Re: [SQU] fakeauth_auth

Hi Craig,
    It won't be going into squid 2.3. (unless someone else wants to
backport it.)
    If I get the time, I will quite likely backport it to 2.4 as patch
set. There are a _lot_ of changes made to squid as part of it, so it
will be a while off before I get the time. (Read several months at
least). The problem with this is that the authentication code hasn't
been stress tested. As people use squid 2.5 it will get it's real
testing and work out any bugs in it. That is unlikely to happen for
squid 2.4 though :[

The most supported fashion today is the squid 2.5 snapshots. They are
unstable code though. The squid developers are pretty careful to keep
the snapshots functional, and they always have the bugfixes from earlier
versions, they just sometimes have new bugs :]

I believe kinkie is using it from the squid 2.5 branch and putting it
through production readiness testing now, (using NTLMSSP not fakeauth),
so I wouldn't worry about stability today. In a month or two as the new
2.5 features get added, I would be careful about checking for know bugs
before downloading new 2.5 snapshots.

Did you try the client test I asked for?

Rob

----- Original Message -----
From: "Craig Fels" <csfels@swbell.net>
To: "Robert Collins" <robert.collins@itdomain.com.au>;
<squid-users@ircache.net>
Sent: Wednesday, February 21, 2001 9:21 AM
Subject: Re: [SQU] fakeauth_auth

> Robert/All,
>
> Have you decided to port the NTLM code to 2.4 or 2.3stable? I really
like
> the ability to block/grant access based on userid, and to have
domain/userid
> appear in the log files.
>
> please reply with the most support method to do this. True
authentication
> is not needed.
>
> Thanks,
> Craig
>

--
To unsubscribe, see http://www.squid-cache.org/mailing-lists.html