an hour ago
Since 14:17 UTC, all outbound HTTP requests from our services in us-east4 (projects “Backend - LIA CRM” and “LIA CORE - API”) to ANY Railway-hosted domain get 429 rate limited from the edge (server: railway-hikari, x-railway-edge: iad1). This includes our own domains (api.registry.liacrm.io, lia-core-api.up.railway.app) and also railway.com and docs.railway.com, 100% of requests, including /health. Requests from the same containers to non-Railway hosts work. The same URLs return 200 from outside Railway. Our request volume is tens per minute. Redeploying did not help. Production is down. Please check edge rate limiting for our egress.
7 Replies
Status changed to Awaiting Railway Response Railway • about 1 hour ago
an hour ago
The only option is to give up on Railway; it’s impacting my service deliveries to clients.
an hour ago
Additional data, measured at 14:42 UTC from inside the container of service lia-crm-api (project Backend - LIA CRM, us-east4):
GET https://api.registry.liacrm.io/health → 429, body “rate limited”, x-railway-edge: iad1, no x-railway-request-id header
GET https://railway.com/ → 429, body “rate limited”, x-railway-edge: iad1
Egress IP of that container: 162.220.234.15 (after a redeploy another container had 152.55.180.55, same result)
Same result from service “lia core - api” (project LIA CORE - API)
GET https://api.github.com/zen from the same containers → 200
First failure in our logs: 2026-11-10 14:16:50 UTC. Partial failures until ~14:24, 100% since ~14:30.
an hour ago
We have same issue, our app is getting 429 and nobody can use it, please help resolve this fast!
it seems some kind of networking issue
an hour ago
seems resolved now
an hour ago
- resolved for me right now
Status changed to Solved tandera-io • about 1 hour ago
an hour ago
Thanks for confirming recovery. We're sorry for the 429s that disrupted LIA CRM and LIA CORE.
An attack prevention rule intended to target domain scanners and scrapers was too aggressive and ended up affecting legitimate traffic. We've disabled that rule, and the issue is now resolved.
Please let us know if you see any further 429s between your services.
Status changed to Awaiting User Response Railway • about 1 hour ago