Project: [PROD] CopierGuy SAP

Environment: Dev

Service: CopierRental

Public domain: https://copierrental-dev.up.railway.app

Current deployment: 16a78456-accc-4156-a5bd-35298f918c72

Region shown in headers: railway/asia-southeast1-eqsg3a

Runtime: V2

Issue:

The CopierRental app recently became extremely slow from the public Railway domain. The delay appears before first byte, even for static files. Other apps in the same Railway hosting setup do not appear to have the same issue.

Evidence:

From local/public access, both static and dynamic routes consistently show about 12 seconds TTFB:

- HEAD /img/big-logo.png: ~12.08s TTFB

- HEAD /Identity/Account/Login: ~12.25s TTFB

Example slow request headers:

- x-railway-cdn-edge: fastly/cache-kul9823-KUL

- x-railway-edge: railway/asia-southeast1-eqsg3a

- x-railway-request-id: YvfkAQ-ZRYW1zhrSV7rehQ

Another slow request:

- Route: /Identity/Account/Login

- x-railway-cdn-edge: fastly/cache-kul9827-KUL

- x-railway-request-id: E1PQIeX2RGiCTzjyAQeqjw

Comparison:

When testing from inside the Railway container / Singapore path, the same public URL returns quickly through the SIN edge:

- Route: /img/big-logo.png

- x-railway-cdn-edge: fastly/cache-sin-wsap440073-SIN

- x-railway-request-id: Wu03D-1IRMetgiGcAXC71g

Origin check:

Direct Kestrel requests inside the container to 127.0.0.1:8080 return immediately, so the app process/origin does not appear to be the bottleneck.

App-side checks:

- Fixed IP restriction is disabled: Security__FixedIpAccess__Enabled=false

- Static files bypass the app’s data filtering and fixed-IP middleware

- Database queries tested fast, around milliseconds

- Service status is SUCCESS for both CopierRental and MySQL

Request:

Can Railway investigate why requests routed through the Fastly KUL edge for this Railway service/domain have ~12s TTFB while origin and SIN edge requests are fast? Please advise whether this is an edge routing/cache/CDN issue, Runtime V2 issue, or domain routing issue.