#Jkas - consumed sources
1 messages · Page 1 of 1 (latest)
Another example: src_1M3oUaEkTO8kaKcKyRrxr9AJ
I'm taking a look but there isn't much I can see besides that both sources show up as consumed. Doing a little more digging
Are they both apple_pay -> apple_pay?
Thanks for your help looking into this with me.
Another example: src_1M7NlhEkTO8kaKcKM2zCfys0
It's too bad apple pay needs to be stored as a source, as opposed to a newer payment method.
The associated Payment Intents show a large number of failing attempts to /pay
Why is this? I wonder if it's buggy because of this fact.
pi_3M6dxoEkTO8kaKcK0aJF0pp8
We wish we could store everything as a pm_
That would make this easier to debug
Those large number of failing attempts is a sidekiq job on our end retrying the paying of unpaid invoices, I think. That will be resolved on our end since we have the 4x auto-retries setting enabled in our Stripe account.
So I am seeing these sources being created with a call to /v1/sources like this:https://dashboard.stripe.com/logs/req_kE3mGz2jUnqz91
I'm reviewing one of the customers involved (cu_1FwiJcEkTO8kaKcKZqKUpbKj) and I am seeing a regular updating of default_source settings, pretty much every month
Jkas - consumed sources
Interesting. It looks like we started seeing this issue around Oct 29. Backend doesn't deal with ApplePay sources - it must be a frontend frame or something that accepts those payment types.
But how they're getting auto-consumed and detached is beyond me. I'm looking through backend commits for any clues.
The source is created with your account's secret key so it must be occurring in the back-end
If that key is exposed on your front-end that is a major security concern as it grants a user full access to your account via the APIs.
So I very much hope it is occurring on the server
Attempting to track down where we'd be creating those in the backend.
Not sure if it helps but you're using the Stripe Ruby client library to make the requests.
That's correct.
I'll need to see if I can reproduce in staging, as I do have an idea of how to fix locally.
Thanks for your help! I'll start a new thread if I run into any other issues.
That makes sense. I just know some larger deployments run multiple languages for separate services so I thought it might help narrow down the codebase a bit. Feel free to come back if/when you have more questions. We'll be here.
Thanks again!