Re: [squid-users] Avoiding incorrect URL interception

On 21/01/11 06:03, Roberto wrote:
> Hi,
>
> I've just installed squid in a debian 5 machine, and most of what I want
> is working, but I´m having one problem.
>
> When someone searches for an URL that does not exist, most modern
> browser get the DNS error and try searching google (or bing, or
> whatever). But, with squid installed, the browser is instead getting
> squid's message that the URL could not be retrieved. As a result, the
> browsers do not try further to find the server.
>
> An example: Without squid, if I type "squid" in Firefox's url bar, it
> will not find a server, but will go to goolge, and find there squid's
> page (ww.squid-cache.org). With squid, the browser is getting squid's
> error page ("The requested URL could not be retrieved"), and stops.
>
> Is there any way to avoid this behavior? To make squid not intercept
> those "incorrect" URL's?

No. Squid is not a web browser and does not contain this web browser
feature. DNS resolvers also do not contain this browser feature.

I think there is a wiki.squid-cache.org FAQ entry about this.

On the side it is also not a very safe feature to have turned on.
There are a numer of viruses based on FunWebSearch which use this
feature to make money out of infected computers on pay-per-view scams.
By simply setting the default search engine to be a website of their own
they can hijack the results page and the cash rolls in from unwitting
advertisers.

Amos

-- 
Please be using
   Current Stable Squid 2.7.STABLE9 or 3.1.10
   Beta testers wanted for 3.2.0.4