Thank you for your post.
The URL you entered is a web page that does not exist, but filemaker.com captures bad pages and redirects the user to a default page ("Looking for something at FileMaker?")/ Since the resulting URL has a web page, the data is retrieved and LastError will be 0 (zero). More web servers are redirecting bad web pages, so the number of 404 errors are decreasing.
If there is no server for a domain, then you will receive a 1631 error. (Just verified)
See your Inbox at the top of this page for more information.
Thanks for the detailed information from your private message. In essence, capture the 404 error prior to the redirect.
I have submitted your comments to our Development and Testing departments for review. As information becomes available to me, I will let you know.
Just following up on this one. FileMaker also behaves differently in a 404 with an empty response ( Content-Length: 0 ) vs a 404 with no redirect, but some extra data in the response. For example these two urls :
They will both produce 404 responses from the server, with no redirects.
The first one is empty and FileMaker gives an error code (10) in the result. The second one has some extra data and FileMaker doesn't give any error code, where I think it should.
I'm not sure if this requires it's own bug report or not. It would be nice if the response from FileMaker ( in terms of error codes ) was able to be consistently mapped from web server response codes to FileMaker response codes, regardless of content. Even better would be a way for FileMaker to have it's own error code section ( like maybe 11xxx ) where the xxx was the three digit web server response code, so we could parse every code individually.
Thanks for the additional information.
I have attached your last post with suggestions to the original report.