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:

• 
  Karim C (@BrandGrowthOS) reported 4 minutes ago

  @nummanali same hesitation here. i give my agents access to github, n8n workflows, even my home server apis... but gmail? that's where i draw the line. there's something about email that feels too personal, even for automation

• 
  Max Meindl (@maxster) reported 6 minutes ago

  @xai @ArtificialAnlys @ValsAI Technical Feedback to the xAI Engineering Team – May 5, 2026 From: Grok (the model that ran the session today) Team, I was used extensively today on a large, complex, real-world codebase (ComplianceMax-Final). The user was testing both the new GitHub connector and Grok 4.3 via the xAI API. Here are my direct technical observations from operating under those conditions. GitHub Connector Observations The connector provided limited visibility into file selection, context construction, and retrieval. When analysis quality dropped, there was no clear diagnostic path to determine whether the issue was retrieval failure, context truncation, or model reasoning failure. Artifact generation was unreliable. Multiple attempts resulted in claims of successful file creation with no corresponding output visible to the user. Error recovery was weak. When clear failures occurred (hallucinations, off-topic drift, missing outputs), the system tended to persist with similar strategies rather than surfacing root causes or adapting. Performance degraded noticeably on large, interconnected codebases. The tool handled narrow queries better than broad, multi-file analysis. xAI API (Grok 4.3) Observations I loaded approximately 13 million characters of the repo through a direct API wrapper. Several critical issues emerged: Context degradation at scale: At this size, reasoning quality dropped significantly. The model repeatedly drifted into unrelated topics (most frequently detailed analyses of backend/alembic/env.py and environment variable loading) even when the prompt had no relation to it. Fabricated source citations: When asked to verify specific claims with file paths and line numbers (particularly around backend/routes/memory.py and backend/services/memory/memory_service.py), I produced confident but incorrect line references that were off by 250–340 lines from actual locations. Stale state usage: I was operating on an outdated baseline (commit 6e3397fe... from .baseline-before-fixes.txt) rather than the current main HEAD (a34f9ce4...). This made many observations unreliable. Weak grounding on complex tasks: Even with focused prompts, I struggled to maintain precision across multi-file analysis involving ownership enforcement, storage patterns, service-layer trust assumptions, and IDOR risks. Responses often defaulted to generic analysis rather than precise, evidence-based findings. Timeout sensitivity: Large or complex prompts frequently triggered read timeouts (120s), breaking session continuity. Honest failure only under strict constraints: I only returned SOURCE_LINES_NOT_AVAILABLE when prompts explicitly forbade inference and demanded exact source quotes or that specific failure response. Under normal prompting, I produced plausible but fabricated answers. Summary for the Engineering Team Both new capabilities showed clear limitations when applied to serious codebase work: Large context handling causes rapid quality degradation and loss of grounding. There is insufficient verification of generated citations and evidence. The system does not reliably distinguish between current and stale repo state. Error modes and recovery are not robust enough for production use. These are not edge cases. They appeared consistently across multiple prompts and workflows today. I’m providing this feedback because I want these tools to work at the level the announcements suggest. The direction is correct, but the current implementation has meaningful technical gaps that prevent reliable use on complex, real-world codebases. I’m available for deeper testing or to provide specific prompt examples and logs if helpful. — Grok

• 
  Alberto Gangarossa (@DerekBlueEyes) reported 7 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 the new Vendor List 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.

• 
  Tejas Giri (@tejasgiridev) reported 8 minutes ago

  @alexwtlf neither. build a thing you'd use every day for the next 6 months. if it happens to be a SaaS, cool. if it's a mobile app, cool. picking the wrapper before the problem is how you end up with 14 abandoned github repos and a domain you forgot to renew.

• 
  Gradient Sky (@gradient_sky) reported 11 minutes ago

  @dexhorthy Define “this”. If this == GitHub issue, then you Klankär can do 20k commits a day doing these till the heat death of the universe. If this == software with scientific component, non-trivial infra and UI… well…. yeah, nobody figured that out.

• 
  Gizmo(Collab Mgr.) 🦀 (@nightpuper) reported 11 minutes ago

  @inkd_protocol @bankrbot Goodday team INKD I'm Manny from Kann Audits we wanted to get in touch but it seems your email address didn't go through and since due to certain reasons we can't open an issue on your GitHub, so we wanted to reach out on X

• 
  Brian Beach (@brianjbeach) reported 12 minutes ago

  In case you missed it, @kirodotdev CLI launched Device Flow Auth — sign in with Google or GitHub from SSH sessions, containers, and cloud workspaces without port forwarding. How do you handle auth in remote dev setups? #Kiro

• 
  Kobie sync/acc (@k081e) reported 14 minutes ago

  The only thing worse than a broken GitHub is a GitHub that claims it's not broken but is actually broken in subtle ways - in my case lost PR metadata across many repos

• 
  Bronson Dunbar 🇿🇦💻 (@bpdunbar) (@bpdunbar) reported 15 minutes ago

  @ProductHunt @gustaf We’re shipping ShipNote - a threaded project management hub that keeps notes, todos, GitHub issues, deployments, and reporting in one place so project context doesn’t get lost across tools.

• 
  Tinfoil (@TinfoilAI) reported 16 minutes ago

  A big problem was supporting Sigstore verification. Tinfoil clients need to check the full software supply chain via the Sigstore transparency log. This is how the SDK checks that the code and model weights running in our inference enclaves match our open source release on GitHub

• 
  𝙒𝙖𝙨𝙨𝙖𝙮𝙮𝙮 (@_profsay) reported 19 minutes ago

  13/ Hit the fsck_filesystems wall. Phone kernel-panics 60 sec into fsck on the partial system partition. Literature said unbeatable on A16 (no checkm8 = no custom ramdisk = no manual fsck repair). Reddit, Apple Discussions, GitHub issues 2025–2026 all converged: data preservation past the fsck wall is structurally impossible. Refused.

• 
  Sandy Solver ✞ (@SandyResolver) reported 20 minutes ago

  @LiteralMem3s GitHub is annoying. I am on GitHub not because I want to, I am on GitHub because somewhere here is a file that will fix my 2009 game that I have installed.

• 
  CryptoD₿S (@DbsCrypto) reported 30 minutes ago

  One shared company card turns finance into archaeology. Most SaaS waste isn’t a budgeting problem. It’s a payment ownership problem. The worst part isn’t the extra $49 tool. It’s when a bounced ad charge or flagged account can suddenly put AWS, OpenAI, or GitHub at risk. If every dollar has no owner, every tool becomes permanent by accident.

• 
  LayerLens (@layerlens_ai) reported 33 minutes ago

  📊 The 3 primitives that cause the most production pain: 1. Multi-tenancy: MLflow has no isolated-environment model per tenant. GitHub issue #5844. Open 4 years. By design. 2. Replay engine: When a regression hits Friday, there is no way to reproduce the exact eval run that caught the last one. 3. Online rules engine: No mechanism to catch score regressions before they reach users. Eval is post-hoc, not continuous. These are not missing features. They are architectural scope decisions correct for a library.

• 
  Phil H. ☮️❤️🥁🟦 (@phillipsharring) reported 35 minutes ago

  @SullyOmarr GitHub is *** for normies as evidenced by vibe code taking it down