linux-kernel - Re: -mm merge plans for 2.6.20

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Message-ID: <200612110436_MC3-1-D49E-FE58@compuserve.com>
Date:	Mon, 11 Dec 2006 04:32:59 -0500
From:	Chuck Ebbert <76306.1226@...puserve.com>
To:	Steve French <smfltc@...ibm.com>
Cc:	linux-cifs-client@...ts.samba.org, Jeremy Allison <jra@...ba.org>,
	simo <simo@...ba.org>,
	Shirish S Pargaonkar <shirishp@...ibm.com>,
	linux-kernel <linux-kernel@...r.kernel.org>, akpm@...l.org
Subject: Re: -mm merge plans for 2.6.20

In-Reply-To: <457CDC38.2090907@...ibm.com>

On Sun, 10 Dec 2006 22:19:04 -0600, Steve French wrote:

> I don't remember any problems reported with plain text password
> support on current cifs and I have certainly seen it negotiated with no 
> problem,
> but I will double check with your reported flag combination.

I played around with it some more.

With SecurityFlags = 0x7 (default) the server asks for a plaintext
password and the client refuses.  That's fine.

With 0x37 the client agrees to send a plaintext password (or at least
fails to reject the server's request for one,) but actually sends:

<         ANSI Password Length: 24
<         Unicode Password Length: 24

<         ANSI Password: 3577D3557009178AFF455A0F7A99C6585CAEF99C515F2F2C
<         Unicode Password: 3577D3557009178AFF455A0F7A99C6585CAEF99C515F2F2C

(my password was aaaaaaaa).  This fails with error -13 (invalid password.)

With 0x30 the client sends:

>         Password Length: 24

>         Password: 61616161616161610000000000000000681C1DCF00002200

and everything works.

> > Also, the client doesn't automatically pick up the domain name from
> > smb.conf like smbfs does.
> >
> >   
> That is true, and is intentional.   cifs sends a domain of null (ie use 
> the server's
> default domain) - but it can be overridden on mount

That's OK then.  I just happened to notice it when I was comparing
traces of smbfs mounts to the ones from cifs.  Maybe the manpage should
mention this difference for those who are converting, though.

-- 
MBTI: IXTP
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/