Webhooks - unexplained 408 timeouts

[sim...@thebutterflyclub.com]

Hi folks,

Beginning at 17:25 UTC on March 15, some of our `attendee.updated` webhooks requests begun failing as 408 (Request Timeout) but with a few inconsistencies from a regular, 'ole fashioned timeout.

In particular, forcing the request to fire again from the Eventbrite interface in `/account-settings/webhooks` does not increment the attempts counter. Also, requests that failed are not reattempted automatically (either that, or your Attempts counter is stuck on "1"). 

It's almost as if the requests were immediately registered as a 408 by Eventbrite, and never attempted again.

Our server logs confirm none of the affected requests ever made it to our infrastructure, but even more mysteriously, this problem only affects only one in every two or three requests - the others merrily getting through without issues.

I'd be grateful for any insight you can provide.

All the best,

Simone

[sim...@thebutterflyclub.com]

Update: the issue has improved, now forcing the webhook to fire again through the Eventbrite website behaves as expected: the request is re-sent, the attempts counter updated, etc etc.

There is a separate issue with the `attendee.updated` event not firing on attendee creation, I will start a new conversation about that to avoid confusion.

All the best,

Simone