4 months ago
Calls to service hang forever per https://station.railway.com/questions/limited-access-a79785fe
18 Replies
4 months ago
curl -v https://ph9km42x.up.railway.app/health
* Host ph9km42x.up.railway.app:443 was resolved.
* IPv6: (none)
* IPv4: 66.33.22.204
* Trying 66.33.22.204:443...
* Connected to ph9km42x.up.railway.app (66.33.22.204) port 443
* ALPN: curl offers h2,http/1.1
* (304) (OUT), TLS handshake, Client hello (1):
* CAfile: /etc/ssl/cert.pem
* CApath: none
* (304) (IN), TLS handshake, Server hello (2):
* (304) (IN), TLS handshake, Unknown (8):
* (304) (IN), TLS handshake, Certificate (11):
* (304) (IN), TLS handshake, CERT verify (15):
* (304) (IN), TLS handshake, Finished (20):
* (304) (OUT), TLS handshake, Finished (20):
* SSL connection using TLSv1.3 / AEAD-CHACHA20-POLY1305-SHA256 / [blank] / UNDEF
* ALPN: server accepted h2
* Server certificate:
subject: CN=.up.railway.app
* start date: Feb 4 14:01:41 2026 GMT
* expire date: May 5 14:01:40 2026 GMT
subjectAltName: host "ph9km42x.up.railway.app" matched cert's ".up.railway.app"
* issuer: C=US; O=Let's Encrypt; CN=R13
* SSL certificate verify ok.
* using HTTP/2
* [HTTP/2] [1] OPENED stream for https://ph9km42x.up.railway.app/health
* [HTTP/2] [1] [:method: GET]
* [HTTP/2] [1] [:scheme: https]
* [HTTP/2] [1] [:authority: ph9km42x.up.railway.app]
* [HTTP/2] [1] [:path: /health]
* [HTTP/2] [1] [user-agent: curl/8.7.1]
[HTTP/2] [1] [accept: /*]
> GET /health HTTP/2
> Host: ph9km42x.up.railway.app
> User-Agent: curl/8.7.1
> Accept: /
>
* Request completely sent off
< HTTP/2 404
< content-type: application/json
< server: railway-edge
< x-railway-edge: railway/us-east4-eqdc4a
< x-railway-fallback: true
< x-railway-request-id: muOXK4qGQwWdN817g4a9AQ
< content-length: 101
< date: Wed, 18 Feb 2026 03:48:35 GMT
<
* Connection #0 to host ph9km42x.up.railway.app left intact
{"status":"error","code":404,"message":"Application not
found","request_id":"muOXK4qGQwWdN817g4a9AQ"}% 4 months ago
sorry that specific language "hostname changed" was the output of an llm in determining what happened. I didn't mean to imply that did happen, just that the llm thought that's what happened.
The curl is directly to my service bypassing cloudflare.
"status":"error","code":404,"message":"Application not found","request_id":"muOXK4qGQwWdN817g4a9AQ"4 months ago
Great, I'll let the team take a quick look as it's awfully close to an incident. Could you double-check that your service has that exact domain attached to it and that it has an active deployment?
4 months ago
how would I check? I use a custom domain but the cloudflare dns points to what I provided. It worked immediately before the incident, that's what I'm going off of. How would I confirm that that was the domain you gave me to map my cname entry to?
4 months ago
Oh, what you shared is a CNAME pointing? That won't work. You have to use your domain instead of that. Railway detects your service by your domain and not CNAME.
4 months ago
that fails as well
4 months ago
so I curled to what cloudflare is trying to find
4 months ago
btw this curl did work yesterday
shallow-alchemy
that fails as well
4 months ago
Do you get any sort of logs on http logs when trying to reach it via Cloudflare?
4 months ago
none no
4 months ago
On my end
Seeing ~15s response times on all requests to my deployment, including endpoints that do zero I/O (auth rejection returns 401 but takes 15s). This started ~75 minutes ago per Sentry.
Health endpoint returning 502 through Cloudflare due to timeout, also ~15s when hit directly.
Single uvicorn worker, no recent deploys. Suspect this is related to the current platform incident even though it's listed as builds-only.
Cloudflare:
curl -I https://caucus-ai.com/health
HTTP/2 502
date: Wed, 18 Feb 2026 03:58:16 GMT
content-type: text/plain; charset=UTF-8
content-length: 15
cache-control: private, max-age=0, no-store, no-cache, must-revalidate, post-check=0, pre-check=0
expires: Thu, 01 Jan 1970 00:00:01 GMT
referrer-policy: same-origin
x-frame-options: SAMEORIGIN
server: cloudflare
cf-ray: 9cfa9a87590aba52-SEA
alt-svc: h3=":443"; ma=86400
jackwelty
On my end Seeing \~15s response times on all requests to my deployment, including endpoints that do zero I/O (auth rejection returns 401 but takes 15s). This started \~75 minutes ago per Sentry. Health endpoint returning 502 through Cloudflare due to timeout, also \~15s when hit directly. Single uvicorn worker, no recent deploys. Suspect this is related to the current platform incident even though it's listed as builds-only. Cloudflare: curl -I <https://caucus-ai.com/health> HTTP/2 502 date: Wed, 18 Feb 2026 03:58:16 GMT content-type: text/plain; charset=UTF-8 content-length: 15 cache-control: private, max-age=0, no-store, no-cache, must-revalidate, post-check=0, pre-check=0 expires: Thu, 01 Jan 1970 00:00:01 GMT referrer-policy: same-origin x-frame-options: SAMEORIGIN server: cloudflare cf-ray: 9cfa9a87590aba52-SEA alt-svc: h3=":443"; ma=86400
4 months ago
Please open your thread.
Status changed to Awaiting User Response Railway • 4 months ago
4 months ago
why is my thread set to awaiting user response? to what?
Status changed to Awaiting Railway Response Railway • 4 months ago
4 months ago
I think I'm experiencing this as well. Getting a 502 BAd gateway from Cloudflare
Status changed to Awaiting User Response Railway • 4 months ago
4 months ago
For redeploying you can hit Ctrl + K -> Deploy latest commit with the service panel opened.
Status changed to Awaiting Railway Response Railway • 4 months ago
4 months ago
My service is back. To be clear I had redeployed 2 times. Seems back up
Status changed to Solved shallow-alchemy • 4 months ago