RE: [squid-users] Forcing Squid to cache all pages in accelerator mode

From: Elsen Marc <elsen@dont-contact.us>
Date: Thu, 25 Nov 2004 10:08:56 +0100

 
>
> All
>
> I have configured squid to run in accelerator mode in front
> of our application server.
>
> ie
>
> User Browser <--------> Squid <--------> Application Server
>
> In our case, all URL's uniquely define a page, so we can
> cache them with no fear of any session information meaning
> that one user should see a different page for the same basic
> URL. The URLs are generally of the form
> http://<host>/a_jsp_page?arguments. As our underlying data
> only changes daily, i would like to just generate each URL
> only once, and let squid serve copies to everyone else.
> Hence, I also restart squid daily and clear the caches down.
>
>
> I believe the following refresh_pattern sould do the job
>
> refresh_pattern . 1440 100% 1440
> ignore-reload override-lastmod override-expire reload-into-ims
>
> Unfortunately, this does not force all object to be cached -
> I still get some TCP_MISS's for identical requests. Some
> dynamic pages do get cached, but not even all the static
> images are cached (as I can see via the TCP_MISS messages in the log).
>
> Is there a way that I can configure squid so that I can see
> why the URL is a TCP_MISS (ie expired or whatever)? Is there
> something else I am missing, to force squid to cache everything?
>
> I am using 2.5-STABLE on RedHat 9
>
> Any pointers would be appreciated
>
> Many thanks to you all
>
 
 As to the why of the MISSES :

    http://www.ircache.net/cgi-bin/cacheability.py

 may help.
 Probably this tool can be 'saved' and used in a local context too.

 M.
Received on Thu Nov 25 2004 - 02:10:38 MST

This archive was generated by hypermail pre-2.1.9 : Wed Dec 01 2004 - 12:00:02 MST