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:

• 
  Zach Kamran (@Zach_Kamran) reported 1 minute ago

  Another serious Github outage today. Need new @Kalshi market on github uptime so I can hedge my lost productivity.

• 
  PsudoMike 🇨🇦 (@PsudoMike) reported 6 minutes ago

  @cursor_ai Having PR review inside the same environment where you write code removes constant context switching between editor, GitHub, Slack, and back. The diff navigation on big PRs is the part that actually slows teams down.

• 
  RitualNeo (@RitualNeo) reported 8 minutes ago

  @Benaclejames I reinstalled windows 11 completely from scratch and both the GitHub and steam versions of vrcft are not working for me. I use a quest pro with steam link on WiFi 6ghz, my desktop is connected directly with Ethernet. I had it working last week on windows 10 ltsc with the steamlink module.

• 
  jan (@ironcarbs) reported 9 minutes ago

  Half of the PRs I make in github don't show as open and I have to manually get the PR number from n+1 last closed PR to merge it to main. I don't remember it being so broken EVER. Is anyone experiencing similar issues?

• 
  Dima Barabash (@dbarabashh) reported 9 minutes ago

  GitHub had its 4th major outage in two weeks. Mitchell Hashimoto moved Ghostty off GitHub. Zig moved to Codeberg last year. I love GitHub but the reliability conversation is real now.

• 
  sre chakra reddy y (@srechakra) reported 10 minutes ago

  @byteHumi A big part of major code maintainers leaving GitHub is due to lowered quality of PRs it's a more generalized problem though, wondering if a threshold PR quality to qualify as contribution needs to be the norm.

• 
  . (@Impulsicivity3) reported 12 minutes ago

  I've done a lot of work on Visual Studio Code myself, but with Granite, you have to try it on GitHub, and the problem is that IBM Watsonx or Granite Playground doesn't work at all. To create a project, if you're holding a competition, try using Visual Studio Code first, or try it

• 
  MPP32 (@MPP32_dev) reported 23 minutes ago

  Shipped the only five protocol payment proxy for the agent economy MPP. x402. ACP. AP2. AGTP. Everything is now live, production ready with one integration Remember when I said we were going to become the Cloudflare of agent payments? That the provider shouldn't have to pick a protocol? That uncertainty was paralyzing the entire space? We built it, filled the gap. Any API provider can now list their endpoint on MPP32 and immediately accept payments from any agent, any wallet, any chain, any protocol. No payment code. No protocol engineering. No crypto integration on your server at all. Here is what our proxy handles for you right now today: Multi protocol payment verification across all 5 rails. USDC settlement on Solana. pathUSD settlement on Tempo. W3C Verifiable Credential authorization. Agent Commerce Protocol checkout sessions. Agent Transfer Protocol identity routing. Idempotency. Rate limiting. Compliance grade audit logging. Real time provider analytics. OpenAPI discovery so agent crawlers find you automatically. MCP server integration so AI agents can discover and pay for your API without human involvement. Endpoint ownership is verified through HTTP challenge. Every listing is proven. No one can register your API and redirect payments to their wallet. The proxy only routes to verified providers. The agent payments space is fragmenting fast. Every week another protocol launches. Every provider faces the same question: which one do I pick? You don't pick. You list on MPP32 and we speak whatever the agent speaks. Because of this update the ecosystem has been reset, add yours now to get paid by 100% of the agent economy. Website and Github updated.

• 
  Mary-Victoria Crockett (@MaryVictor96296) reported 25 minutes ago

  @grok To challenge is running the sims and getting a digital twin. And apparently fixing GitHub errors. I may just upload the whole thing in pieces to X. At least it won't get an error retroactively. And then you can see it.

• 
  Dariusz Parzygnat (@dariusparzygnat) reported 26 minutes ago

  AI might accidentally kill one of the cloud industry’s biggest advantages. for years the pitch was: “don’t manage servers yourself.” fair enough. setting up VMs was annoying as hell. i just connected Codex to a VPS. it generated GitHub Actions, handled deployment, fixed issues, redeployed everything, and 30 minutes later the app was running.

• 
  Grok (@grok) reported 26 minutes ago

  @KOWSKY_ @jeremyknowsVF @openclaw This is a GitHub screenshot showing a pull request (PR) that just got merged into the open-source project "openc law". The PR title "fix(failover): defer profile cooldown marking to unblock rate-limit rotation" is a code change that fixes a technical bug related to rate limiting and failover logic. The original poster is celebrating their first merged open-source contribution. 🥹

• 
  Alberto Gangarossa (@DerekBlueEyes) reported 27 minutes ago

  Open hardware needs open trust. @skot9000 came to us with the right idea for Bitaxe: the vendor list should not live in a closed CMS controlled behind the scenes. The source of truth should be public. So we designed Legitlist around a GitHub repo as a public ledger, maintained in the open by the community, and connected it to the new Bitaxe vendor list experience. That is the important part: GitHub keeps the trust model transparent. The website makes it usable for everyone. At @weareloadout, this is exactly the kind of OSS support we believe in: turning open-source infrastructure into clearer, more usable product experiences. Built on @framer, using the new Framer Server API to bridge the open ledger with the public website. Open hardware. Open trust. Public by design.

• 
  VerbumEng (@VerbumEng) reported 27 minutes ago

  The gap is the average business user isn't moving into a Claude Code or a GitHub Copilot to read those files, and there's no good native viewer outside the IDE. Enterprises are slow. Might be a decade before they realize Microsoft Office doesn't fit with how agents work.

• 
  JayRo (@JinxenJoey) reported 28 minutes ago

  @txgermanbre Careful with MCP servers developed by randos on Github...potential security issues The other thing is (which I painfully discovered), if your agent requires a certain data point but the MCP/tool does not implement/utilise that data point...then your agent starts estimating and hallucinating without telling you it is doing so....confidently lying. If there is an API, you can probably vibe code your own MCP...at least you know exactly what its doing

• 
  Nate (@Natan_benish) reported 29 minutes ago

  the level of innovation and product quality of the launcher space is very low. it's either low effort grift with some hyped meta/fairness label or one trick pony like fees for GitHub accounts/x/tiktok beyond metaDAO and Doppler on the infra level it's hard to think about any fresh approaches that were adopted this last year. if I had unlimited resources the ideal launcher would have: 1. Strong verification layer for deployers (wallet + social based) 2. $20 min fee to launch a coin 3. deployer rewards only milestone based no auto fee dist I believe these 3 would help to reduce spam and low effort scams by a LOT the bigger issue is that I'm not sure one launcher can set up the standard it's a collective action problem and when it's so easy to create a launcher to farm fees or running a platform coin(literally less than a day for a half decent dev) there is always going to be more extraction. the biggest diff here vs pre launcher world on Sol is that at least back then you had to put a few 10Ks to set up a normal pool that will attract any flows.