Receiving NoHttpResponseException for Verified and Enabled Fitbit Subscriptions

[varun271]

Hi, I have created a private issue using google issue tracker as well, but thought it would be good to post in the forum in case anyone else has seen this issue. Our application is currently set up to use Fitbit subscriptions to update a customer’s data on our end whenever new data is detected. We have set this up in our Dev and QA environments which is currently working.

We have a separate application on the Fitbit Developer Portal for Prod which uses the same code. The Prod application the Subscribers endpoint was verified on January 24, 2025, and was working at some point since we have logs of subscribing some users.

However now we keep getting NoHttpResponseException errors. None of the requests are going through to our application from the Fitbit developer portal side, but when we try making the call to our exposed application endpoint manually it is going through and we are receiving logs. I was hoping to find some clarity for why the POST requests from Fitbit aren’t going through to our endpoint.

Another thing to note is that the subscriber automatically gets disabled without any warming.

We are not seeing SSL errors, just NoHttpResponseException errors as shown in the image below. Is there a way to get more detailed logs of what is being sent out to help us track this down.

[GordonFitbit]

Just to circle back on this question based on the private conversation.

Our logs indicate that notifications are being sent from Fitbit to your subscriber successfully. For example, this is a log of a notification being sent to your application.  The SocketTimeoutException is telling us that your application is not returning an HTTP 204 No Content when the notification is received, as documented in our Responding to Notifications documentation.

This is then followed up by a NoHttpResponseException error indicating that your application failed to respond to the Fitbit server with the 204 No Content status code. 

The user is checking with their middleware team to see if the firewall is letting the requests through.   We have to be able to communicate with the subscriber for it to receive the notifications.