[Feature request]: Support S3 backend for file server

[shane-davidson]

Use Case

I'm trying to deploy catalog servers at scale and want a nice / easy way to server files.

Also offloading the file downloads to S3 would reduce the load on out puppet servers which hangs/return lots of EOF responses when pushing out new binaries to thousands of servers and I don't want to scale my servers based on the required load only when pushing new binaries.

Describe the solution you would like

Support S3 backend (maybe presigned URLs or proxied access) allowing openvox to serve files from a private S3 bucket

Describe alternatives you've considered

rsync files, mounted NFS, S3 fuse. All of which would are more complex and don't offload the download/HTTP requests

Additional context

No response

[bastelfreak]

Hi, thanks for the issue. Do you want to pass S3 URLs to every openvox agent and the agents connect o the S3 bucket / do you want to get rid of static files on the openvoxserver? Or do you want to speed up openvoxserver by bypassing the jetty Webserver for static files?

There are already a few tuning options available. For example https://docs.openvoxproject.org/openvox/8.x/static_catalogs.html. Another option that some people run is an haproxy or nginx in front of the openvoxserver. Those tools can terminate TLS and serve the static files.