On Fri, 13 Jun 1997, Wojtek Sylwestrzak wrote: > Most of the bigger caches I know try not to cache cgi scripts > (based on regex against the url of the cached object). Because > they are afraid cgi means dynamic contents that might not be > intended for caching.
Hmm, that's a bad decision, as some CGI output is eminantly cacheable. The CGI author should be responsible for setting appropriate last-modified or expires headers. > one more remark on caching - it's important that mirrors > look *exactly* the same as the original site (that's why > we call them 'mirrors' after all :-)) > Because some caches try to redirect to local resources. > E.g. I redirect all cache users in Poland to our mirror > of www.apache.org (this is done on a top level national cache), > so that when a request for http://www.apache.org/ comes to the cache, > it substitutes it with http://sunsite.icm.edu.pl/pub/www/apache/ > (so it's actually not a redirect but contents substitution). That's also not a wise move. How does the bug database work? http://www.apache.org/bugdb.cgi No wonder we haven't gotten any user complaints from Poland :) Brian --=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-- [EMAIL PROTECTED] www.apache.org hyperreal.com http://www.organic.com/JOBS
