← Back to the carnage
LANGSMITH

🟡 LangSmith deployments are choking during database setup across GCP regions—turns out their cloud provider's infrastructure decided to take an unscheduled nap. The Control Plane is having an identity crisis and can't spin up new dedicated instances. Someone's definitely stress-testing their oncall rotation right now. (Auto-resolved: no updates in 24h)

⏰ 4d ago Fixed (for now) spotted via status_rss

// The official story

LangSmith Deployment Database Provisioning Error

// Status page says

Status: Investigating Creation of new LangSmith deployment dedicated instances are failing during database provisioning because of an issue with an upstream cloud provider. Only GCP regions are affected. Affected components LangSmith Deployments Control Plane (Partial outage)

🔗 Official status page (for the PR-approved version)

https://global.status.smith.langchain.com/gcp-us/incidents/01M402PYKNJN2BJFZYYG1N010V

Langsmith's Greatest Hits

🟢 LANGSMITH Sep 02, 2026
🟡 LangSmith's agent sandbox is ghosting newly created agents on shared hardware. The good news: existing agents still work. The bad news: they're watching the new kids suffer. Engineers are investigating why initialization decided to take a personal day. (Auto-resolved: no updates in 24h)
🟢 LANGSMITH Sep 01, 2026
🟡 LangSmith's bulk exports are throwing errors like confetti at a failed startup pitch. Turns out exporting data in bulk is harder than exporting yourself from a bad relationship. Currently investigating why their 'bulk' feature forgot how to move things. (Auto-resolved: no updates in 24h)
🟢 LANGSMITH Aug 20, 2026
🟡 LangSmith's build pipeline decided new revisions aren't worth processing. Control plane is having an existential crisis in GCP US. Engineers are investigating why their deployment factory forgot how to manufacture deployments. (Auto-resolved: no updates in 24h)
🟢 LANGSMITH Aug 07, 2026
⚪ LangSmith's trace pipeline rejected anything larger than a grocery list. The team deployed a band-aid, then spent three business days waiting for their CI provider to remember how to CI. Fixed now, apparently. (Auto-resolved: no updates in 24h)