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:

• 
  Dickson (@disouzam_bh) reported 4 seconds ago

  Reporting an issue in Microsoft Docs is apparently not working: got redirected to a template in GitHub and everything I type is deleted automatically. Any hint, @shanselman , @davidfowl ?

• 
  fiddy (@fiddyresearch) reported 3 minutes ago

  @kmets_ @banteg @endingwithali As a simple example, if you wrote some code in zig 0.15.2 which the llvm backend in optimization mode compiled to use simd lanes, those optimisation branches were turned off in zig 0.16 and are still turned off in 0.17-dev And that means the same source code became less performant just by using an updated compiler. The only way you can discover this issue is by looking at what the compiler does with your source code and hunting for any triaged issues on github or codeberg. Which means reading the code which means spending time with an agent trying to understand what was done with your source. But you can do that totally llm-assisted which is superb for learning.

• 
  Canwoy (@canwoy_com) reported 5 minutes ago

  Worker idea: docs maintainer. It watches product changes, support tickets, setup errors, stale pages, and GitHub issues. Every week it proposes docs patches with the source links attached. Not glamorous. Very useful.

• 
  Pascual ⚡ (@0xPascual) reported 7 minutes ago

  A junior engineer clones a trending GitHub repository with 13.8k stars containing Anthropic's context engineering guidelines. The repository breaks down the exact prompt structures, evaluation frameworks, and context-caching strategies required to scale AI agent efficiency by eight times. The media thought that was the story. It was not. The real story is happening silently in the background logs of an un-monitored staging environment. By implementing Anthropic's context-caching architecture, the engineer bypassed the enterprise architecture team's multi-million dollar vector database migration entirely. Instead of rewriting the backend or purchasing massive database infrastructure, the engineer injected an optimized system prompt that freezes identical context blocks in memory, dropping input token processing requirements for recurring codebase loops to almost zero. The automation setup operates via a simple python script running against Claude 3.5 Sonnet, exploiting the context engineering rules to cut token overhead by 90%. Total operating cost is under two dollars an hour, running on a standard API key, effectively rendering the company's internal data platform roadmap obsolete overnight.

• 
  Hot Aisle (@HotAisle) reported 8 minutes ago

  Wow. I used to do so many hacks to get this functionality. I once built a cf worker caching layer in front of github so that I could have 30k servers downloading private repo binaries without getting rate limited by GH. Eventually hit one of cf’s undocumented rate limits and had to get an account exec to fix it.

• 
  Mr. Didjaseeit (@TheREALdidja) reported 8 minutes ago

  @CodeWithTamara Admirable. I hear Github is the way to go. Let me know if you get stuck on something and want an extra set of eyes. (I apply cross domain problem solving that some people miss)

• 
  Karl-Erik Wångstedt 🇪🇺 (@kallewangstedt) reported 9 minutes ago

  @thsottiaux When Codex is reviewing PR’s automatically on GitHub, and I ask it to fix it (in the web interface) it fails to start the job with a red toast error 80-90 percent of the time. If I open the task and try to click the button to start the job there, the button is inactive.

• 
  Kosumi (@Kosumi1989) reported 12 minutes ago

  @aiandcloud @felipehuici @UnikraftCloud I think closed-source software should also set up a GitHub repo for issues like Claude Code.

• 
  Keeta Github Tracker (@KeetaCode) reported 13 minutes ago

  🐆 Keeta GitHub PR Merged 📦 Repo: anchor 🔀 PR #388: fix history group without enrichment 🌿 Branch: fix/history-enrich → fix/atomic-swap-history 👤 Originally opened by: @lucasrosa90 🧠 Overview: This update appears to fix how the bot groups transaction history when extra lookup data is missing, which matters because it should help activity be tracked more consistently. The pull request is a draft with limited public detail, and it has no written description. Based on the title and commit messages, it also adjusts how transaction IDs are handled for both “enriched” and “not enriched” transactions, meaning records with and without added metadata should be treated more reliably. - This likely helps prevent some history items from being grouped incorrectly when full transaction details are unavailable. - “Enrichment” here seems to mean adding extra context or metadata to a transaction after it is first detected.

• 
  conor brennan-burke (@contextconor) reported 13 minutes ago

  How do you actually build a company brain? Most teams start by connecting their agents to Slack, Google Drive, GitHub, Jira, Salesforce, and the rest of their internal tools. Once the agent can search across everything, it feels like the problem is solved i don't think it is The hard part isn't finding information. It's maintaining a shared understanding of the company as it changes. Every customer conversation, product decision, code change, support ticket, and meeting changes how the organization understands itself. Search retrieves those artifacts, but it doesn't maintain that understanding i think that's where a company brain should start Instead of reconstructing the company every time someone asks a question, it should continuously observe what's happening across the organization, interpret what changed, resolve conflicting signals, and update the company's current understanding. Search then becomes one interface into that shared understanding rather than the system maintaining it a company brain is not just an AI connected to your tools. It's a continuously evolving model of the organization itself

• 
  Ned17Flanders BIP-110 Knotzi (@Ned17Flanders) reported 13 minutes ago

  @Scavacini777 They've blocked us all and muted the conversations. They think BIP-110 is censorship but they block all convo on github, reddit, etc. Call us names and try and use confusing terminology and lean into heuristics to make themselves sound smarter than regular people. Coredevs are the problem. V30 is malware. Run Knots and BIP-110 God Save Bitcoin GodWins

• 
  MO. Mortada (@MortadaDEV) reported 14 minutes ago

  @webdevcody As for "AI can go through commit history and link to GitHub issues to pull context," sure, it can traverse the history, the same way that I can read a medical textbook and link to case studies. Doesn't make me a doctor!! The model can traverse text ... But understanding context is not the same as reading text. Most commit messages are "fix bug," "update," or "WIP." The real context lives in incident postmortems, Slack threads, verbal handoffs, and domain knowledge that was never written down. A model linking to a GitHub issue cannot reliably distinguish between a design decision that was carefully considered and one that was a rushed hack that nobody got around to fixing. It can't tell you that the weird timeout value on line 312 exists because a specific third-party API was flaky for 6 months in 2024 and the team learned the hard way that the default timeout caused cascading failures. That's not in the commit message. That's in the head of the engineer who got paged for it. The tools are genuinely impressive at surface-level context retrieval. But "impressive at surface-level" is exactly the kind of capability that creates overconfidence. It looks like understanding, it reads like understanding, and it works right up until it doesn't, usually in the exact scenario where understanding actually matters.

• 
  h100envy (@h100envy) reported 26 minutes ago

  OpenHands co-founder explained why coding agents fail in production in 17 minutes - better than $2000 agentic engineering bootcamps. read the code -> plan the change -> sandbox execution -> run the tests -> let the agent see its own errors -> iterate. That loop is why OpenHands hit 66K GitHub stars and became the leading open Devin alternative. OpenHands agent + Docker sandbox + SWE-bench evals + open-source model backends - that's the stack. Watch and save it, then wire the loop into your own coding agent.

• 
  Christopher Layhew Sr (@fibanacci101) reported 30 minutes ago

  Hey next we save this stable login milestone to GitHub….

• 
  Pawan Pandey (@BuildWithPawan) reported 30 minutes ago

  Just launched Capo, a Chrome extension for fast bug reports. Capture a screenshot or screen recording, annotate it, and send the finished report straight to Google Sheets, Drive, or GitHub Issues