Hi CF Team!
I am really enjoying exploring the Workers functionalities and I am starting to imagine all the possibilities that they will enable us to do.
So here is my idea to which I would like to get your opinion if it is a good case for Workers, and if it’s reasonable to expect it to work within the CPU time boundaries.
I have a project with some files that have to be served to users. Files are being stored on several file servers. Eg.
But, for security reasons, we don’t want the end user to know the exact server/URL path where the files are being stored (like the ones above).
That’s why we are using a reverse-proxy server to handle the requests, lookup the proper file server for particular file, and serve it back to the user. All that is being handled through a common domain, the links that users see and use are eg.
Do you think it’s possible to use Worker to replace the reverse-proxy server, request the proper file server name using some cached JSON data file and start passing through the file data from the storage server to the browser, hiding the real file location from the visitor?
As files can be several tens of megabytes big, and the download could take some minutes or an hour, I guess the Streaming Responses approach to pass-through the file data stream would be the best way to go?
Thanks for your opinions and any advice in advance!