0007485: Order number 0 is not updated when you get stuck in the payment process.

ID	Project	Category	View Status	Date Submitted	Last Update

0007485	module Unzer	General	public	2023-06-16 15:27	2023-09-12 14:14

Reporter	Spritje	Assigned To
Priority	low	Severity	feature	Reproducibility	always
Status	resolved	Resolution	fixed
Product Version	1.1.0
Target Version	1.1.2	Fixed in Version	1.1.3

Summary	0007485: Order number 0 is not updated when you get stuck in the payment process.
Description	According to the OXID-ESales UNZER module log, the UNZER API was unreachable around 23:11 - 23:13. Apparently, the module was not able to write the order number correctly after the API was available again. Thus, first an error was received in the log and later a success was reported in the log for the order. However, no order number was written, but the order number remained at "0".
Steps To Reproduce	Not reproducable. The order number is written when you come back from Unzer. So in the middle of the payment process. The way in description described, the customer seems to be stuck exactly in this hole.
Additional Information	- es - A possible solution: I could imagine that for webhook calls that find an entry, we could look to see if an order number is there. If no, then it will be set. Just like the "paid" date. The order number is in turn transferred to Unzer later, in the "order shipped" process.
Tags	No tags attached.

mario_lorenz 2023-07-11 11:34 developer ~0015416	If I understand correctly, Unzer's API didn't respond at a time when Unzer would even know about the order. This means that Unzer does not know anything about the order and will therefore not send any webhooks. In my opinion, the solution would be to pack this process into a transaction, since it is the initial moment that has to work so that webhook techniques can take effect later. In this case, which the customer describes, a transaction rollback would have had to take place and the customer would have had to be brought back into the shopping cart. That's why I see this as a feature request, since the problem occurs whenever sub's API is down.