On Fri, 2011-09-09 at 08:29 -0700, Patrick Lightbody wrote: > Thanks for sharing the new async stuff. I'm curious about the example you > linked to: NHttpReverseProxy.java. > > Can you share a bit more about what that does exactly?
You can point NHttpReverseProxy at an external site (say www.apache.org) and use a browser to access content of the target server through the proxy. The client (browser) does not even need to know that the site it is interacting with is actually a proxy. Hence the term reverse proxy. > I always get a bit confused when talk of "proxies" and "reverse proxies" > come up. I ask because I have an open source project that is using an old > Jetty 5 proxy (w/ mods) that was capable of doing SSL man-in-the-middle > proxying (provided your browser had a special CA installed). Wondering if we > might be able to simplify the codebase with some of the stuff you've been > working on! > Proxies based on an async I/O model may be better suited for handling very large number of connections most of which stay idle most of the time. However, if you need to handle data transferred through the proxy using standard InputStream / OutputStream abstraction the asynchronous I/O model immediately becomes a royal pain in the rectum. You'll be much better off using a servlet engine such as Jetty if your proxy is intended to handle only a fairly moderate number of concurrent connections (<500) and its content processing logic is based on InputStream / OutputStream abstraction. Hope this helps Oleg --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
