rproxy is an HTTP file caching proxy intended to save bandwidth by caching updates to packages in Linux repositories. When a device connects to rproxy and requests a file, it will be downloaded to the device and stored on the machine running rproxy. Any later requests for the same file by other devices will be served from the local cache, saving bandwidth and potentially allowing for faster downloads.
rproxy is a typical Cargo project.
Building is as simple as:
cargo build --releaseTo build with HTTPS support:
cargo build --features https --releaseThe binary will be built in target/release/rproxy.
rproxy needs to know where to store and look up any cached files it downloads.
You can set this by defining the X_PROXY_CACHE_PATH environment variable
to a folder that allows read/write permissions.
If the path doesn't already exist, rproxy will attempt to create it.
export X_PROXY_CACHE_PATH="/tmp/rproxy"
./rproxySET X_PROXY_CACHE_PATH="C:\Temp\rproxy"
rproxy.exerproxy can optionally bind to a particular network address.
You can set this by defining the X_PROXY_HTTP_LISTEN_ADDRESS environment variable
to an address and port.
When X_PROXY_HTTP_LISTEN_ADDRESS is not set,
rproxy will default to listening for any address on port 3142.
X_PROXY_HTTP_LISTEN_ADDRESS="127.0.0.1:8080"X_PROXY_HTTP_LISTEN_ADDRESS="[::1]:8080"
To test that the proxy is working on the same machine with wget run the following command twice
http_proxy=http://127.0.0.1:3142 https_proxy=https://127.0.0.1:3142 wget --no-check-certificate https://github.com/Lethja/rproxy/archive/refs/heads/master.zip
Using
--no-check-certificateto test the proxy is safe in this case since127.0.0.1is a loopback address to the same machine.
rproxy will still verifygithub.coms certificate when it makes the request.
Cached content never expires. If the rproxy cache disk has low free disk space, you will need to manually delete files.