Hi, I have a very strange problem that I don't know why it happens. I'm creating a REST API service that wraps another REST API payment service, let's call the other services XPAY. It has various payment options (Card, eWallet, Kiosk, and Cash). The steps required to use the services (to make a transaction)
1. authentication 2. order registration 3. payment key 4. prepare frontend based on payment type 5. Callback URL 6. Redirect URL and each step depends on its predecessor. The first 4 steps are performed through HTTP Requests that I issue from my service. I store some of the information I receive in those steps in a Model called Transaction. 5th and 6th steps are Callback endpoints that I create to receive further information about the status of the transaction that I started in the first 4 steps. My endpoints design is: myservice/payment/card myservice/payment/wallet myservice/payment/kiosk myservice/payment/cash myservice/payment/callback-url myservice/payment/redirect-url These endpoints are wrappers for the payment methods available in XPAY's service that my frontend apps will use to provide those payment options for users. I implemented the first 3 endpoints with no problem. inside those endpoints, I create a transaction object with a unique id and then update it from inside the *callback-url endpoint* after I receive the information from XPAY's service. Note that it takes some time between creating the Transaction object and saving it into the database from inside my endpoint, and updating it from inside the callback url endpoint. The weird thing happens in the 4th endpoint `cash` endpoint. XPAY's service responds very quickly on my *callback-url endpoint* and when it gets to the updating step, I get *Transaction.DoesNotExist *Exception while getting the transaction object using Transaction.objects.get(id=id). After some debugging time using ipdb I found that the instance has already been saved to the database! and I can retrieve it with its id. but when I let it run ordinarily the saving operation doesn't finish until that time (I got to know that using some print() statements), which is very strange, it's impossible that my laptop is that slow! Solutions I tried: 1. atomic transaction 2. time.sleep some time after making the request and saving the transaction object to the db 3. time.sleep in the callback-url before getting the transaction from DB I checked any possible logic error related to it and there's nothing illogical. the issue is related to Postgres and how it operates, and also with Django's ORM. I don't know. Main technologies I use: Django 2.2 Postgres Docker container DRF I hope someone gives me an insight into why this is happening and how to work around it. Thanks -- You received this message because you are subscribed to the Google Groups "Django users" group. To unsubscribe from this group and stop receiving emails from it, send an email to django-users+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/django-users/e43be322-bab9-4162-9fbc-05c6c9b95c94%40googlegroups.com.
