Cap0n3
Database migration -
This morning I received this email below explaining to me that my database will be deleted on 31 of january if I don't migrate it (a bit harsh isn't it ...). The thing is that I already migrated all my databases one week ago and I don't have any banner action required anymore. I suppose this is an error on Railway's part but I need to be absolutly sure !
Could you enlighten me ? Thanks !
Project ID : 67ca2275-28a1-45b1-a973-b920d49e7ce1
=== Email ===
Hello,
We are reaching out to let you know that you have one or more databases in your Railway projects that need to be upgraded to the latest version.
Your database will be deleted by January 31, 2024 if you do not take any action.
Please sign into the Railway Dashboard and browse to your projects with an "Action Required: Migrate Database" banner to start the migration process immediately. Migration is a one-click process.
For further information, refer to our Migration Guide and our blog post on how we're improving databases on Railway.
If you have any questions, please ask them in the Databases V2 Migration discussion in our Community Forum.
Best,
The Railway Team
6 replies
Gunicorn timeout when sending POST between two apps
Hi there,
My project have two apps, APP_1 for handling everything concerning my website and APP_2 which acts like a proxy for all external API calls (to avoid csrf related issues). Everything run smoothly locally but when I'm in production on Railway, gunicorn seems to time out and crash when request between APP_1 and APP_2 is made.
In this specific case, APP_1 handles a form and send a POST request to APP_2 and it then communicates with Stripe API. The timeout crash happens when APP_1 uses request.post (python library) and POST request never reaches APP_2. (Actually, request does reach App2 but only after a timeout crash ... check logs below ...). I should add that in production, APP_2 online endpoint is working fine and I can reach it with curl command and proper authentification.
Could you help me there ? Or give me some pointers ?
LOGS
=> APP_1 receive form and send post request to APP_2
[2023-08-15 09:57:06,324] [INFO] {views -> post} - Will send request to https://www.mywebsite.com/api/proxy/stripe/
[2023-08-15 09:57:06,330] [INFO] {views -> post} - Membership form is valid
=> Here It's waiting to hear from APP_2 but it timeout
[2023-08-15 09:57:36 +0000] [1] [CRITICAL] WORKER TIMEOUT (pid:9)
[2023-08-15 09:57:36 +0000] [9] [INFO] Worker exiting (pid: 9)
[2023-08-15 09:57:36 +0000] [11] [INFO] Booting worker with pid: 11
=> And then, POST request is finally received after App1 has crashed ...
[2023-08-15 09:57:37,212] [INFO] {views -> post} - Stripe proxy is handling form submission ...
[2023-08-15 09:57:37,212] [DEBUG] {views -> post} - request.POST: <QueryDict: {...DATA...}>
[2023-08-15 09:57:37,213] [INFO] {views -> post} - Checking if customer already exists ...
[2023-08-15 09:57:37,418] [INFO] {views -> post} - Getting prices from Stripe ...
[2023-08-15 09:57:37,510] [INFO] {views -> post} - Creating checkout session ...
=> Here it sends back Stripe checkout session url but since APP_1 has crashed noting happens on website.
17 replies