Try doing the following:
1.) Download the library files instead of using a CDN.
2.) Increase your page time-out time on your AWS server. It’s possible that you don’t have enough time to process all those records.
4.) Try another server. There might be restrictions on AWS that you can’t control (e.g. memory or “time-to-live” for your connection), but if you set up your own personal dedicated server for testing, you could rule that out.
5.) Disable your “ALL” option. Do you really want people to pull 16k records at once?
6.) As a last resort, try making a server-side pagination script.
This looks to be a problem with the S3 request expiration:
<?xml version="1.0" encoding="UTF-8"?>
<Message>Request has expired</Message>