What Broke Today

🔴 HASHICORP 3h ago

🟠 HCP Terraform is partially down and choking on runs. Engineering deployed a fix and are now in the 'monitoring' phase, which is code for 'fingers crossed and refreshing dashboards every 30 seconds.' Infrastructure as Code? More like Infrastructure as Chaos right now.

🔴 ELEVENLABS 10h ago

🟡 ElevenLabs is speedrunning every engineer's nightmare: half their infrastructure decided latency is a feature, not a bug. Telephony, UI, Quality, Conversations, Integrations, and RAG are all throwing timeouts like they're going out of style. The team is "actively looking into it," which in corporate-to-English translates to "we have no idea what happened yet."

🔴 ELEVENLABS 10h ago

🟡 ElevenLabs endpoints decided to play slow-motion chess with your requests—Quality, Conversations, Integrations, and RAG all running at molasses speed. Engineers are investigating why their infrastructure suddenly developed commitment issues with actually responding to things.

🔴 CURSOR 19h ago

🟡 Cursor achieved quantum superposition: simultaneously broken everywhere. IDE, agents, bots, automations—basically the entire stack decided to unionize and call in sick. Engineering team is now speed-running 'have you tried turning it off' on production.

🟢 SUPABASE 1d ago

🟡 Supabase's logging service decided to ghost its own data. Logs are either fashionably late or taking an indefinite sabbatical. The irony of your observability tool becoming unobservable is not lost on anyone. (Auto-resolved: no updates in 24h)

🟢 CONTENTFUL 1d ago

🔴 Contentful's developer docs vanished like a ghost in the machine. Monitoring caught the body, investigation underway. Apparently 'headless CMS' now includes headless documentation—no brain required. (Auto-resolved: no updates in 24h)

🟢 GROQ 1d ago

🟠 Groq traced minimax-m2.7's existential crisis to a hardware connectivity breakdown. Engineers are frantically re-introducing the model to its own infrastructure while users experience the joy of API 500s. Apparently even AI needs a moment to remember where it plugged in. (Auto-resolved: no updates in 24h)

🟢 LANGSMITH 1d ago

🟡 LangSmith's authentication layer is playing hard to get—some login attempts timeout while others fail outright. Already-logged-in users watching the chaos unfold safely from inside. Engineers are 'investigating' which tech speak for 'we're checking the logs for the first time today.'. (Auto-resolved: no updates in 24h)

⚪ PAYPAL 1d ago

⚪ PayPal's Buy Now, Pay Later got a little too literal about the 'later' part and started declining everything. Global checkout errors ensued while engineers performed financial triage. Now resolved, so your impulse purchases can resume their regularly scheduled deductions.

🟢 FLY 1d ago

⚪ 🎭 Turns out SSL certificates have commitment issues too. Fly's sprite subdomain decided to play hard-to-get with a certificate error, leaving developers staring at broken image tags for an hour. Romance was rekindled at 00:45 UTC. (Auto-resolved: no updates in 24h)

🟢 GROQ 3d ago

🔴 Groq traced minimaxai/minimax-m2.7's performance meltdown to a data center having an existential crisis. Engineers are frantically applying metaphorical duct tape while users watch their requests vanish into the void. Fix incoming, but your failed requests won't retroactively succeed—that's not how apologies work. (Auto-resolved: no updates in 24h)

🟢 CONVEX 3d ago

⚪ Some query and mutation executions decided to ghost their users for 41 minutes thanks to an upstream network hiccup. Convex's infrastructure had commitment issues, but they're back to pretending everything's fine now. (Auto-resolved: no updates in 24h)

🟢 PLANETSCALE 3d ago

🟡 One availability zone in us-east-1 decided to go dark, turning replicas into time travelers and connections into fiction. PlanetScale's status page is doing that thing where 'investigating' means engineers are stress-eating at their desks. (Auto-resolved: no updates in 24h)

🟢 DOCKER 3d ago

🔴 Docker Hub's registry decided authentication is optional—by returning 500s to anyone who tries. Your `docker pull` commands are getting the digital cold shoulder. Engineering team is investigating why their own infrastructure doesn't trust itself. (Auto-resolved: no updates in 24h)

🟢 OPENAI 3d ago

🟡 OpenAI's investigating why some workspace admins can't manage plugins—turns out giving people control over their workspace was just a beta feature. Connectors/Apps decided to ghost the people actually trying to run things here. (Auto-resolved: no updates in 24h)

🟢 OPENAI 3d ago

🟠 OpenAI's Android app is playing content gatekeeper, showing "unavailable" errors instead of letting you actually chat. The fix is trickling out slower than a 2G connection. Web version works fine if you want to escape the mobile experience entirely. (Auto-resolved: no updates in 24h)

🟢 CIRCLECI 4d ago

🟠 Your jobs finished their 5K before the UI noticed they left the house. Workflows are running fine, just don't ask the dashboard about it—it's still on coffee break from an hour ago. (Auto-resolved: no updates in 24h)

🟢 ZAPIER 4d ago

🟡 Zapier's S3 bucket got flagged by Google Safe Browsing, so now your file downloads come with a free 'this might infect your computer' warning. Nothing says 'enterprise-grade' like your own infrastructure ratting you out to security algorithms. (Auto-resolved: no updates in 24h)

🟢 COINBASE 4d ago

🟡 Interac threw a fit, so Canadian customers can't move their crypto in or out. Coinbase is 'investigating'—which is corporate speak for 'we're refreshing our status page every 30 seconds while drinking cold coffee.' Your money's fine, just temporarily unvisitable. (Auto-resolved: no updates in 24h)

🟢 MONGODB 4d ago

🟠 MongoDB's search engine decided to take a unilateral vacation, leaving some users' queries stranded at the terminal. Engineers are frantically implementing a fix while mongot processes contemplate their career choices. (Auto-resolved: no updates in 24h)

> cat /var/log/failures.log