New "via: 1.1 Varnish" header in api responses
ctrlshifti
PROOP

2 months ago

hello! i discovered a new header "via1.1 varnish" being added to all my service API responses. what is it, and why is it there? i don't use varnish. (i assume it is related to the collaboration with fastly)

Solved$10 Bounty

Pinned Solution

domehane
FREE

2 months ago

Hello,

your assumption is correct. railway moved all customer domains behind fastly on february 19-20 2026 as an emergency response to a series of ddos attacks that hit their infrastructure. fastly is now acting as a waf/proxy in front of all railway-hosted services platform-wide, which is why you're seeing that new header. you didn't do anything wrong, it's a railway infrastructure change

source: https://blog.railway.com/p/incident-report-february-19-2026

5 Replies

Varnish is an HTTP reverse proxy that works by caching frequently requested web pages as well as APIs. It's probably used somewhere, that's why you see its header in the response object.


darseen

Varnish is an HTTP reverse proxy that works by caching frequently requested web pages as well as APIs. It's probably used somewhere, that's why you see its header in the response object.

ctrlshifti
PROOP

2 months ago

I understand, but why is it widely used on railway infrastructure? Isn't it bad practice to force a caching layer on a user's API calls without their knowledge or consent?


ctrlshifti

I understand, but why is it widely used on railway infrastructure? Isn't it bad practice to force a caching layer on a user's API calls without their knowledge or consent?

domehane
FREE

2 months ago

Hello,

your assumption is correct. railway moved all customer domains behind fastly on february 19-20 2026 as an emergency response to a series of ddos attacks that hit their infrastructure. fastly is now acting as a waf/proxy in front of all railway-hosted services platform-wide, which is why you're seeing that new header. you didn't do anything wrong, it's a railway infrastructure change

source: https://blog.railway.com/p/incident-report-february-19-2026


domehane

Hello, your assumption is correct. railway moved all customer domains behind fastly on february 19-20 2026 as an emergency response to a series of ddos attacks that hit their infrastructure. fastly is now acting as a waf/proxy in front of all railway-hosted services platform-wide, which is why you're seeing that new header. you didn't do anything wrong, it's a railway infrastructure change source: <https://blog.railway.com/p/incident-report-february-19-2026>

ctrlshifti
PROOP

2 months ago

since railway now uses fastly by default, is it better to switch from cloudflare to fastly for my own edge layer to minimize latency via a single-provider network?


ctrlshifti

since railway now uses fastly by default, is it better to switch from cloudflare to fastly for my own edge layer to minimize latency via a single-provider network?

domehane
FREE

2 months ago

not worth switching for latency reasons yet. i think railway added fastly only for ddos protection, the cdn isn't live yet. so there's no single-provider edge layer benefit to gain right now. if you have no specific cloudflare rules you rely on, you can already turn off the cloudflare proxy, but wait until the cdn is fully rolled out before making any latency-based decisions


Status changed to Solved brody 2 months ago


Welcome!

Sign in to your Railway account to join the conversation.

Loading...