-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Captcha Harvester "ERROR for site owner: Invalid domain for site key" #204
Comments
Might need your expertise on this when you get a moment. Captcha window is useless till this gets patched. |
@pr1sm ^^ sorry forgot to tag you. Get to this whenever you can! Or if you have any tips, let me know! |
Hmm, yeah I remember this being a problem when I was first implementing the captcha page, not sure how I initially solved it... Will take a look back at the commit history and see if there are any relevant messages. |
Sounds good! Yeah I'm not sure either :/ |
Can't seem to download your debug file -- says server not found... How big is it? can you dm it to me? |
Hmm that’s weird.. Yeah I’ll dm right now |
I'm unable to reproduce this error message in our frontend. I'm using the debug commands (available in Is there a particular site on which the captcha harvester is failing? Does this work for you in |
I believe you’ll need to actually run this on a task for it to reproduce why it’s failing. I think the problem lies with the fact we’re solving the captcha on our machine, not the respective site domain. We may need to reconfigure a few things to get it to work. I think if you run enough tasks on kith it’ll require a captcha from you. That’s where the debug file is from |
I'm not sure how much I can help with this issue since I can't seem to get a captcha to trigger for me and when testing manually (via the debug commands) everything works fine. From the html, I'm not sure if the error message loads when the captcha loads or after attempting to solve the captcha. I'm unclear about what you mean by "not on the site domain" since the js for a catpcha is always run on the local machine. Is it possible our shopify captcha key is wrong? It seems like we are always using the same key ( A solution that might work is having our captcha server act as a proxy for the actual site domain. We could attempt to go to the specific site, hit our proxy (the captcha server), then serve the same html instead of proxying the request. This would allow the |
Yeah this is what I meant by "not on the site domain." Sorry the lack of clarity, it was late and I was pretty sleepy. I do think this is our only issue though, the site key should be fine. |
I've gone ahead and created a way for us to always receive a task:
|
Hmm, I'm not able to reproduce, but it's because I can't even create a checkout session (always responds with 422 response). I think I might have to wait until I'm back in the States to investigate this further... |
Here's a recently updated captcha harvester we can compare against: https://github.com/ItsFuego/Captcha-Solver |
Thanks for linking the updated harvester! I wasn't familiar with Electron's I'm gonna push what I have now, but there's still some code cleanup to do -- will put up a PR tomorrow. |
This commit adds a protocol interceptor to load our captcha html instead of the regular html for other sites. This commit also updates the captcha server middleware stack to better handle loading assets such as images/styling/js while still redirecting everything else to the captcha html page. Issue: #204
* Handle Captcha State This commit adds a missing handler for the captcha state. This uses the checkouts handle request captcha method to request a captcha and wait for the token. The handler has also been updated to include error checking. * Use Protocol Interceptors to Load Captcha Page This commit adds a protocol interceptor to load our captcha html instead of the regular html for other sites. This commit also updates the captcha server middleware stack to better handle loading assets such as images/styling/js while still redirecting everything else to the captcha html page. Issue: #204 * Add Temporary Workaround for Unhandled Size Option Formats Some sites include unsupported size option values, which affects our variant group matching. This commit adds a temporary workaround until the actual issue (#239) is addressed. Additionally, a linter error is fixed, some unnecessary log statements are removed, and another log statement is adjusted to only display relevant information. * Fix Various Captcha Related Bugs This commit addresses some (not all!) captcha related bugs: - Captcha Window Reloads are now handled properly - Redundant Captcha Reset calls are removed - Attempting to add >5 captcha windows no longer crashes the frontend These are some small bugs that fall under the scope of this branch. Other larger bugs that require separate issues are not fixed in this commit.
Describe the bug
The captcha harvester works really well, but seems to have a small bug when trying to validate the site domain. I'm assuming this is due to the fact that we are technically the domain of the harvested token, not the respective task site.
To Reproduce
This is a little hard to reproduce, so I've attached a file (see end of OP) to help show you exactly what I'm experiencing.
Expected behavior
The harvested captcha response should validate properly and be accepted by the post payment method on the task runner side. Since there's no changes that need to be done there (or so I hope) I'm just opening up this issue to track the frontend changes.
Additional context
debug file –– rename to
debug.html
and open in browser to see what i meanThe text was updated successfully, but these errors were encountered: