At 20:51 01/05/2002 +0200, Henrik Nordstrom wrote:
>On Wednesday 01 May 2002 19:21, Richard Barrett wrote:
> > I am embarrassed to say that I've discovered the cause of my
> > problem and it looks to be a mainly non-Squid problem.
>
>Thanks for letting us know.
>
> > But in one respect I'm not entirely sure that Squid's hands are
> > clean in this matter. In some instances, the requests pushed done
> > the same connection by Squid seem to be incompatible. In
> > particular, some requests containing no user authentication and
> > others that do seem to be made down a common connection by Squid. I
> > do not believe Squid should do this. If I understand correctly,
> > some of the information associated with authentication is lodged in
> > the Apache connection record rather than the request record and I
> > presume that connection sharing by Squid might thus lead to
> > problems similar to those I have experienced. I've not tried to
> > prove this. I'm just hypothesising.
>
>The HTTP specification is quite clear on this.. Authentication and
>user information is a per-request property, not a property of the TCP
>connection.
>
>A server erroneously assigning per-user properties to the connection
>will fail in precense of proxies, similar to how your X-Forwarded-For
>log module failed.
I will not dispute you contention that Squid is operating within the letter
of RFC2616.
Looks like Apache is not RFC compliant in this respect. I'll let the folks
there know.
Received on Thu May 02 2002 - 03:45:08 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 17:07:52 MST