f740c960-4ea5-4e0d-a8fc-7b3efef00d50

I used to get an error SSL SYSCALL error: EOF detectedError but now it just states connection already closed

going to need more info, what kind of app is this?

I have a Django backend application that is deployed on your platform that interacts with a postgreSQL database on your platform as well. I have Django handling the backend and React handling the frontend. It is a SPA application that offers user authorization, interaction with forms where the user enters a numerical set and that set is used to create a matrix and then deconstructed into a set of numerical sets and then displayed to the user on the frontend. Each time a user enters a set the backend/frontend updates (decrements a daily entry count). There are several pages that are injected via React router. I have not had any issues with this part of the application. However, when a standard user navigates to the payment page and makes a purchase using the Stripe API, the transactions go through and the webhook executes after a successful payment ( I have no issues there either). The webhook uses an apScheduler to setup a cron job 30 seconds after a successful payment, and it is during this time that causes the error. So the cron job executes 30 seconds after the webhook finishes execution. In this function the user is accessed via stripecheckoutid (foreign-key) and then updated in the user_table (changes from standard user to whatever membership was purchased) and updates entry fields as well. It is only during this process that I see the error. I am sorry if I am being to Verbose.

What does not make sense is that I am able to update the users during their sessions in the database while interacting with the frontend (decrementing available entries). But as soon as a payment is made and I go to update the user so they have these features I get an error that connection is closed.

is your app using different database clients for the different parts of the app?

Do you mean is the database that I have been using that is on your platform providing everything I just mentioned?

for the record, i dont work for railway, so it would not be my platform

Sorry lol I just assumed

My mistake

not what i mean, does your app use different database clients for the different parts of the app?

No

I see people complain about psycop2. Could that be the issue?

the issue is that your database client is trying to use closed connections

I read on a thread that closing unused connections every hour worked for them. Unfortunately it is not working for me

Do you have any suggestions?

make sure the client isnt trying to use a closed connection

By the way I am new to full stack development and I appreciate your patience. I just graduated last may and built the entire application from scratch so…

Have you worked with Django

i have not

So here is the info from the Django documentation. The default value is 0 , preserving the historical behavior of closing the database connection at the end of each request. To enable persistent connections, set CONNMAXAGE to a positive integer of seconds. For unlimited persistent connections, set it to None .

So maybe I should change the settings… What you do think? Or what would you suggest

it is clear from the information that you provided that database connections are not being fully closed, the client tries to use a closed connection and that is where the error is coming from

Thanks. By the way, I appreciate you taking the time to interact with me even though you do not work for Railway

happy to help where I can.

while I unfortunately can't give you a direct solution, I can at minimum tell you where the problem originates from.

👍