findItemsAdvanced "The remote server returned an error: (500) Internal Server Error."
I'm getting a TON of "The remote server returned an error: (500) Internal Server Error" errors from findItemsAdvanced API calls. - -No changes to my app in months. - -I get the error in the sandbox and in production. - -I do not get the error when calling the service endpoint with my credentials in a browser. The difference is parameters via URL in browser vs. serialized XML (SOAP) sent via my app.) - -I get the error from my app 60-70% of the time and am unable to repro on demand. Is anyone else noticing a high volume of failed API calls? What is the correct support avenue for this type of issue?
This just starting happening to my app. I have a Microsoft Azure-hosted
VB.Net webjob which makes findItemsAdvanced API calls. It runs 24x7 and works for several days, and then it aborts with the following error: The remote server returned an error: (504) Gateway Timeout. Any ideas why? Any ideas how to troubleshoot? Thanks in advance!
Sometimes a 500-error also includes body error message text in the response. Can you see/capture that? Also check the link/thread in my original response here in this thread, whereby someone suggests that adding a small header line in the request fixes things.
I added the header line and it did not resolve the issue. It's odd because the app makes several thousand calls per day and only a handful are experiencing this issue. I'll check to see if I can capture additional info pertaining the the error. I'm still not sure if the issue is with Microsoft Azure, eBay or my app. Thanks!
Turns out I wasn't trapping the error properly. I resolved the issue and now when I get a 50x error, I retry the API call and it works. Bottom line: a 50x error can result from an API call. Trap for it and retry. Typically only one retry is necessary.
https://httpstatuses.com/503 says plainly that this http code is returned when the service is not available. This means there the whole service is down -- unlikely -- or there are too many calls for the provisioned capacity (the number of computers in ebay's cloud tasked to handle these calls) and some are aborted with a 503. Bottom line is you cannot do anything about this except retry after a small delay. I suggest not trying *immediately* since that will only cause even more requests to be aborted with 503.
Proper trap & retry coding resolved the issue. My conclusion is that the issue is Azure-related. When running on my local dev box, I never get a 50x error. I only get the error when running on Azure. I'm guessing it's caused by an Azure maintenance task. Thanks