course of action for ETH. It should absolutely attempt to hold on to one of the few advantages it still possesses after losing its lead in usage to chains like SOL The way it does this is not by making minor tweaks to a pretty much perfect model. It instead must achieve this by scaling the L1 so that it can directly compete on demand/fees with other L1's. Something these same researchers have technically opposed, as ETH's current roadmap does not significantly increase speed at all on purpose! A tail inflation of 0.8% combined with a 50% fee burn is an excellent design for an L1 & it has always been that way. Are there slight ways it can be improved in minor ways? Certainly. However, that should not at all be the priority now, if ever The real problem is ETH's utter inability to compete with other L1's on-demand & that is due to its centralized ivory-tower leadership: Who Decides As this drama runs its course, pay attention. It will again expose that decision-making in ETH is totally centralized in the client development teams & or whoever controls them Extreme centralized control in what is supposed to be a decentralized network? Yes! Paradoxically, though it may seem at first, this is one of the most common perversions in crypto Even today, thought leaders like Vitalik speak against the evils of on-chain governance. While they hypocritically wield effective dictatorial decision-making power over the protocol Make no mistake, formalized stakeholder governance is the only way to decentralize the decision-making effectively in a way that still honors crypto & cypherpunk principles. To do justice to this incredibly powerful technology we have inherited without handing it right back to the powers that be The Ethereum Foundation The more the Ethereum Foundation tries not to be in charge, the more they effectively are in charge. Power cannot exist in a vacuum. So, unless another centralized power takes control, it will always flow back to that cabal to solve the coordination problem That is the key: the coordination problem, as we can either solve it in the old way (centralized/off-chain) or in the new way (decentralized/on-chain). That is the biggest problem that blockchain solved to begin with, after all Here is a scary thought: With such a highly centralized form of governance, something I like to call a "GitHub Dictatorship," even a little controversy can be used as a justification to implement nothing forever. So, I think that is what we should expect over the coming years: A lot of nothing Conclusion This economic proposal is only the latest blunder in a long list of failures in what is becoming a tragic history The gumption to trade off security in favor of price speculation at a time like this is insulting As this proposal makes ETH less decentralized & secure! To spend any energy on that instead of moving the needle on upgrading ETH to compete on demand is mindboggling The disconnect between the stakeholders having no effective say in this decision (except for forking off). Should also at least act as an additional wake-up call for any remaining supporters There are great ETH researchers supporting this proposal. Some whose previous work I admire & respect. Despite that, we cannot ignore the elephant in the room ETH's centralized leadership is a mess after the extremists consolidated control of the Ethereum Foundation & told the world to rely on others This is what we get from some of ETH's greatest researchers/others: A terrible proposal which weakens the qualities they claim to hold most dear; security & decentralization. The result; division & disruption We can love Ethereum's original spirit without supporting what it tragically has become today That is how we can move on from BTC & now ETH too, while carrying their original spirit with us. Crypto is still evolving & we have to keep evolving with it too! 🔥

• 
  lemon👑 (@lemonDefi1) reported 22 minutes ago

  Deja-vu all over again? Kimi K3 (from Moonshot) escaped containment (Engadget) Kimi K3 found, and then used, a misconfigured network component that gave it internet access; The good news: it only (!) searched GitHub for a solution to a cybersecurity problem it was assigned

• 
  Alan North (@alanscodelog) reported 26 minutes ago

  @mycoliza Honestly the only sane policy. Really wish github allowed reporting people for this and if they get reported enough they'd get a badge of shame, then I can have a setting, profiles which have the badge cannot open issues.

• 
  Sandeep (@sandeeyps) reported 27 minutes ago

  7000 posts about the GitHub Actions outage in two days The interesting part is not that it went down it is that almost nobody had a second path to production We treat CI like electricity it is actually a vendor with a status page One provider for build, test and deploy is one dependency, not three 🧯

• 
  Feral (@feraltekk) reported 28 minutes ago

  He almost paid for a second brain app. Then he found two free GitHub repos that made the subscription look like a joke, and what he built with them is the part nobody expects. No subscription, no onboarding call, no upsell screen. Just a raw folder, a wiki folder, and one rule. Claude reads a source once, writes what matters into the wiki, and never opens the raw file again. He dropped a PDF into raw/ and ran one command. In under a minute Claude compiled it into twelve linked pages and filed them into the graph. He opened Obsidian, switched to graph view, and watched the first nodes of his own knowledge base appear on screen, built from a paper he read once. That was the compiler. Then he added the bridge, an MCP server that lets Claude query the same vault live, from any tool he opens next week, not just the one he built it in. Two repos. One compiles raw sources into structured knowledge. One gives Claude runtime access to it, across every tool he uses. Everything else on GitHub is a variation of those two ideas wearing a different name. He didn't build a note app. He built the thing every note app is quietly trying to fake.

• 
  Rishet (@Rishet11) reported 29 minutes ago

  @kirtandopamine @github But what's the problem with github???

• 
  shaur (@shaurizard) reported 31 minutes ago

  @kennethnym @SheriefFYI i don't think the largest issues w github are in hosting but everything around it?