NTLM problem on Sparc (was Re: 2.5.PRE9 released)

Hi Henrik,

Il 17.07 21/07/2002 Henrik Nordström ha scritto:
>On Sun, 21 Jul 2002, Guido Serassio wrote:
>
> > Hi Henrik,
> >
> > Only a question, what squid debug level it's better ?
> > ALL,9 or limited to some sections, 29,9 and may be others ?
>
>The following should do just fine:
>
>debug_options ALL,1 29,9 84,9
>
> 29 = auth
> 84 = helpers
> (actually one of them should be sufficient, but having both is nice)
>
>or just use "squid -k debug" to temporarily turn on full debugging while
>making the request... what I usually do.
>
>Regards
>Henrik

Here are the logs. There are Alpha successful logs and Sparc failure log,
with full debugging and only 29,9 and 84,9 logging.
All was made with the same Windows user, with IE cache clean, with same
Squid release compiled with same configure options:

serassio@io:~/2.5/src$ ./squid -v
Squid Cache: Version 2.5.PRE9-CVS
configure options: '--disable-wccp' '--enable-underscores'
'--mandir=/usr/share/man' '--enable-basic-auth-helpers=winbind'
'--enable-ntlm-auth-helpers=winbind SMB fakeauth'
'--enable-default-err-language=Italian' '--enable-err-languages=Italian'
'--enable-auth=ntlm,basic' '--enable-external-acl-helpers=winbind_group'
'CC=gcc' 'CFLAGS=-s'

Both machine runs Samba 2.2.5 member of the same Windows domain.

I have already looked on logs, but i can't found anything interesting.

Only a thing: Sparc logs remember to me the same log that I had on first
2.5 native NTLM work on Windows, when I had problems with
stateful_helper_reserve_t bit fields structure.

Regards

Guido

-
=======================================================
Serassio Guido
Via Albenga, 11/4 10134 - Torino - ITALY
E-mail: guido.serassio@serassio.it
WWW: http://www.serassio.it