Re: caching dynamic pages (CGI)

DOH! I just figured it out. My form uses an image as a submit button,
and clicking the image sends something like image.x=10&image.y=7 as part
of the URL. Every time these numbers are different, they get cached
separately.

Does anyone know a way to fix this problem by ignoring the name of the
image in the URL? Do I need to use a redirector?

- Perrin

Henny Bekker wrote:
>
> Hi Perrin,
>
> > > > Sorry to ask what seems like a stupid question, but how do I get Squid
> > > > to cache dynamically generated pages? I'm trying to use it for
> > > > acceleration of an existing server, and want to cache the results of CGI
> > > > requests for 24 hours, even though they don't send last_modified or
> > > > expires headers.
> > > >
> > > > The basic proxying/acceleration is working fine, but it isn't caching
> > > > the CGI pages. Do I need to turn off some stopwords somewhere?
> > > >
> > > For squid v1.1.* look in the squid.conf file for "hierarchy_stoplist" and
> > > "cache_stoplist".. The default setting is not to cache URL's with a "?"
> > > or "cgi-bin".. You also have to set the "refresh_pattern" for the URL's
> > > containing "cgi-bin" ..
> >
> > I removed the hierarchy_stoplist lines and the acl QUERY lines and I'm
> > still getting nowhere. My refresh_pattern looks like this:
> >
> Eh.. Removing the hierarchy_stoplist lines does not do the job because:
> The default is to directly fetch URLs containing 'cgi-bin' or '?'.
> and in the default situation, the hierarchy_stoplist are escaped (e.g. does
> not exist)..
>
> > refresh_pattern cgi-bin 600 0% 4320 ignore-reload override-expire
> > override-lastmod
> >
> > I know that's extreme, but I just want to see it work before I start
> > toning it down.
> >
> > I'm using 1.2beta25.
> >
> Also with Squid v1.2-beta.25 the remove of the "hierarchy_stoplist" lines
> activates the default situation, thus disabling URL's containing 'cgi-bin'
> or '?'...
>
> Cheers, Henny
Received on Tue Sep 29 1998 - 17:47:00 MDT