Antony Stone wrote:
> 
> Hi.
> 
> I've just set up a squid system (2.5stable3) under Linux, and I've changed
> only the access control setting in the squid.conf file, to allow me access.
> It's working fine, and passing requests and pages through nicely.
> 
> However, I can't really see that it's doing much cacheing, when I would
> expect it to.   So far I'm the only user, testing the system, and I'm using a
> website mapping application which walks through every page on a website,
> following all the local links, as a way of generating some traffic through
> squid.
> 
> I'm visiting a couple of sites which have nearly all static content, so it
> should be cacheable okay.
> 
> The first time I run the app, and I watch squid's access.log output, I see
> lots of TCP_MISS entries, with the access going DIRECT to fetch the webpage
> from the server.
> 
> But it looks pretty much the same the second time I do the same thing - I get
> very few TCP_HIT, TCP_IMS_HIT or TCP_MEM_HIT entries, and my inbound traffic
> to the squid server is pretty much as high as it was the first time round.
> 
> I have two questions, which I can't find in the FAQ:
> 
> 1. Can I get any statistics out of squid to tell me what size cache it's
> using, how full it is, and how quickly it's churning the entries?

 Use squid's cachemgr interface to obtain various stats/info
 about the cache.

> 
> 2. Is there anything else I should be changing in the squid.conf file to get
> good cacheing as well as proxying?
> 
  Not really, 'refresh_pattern' is related , though read the comment in
squid.conf.default completely before tweaking this parameter.

  You can also use :

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

  as a 'tool' to verify cacheability stats for objects returned
  by webservers.

  M.

Reply via email to