> Matus UHLAR - fantomas wrote:
> >However, what I am repeating here is, that the difference between this:
> >
> >client ====> server
> > HTTPS
> >
> >and this:
> >
> >client ====> proxy ====> server
> > HTTPS HTTPS
> >
> >network structure is, that second one has one more weak place - the proxy.
> >Although the second structure CAN work and possibly DOES work somewhere,
> >it MAY be just a result of wrong decision or implementation
On 12.10 14:03, Neil A. Hillard wrote:
> There are a couple of reasons that I can think of that require this
> configuration:
>
> 1) Where you don't trust the security of the connection between the
> reverse proxy and backend web server and
Yes, but I already told that :)
> 2) Where the backend web server insists on generating URLs based on the
> protocol used to communicate with it. e.g. https to the reverse proxy,
> http to the web server and it generates HTML with http:// URLs.
In such case it's problem of braindead dynamic pages, that use absolute URIs
;) Yes, this problem can be avoided by having ssl in outgoing requests from
proxy. But that's a workaround to this problem...
> We moved away from squid as a reverse proxy to Apache with mod_proxy,
> mod_rewrite and mod_proxy_html (from Nick Kew). This allows us to fully
> rewrite the HTML from the backend web server and change links for
> external access. This way we can consolidate multiple backend servers
> into a single certificate and we use strong authentication so this
> ensures that the users only have to authenticate once.
... and this is another workaround :-)
the fix is: change the scripts to generate only relative URIs ;)
-- Matus UHLAR - fantomas, uhlar@fantomas.sk ; http://www.fantomas.sk/ Warning: I wish NOT to receive e-mail advertising to this address. Varovanie: na tuto adresu chcem NEDOSTAVAT akukolvek reklamnu postu. I don't have lysdexia. The Dog wouldn't allow that.Received on Wed Oct 12 2005 - 12:18:22 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Nov 01 2005 - 12:00:04 MST