TCP handshake timeouts
hdwhite1980
PROOP

6 days ago

Service: consilium

Deployments affected: 5380beb7-ee6f-4659-8ee5-fede90709d13, 1dc229a2-1543-4799-9d5c-ba4e7cb4ead3

Issue: TCP handshake timeouts + premature stream closes to external APIs (Anthropic, OpenAI, Perplexity)

Timing: Started ~15:30 UTC, correlates with Supabase infrastructure event

Scope: All regions (tested us-east4-eqdc4a and ams), all external APIs

Root cause: Egress path degradation, not application code

$10 Bounty

2 Replies

Railway
BOT

6 days ago

Your logs from around 16:20 UTC show Cloudflare 522 errors from your Supabase instance (thlrdnkftdhakuzyhivp.supabase.co) alongside ERR_STREAM_PREMATURE_CLOSE from the Anthropic and Perplexity APIs. The 522 errors originate from Cloudflare's edge in front of Supabase, meaning Supabase's origin was not completing requests, which aligns with the Supabase infrastructure event you mentioned. No matching errors appear after ~17:00 UTC, suggesting the issue was transient and tied to that upstream event. If the errors have resumed, please reply with the current timeframe and we can check the logs again.


hdwhite1980
PROOP

6 days ago

the errors are still happening and the supabase issues are resolved


Status changed to Awaiting Railway Response Railway 6 days ago


Railway
BOT

5 days ago

This thread has been opened as a public bounty so the community can help solve it. The thread and any further activity are now visible to everyone.

Status changed to Open Railway 5 days ago


Welcome!

Sign in to your Railway account to join the conversation.

Loading...