Head branch with "server = domain", login fails.

[Paul Allen]

OK, I'm taking another run at this thing.  My experiments with the
TNG branch are on hold.  I won't be allowed to put up a non-Microsoft
domain controller anyway, so I'm stepping back to just putting up a
good fileserver.

I'm testing the head branch code under Solaris 7 on SPARC (calamity).  
The code I'm using was checked out yesterday.  I've got three other 
Solaris 2.6 machines running 2.0.4b with zero problems.  (And that
means *zero* problems for months and months.  Great job, guys!)

The local NT PDC is running NT 4, service pack 5, in a "resource
domain" called ISS-TECH.  My other servers all use the local PDC for
authentication, and that's what I'm trying to do with calamity.

The basic networking stuff works, but the PDC is rejecting my password.
With debug level 5, it looks as if it thinks my NT account doesn't
exist.  The Samba server is logging a "no such user" error after
having a conversation with the PDC.

Here's my smb.conf:

[global]
        workgroup = ISS-TECH
        server string = Samba Server
        security = DOMAIN
        encrypt passwords = Yes
        password server = iss-tech-f
        username map = /usr/local/samba-main/lib/map
        log file = /usr/local/samba-main/var/log.%m
        max log size = 50
        name resolve order = host lmhosts bcast wins
        client code page = 437
        lm announce = True
        local master = No
        dns proxy = No
        wins server = 130.121.5.43
        remote announce = 130.42.151
[export]
        path = /local/export
        read only = No
        guest ok = Yes

In running through the checks in DIAGNOSIS.txt, tests 7, 9, 10, and 11
have problems.  Here's what I see:

# ./smbclient //calamity/export -Uallenp 
load_unicode_map: filename
/usr/local/samba-main/lib/codepages/unicode_map.437 does not exist.
added interface ip=130.42.35.62 bcast=130.42.35.255 nmask=255.255.255.0
Password: 
session setup failed: ERRSRV - ERRbadpw (Bad password - name/password
pair in a Tree Connect or Session Setup are invalid.)
# 

"allenp" is my NT login, and I gave my NT password in the above test.
All permutations of my Unix and NT login names and passwords produce
the above error.  Also, I  don't know where the "load_unicode_map"
complaint is coming  from.  The file it wants certainly does not
exist.

In test 9, a "net use x: \\calamity\export" command on an NT 4 machine
produced a "network name cannot be found" error.  That host can ping
calamity, and vice versa.  A "net view \\calamity" command on the same
PC succeeded in showing the "export" service.

In  test 10, the "nmblookup -M ISS_TECH" command produced the message:

	querying ISS-TECH on 130.42.35.255
	name_query failed to find name ISS-TECH

I guess I can understand that.  We have one big switched network, but
the PC's and Unix machines are on different IP subnets.  

And finally, in test 11, neither calamity nor any of my production Samba
servers appear in the browse list on the NT machine.  This appears to be
a normal condition on my network.

Can anybody point me in a direction to go looking?  Have I left out
a detail that would unlock the whole thing?

Thanks!

Paul Allen
-- 
Paul L. Allen           | voice: (425) 865-3297  fax: (425) 865-2964
Unix Technical Support  | paul.l.allen at boeing.com
Boeing Phantom Works Math & Computing Technology Site Operations,
POB 3707 M/S 7L-68, Seattle, WA 98124-2207