[FIXED] Proxy Redirect Temporarily Breaks when Backed Restarts

Jul 6, 2009 at 9:29 PM

I have URL Rewriter successfully proxying requests for a Ruby application to an instance of Thin running on port 3000. I have the Thin server setup as a Windows service as well using SRVANY. I am running into a minor problem though when I restart the backend Thin service. The frontend displays an error. If I wait a few minutes everything goes back to normal. Perhaps this is not directly related to the Managed Fusion product, but is there a way to increase the frequency that the frontend checks to see if the backend is responding so that I can make the window for seeing this error as small as possible?

Jul 6, 2009 at 9:35 PM

Nevermind, this was an ID10T error. I just did a bit more testing of the proxy with the Windows service component out of the loop and the delay is because Ruby is taking a bit longer to start up as a process than running the command directly in the console. The error is only visible while the backend is legitimately unreachable during thin's startup.

One possible suggestion though from this. Would there be any interest in fleshing out the reverse proxy capabilities to directly support mod_proxy style functionality to support multiple backend service points?

Coordinator
Jul 12, 2009 at 11:26 PM

I would be interested in understanding more about mod_proxy.  I found this resource on

http://httpd.apache.org/docs/2.0/mod/mod_proxy.html

But can you give me some concrete examples of how it is used?