#Jelle-express-onboarding
1 messages ยท Page 1 of 1 (latest)
Sure ๐
can you share the example request IDs for creating an account and an account link?
Yep one moment
Account create:
curl https://api.stripe.com/v1/accounts \
-u sk_test_[blank]: \
-d type=express
Account link create:
curl https://api.stripe.com/v1/account_links \
-u sk_test_[blank]: \
-d account=acct_[blank] \
-d refresh_url="https://example.com/stripe/reauth" \
-d return_url="https://example.com/stripe/return" \
-d type=account_onboarding
There are various reasons why it'd directly take you to the refresh_url
https://stripe.com/docs/connect/express-accounts#refresh_url
- The link is expired (a few minutes went by since the link was created).
- The user already visited the URL (the user refreshed the page or clicked back or forward in the browser).
- Your platform can no longer access the account.
- The account has been rejected.
It has been rejected
Because it lacks data
The TOS date and acceptance and external account, but this is exactly why I should redirect them to the stripe onboarding url, for them to enter the bank account etc?
Sorry it actually says rejected in the dashboard and disabled in the response from account/create
Have you tried creating a new account link for re-onboarding?
I haven't
One sec ๐
When redirecting to stripe onboarding I don't need to pass any query params, do I?
Nope. account links are unique and should have all the info needed
Sorry what do you mean by re-onboarding? Should I set type: re-onboarding?
I thought I should call it again with the same params to do that
that's not a type sorry
you'd basically use the same params
Okay thanks ๐
I had a demo 3 hours ago for this lmao
The re-onboarding does the same thing, straight to the refresh url
Did you want like actual request Ids?
This is a re-onboarding response
<Stripe.StripeResponse status=200 Request-Id=req_n80AgEy09Dckoh Date=2022-05-24T14:19:48>
Thanks for the request ID, checking in to it...
Thanks, let me know if you need anything from me ๐
Will do, thanks for your patience so far. I am still not finding much on this and have reached out to my colleagues for help looking in to this.
Are you sending the link anywhere via slack or Discord or anywhere that will auto-expand the link? Even accessing the link like that can expire it
Nope I get this fresh from the api call and immediately go to it myself
I talked to stripe support about it and they say it's unexepected behavior and they will get back to me
But my explanation here is a lot more technical
I wanted to see here if maybe if I explain more in detail and more clearly you might be able to see that I'm doing something wrong
Definitely. I am checking now to see what our logs say about how/when this link is accessed. From what I can tell the account and connection to the platform are fine
And the requests I'm sending? They should work, no?
Curl works ........
zzzzzzzz
Wait, as in when you make the call with curl the link works?
But not when you make it with .NET?
yep
Can you send me the request ID that your curl command made? You may have to go to your dashboard for it https://dashboard.stripe.com/test/logs
Sign in to the Stripe Dashboard to manage business payments and operations in your account. Manage payments and refunds, respond to disputes and more.
req_EZdZXOUwEgPQew
The requests and responses look the same to me.. haha
Just made another one from .NET: req_d5BkTQozDzGICE that didn't work
Interesting. I am not seeing a difference in what our backend receives. This is really stumping me at the moment on what the difference might be.
And you said that your .NET code just basically passes you the URL and you navigate right to it?
I am actually seeing a facebook bot trying to access the URL from your .NET library. Are you sending these links to yourself on Facebook?
Anything that tries to preview the link like this can invalidate it
Facebook?
I dont send them anywhere
I see them in my IDE and then paste them in the browser?
Hi ๐ I'm stepping in for @winged crest I'm just catching up gimme a sec
Im getting the entire stripe staff on board here :D
A Facebook bot accessing my url that i get directly from my IDE after making a call... ๐ค
So you are creating onboarding links directly in your IDE, copying the URL into a browser, and still getting a redirect?
Yes
Straight to the refresh url
But in curl it works and I copy the url from the terminal in the same IDE
How do you know it's a facebook bot? Is the link being shared on facebook and clicked on from there?
I am unclear on why a facebook bot navigated there but I see something with a user-agent of facebookexternalhit/1.1 Facebot Twitterbot/1.0 trying to access the URL that you created
And I don't see that for the one you created via curl
What the .....
Omg
No way
I think i know
I have an app on my mac that keeps track of my clipboard history
And it resolves the link to get a preview
oh wow
I think that might be it
I have never seen that edge-case but I'll add this to my list of "what can expire my account link" now ๐
But then why doesnt it happen from curl
Well yeah anyway it's something like that I think
yeah it's quite common that weird behaviour. In a way it's good because it forces everyone to implement the refresh_url flow :p
I know!!!
Thanks so much you guys ๐ฅฐ
Happy to help, glad you figured it out1