Open candrews opened 12 years ago
Actually, my proposed solution wouldn't work if there are multiple web servers using one DFS root (which is one of the recommended RR setups, and one that I'm using).
I think a better solution would be for RR to add cache dependencies to the response for each RequestReduceContent file used in that response. That way, if the RequestReduceContent file is deleted, ASP.NET would invalidate the cache for the page that uses it, causing RR to reprocess the page.
The scenario is: An ASP.NET page has output caching enabled (say the duration is 10 minutes). The page has RR-able content on it. The first request, the non-RRed page is cached (see #190) Wait until the cache entry has expired (10 minutes in this example) Now that ASP.NET page is cached with all js and CSS RRed Go the the RR dashboard, and flush content View the ASP.NET page again - all the RR references are now 404s
This happens because when the ASP.NET page is cached, the RR filter isn't executed, so the RR content isn't regenerated.