On Mon, Apr 30, 2012 at 8:15 AM, Plüm, Rüdiger, Vodafone Group <ruediger.pl...@vodafone.com> wrote: >> -----Original Message----- >> From: Kaspar Brand [mailto:httpd-dev.2...@velox.ch] >> Sent: Sonntag, 29. April 2012 09:59 >> To: dev@httpd.apache.org >> Subject: Re: [users@httpd] SNI with apache 2.4.1 reverse proxy >> >> Whether that is desired or not probably depends on a judgement of >> possible use cases. As far as I'm concerned, I'm not convinced that >> there are really good reasons for "playing the ProxyPreserveHost trick" >> (neither for https nor for http, actually), but YMMV. > > The main reason I see for the "ProxyPreserveHost trick" is a reverse proxy > setup, > where the backend application builds full qualified links into its HTML > output based > on the information it gets from the Host header. The only other approach in > this case > I can think of would be working with host entries on the reverse proxy to > resolve the name > of the reverse proxy to the IP of the backend which is not really nice either. >
All of our sites (229 and counting) are generated by application servers that are reversed proxied by a main Apache server, and every single one of them uses ProxyPreserveHost for precisely this reason. Cheers Tom
