Fwd: [REG : 112040380892433]: Is there any requirement when handling an NT_TRANSACT_SET_SECURITY_DESCRIPTOR to store the DACL exactly as presented on the wire?

[Richard Sharpe]

Hi folks,

This seems to be the answer. Now I am off to look at MS-FSA.

---------- Forwarded message ----------
From: Tarun Chopra <Tarun.Chopra at microsoft.com>
Date: Mon, Apr 16, 2012 at 7:41 AM
Subject: RE: [REG : 112040380892433]: Is there any requirement when
handling an NT_TRANSACT_SET_SECURITY_DESCRIPTOR to store the DACL
exactly as presented on the wire?
To: Richard Sharpe <realrichardsharpe at gmail.com>
Cc: MSSolve Case Email <casemail at microsoft.com>


Hi Richard

Per our analysis, Yes,  there is a requirement mentioned in section
3.1.5.16 of MS-FSA to store DACL (passed in InputBuffer parameter)
presented by higher-layer protocol as-is without any modification.
Excerpt is as follows:

       The object store MUST set Open.File.SecurityDescriptor to InputBuffer.

As a result, NTFS volume will store the extraneous zeros passed in
DACL and returns the same DACL in Query Security Information
operation.

Kindly let me know if this answers your query or if you require
further assistance/clarification on below issue.

Thanks
Tarun Chopra.

-----Original Message-----
From: Richard Sharpe [mailto:realrichardsharpe at gmail.com]
Sent: Saturday, April 14, 2012 5:33 PM
To: Tarun Chopra
Subject: Re: [REG : 112040380892433]: Is there any requirement when
handling an NT_TRANSACT_SET_SECURITY_DESCRIPTOR to store the DACL
exactly as presented on the wire?

On Sat, Apr 14, 2012 at 5:25 PM, Tarun Chopra
<Tarun.Chopra at microsoft.com> wrote:
> Hi Richard
>
> I am able to repro the scenario against win2008r2 and the information
> shared by you is sufficient. Win2008r2 also exhibits same behavior as
> you observed against win2003.
>
> I have done some research and will share the same by tomorrow which
> will be helpful to understand the behavior further.
>
> Per analysis, so far, this is file system behavior, ntfs in this case,
> and windows default file system applies and returns the blob passed in
> set/query respectively.
>
> I will share more information and the excerpt from ms-fscc which is to
> some extent relevant.
>
> Hope that's fine, please share your inputs.

Hi Tarun,

Thanks for your reply.

I look forward to receiving your answer. It will allow us to decide
how to handle the situation. Ie, whether or not to emulate the
behavior of Windows in Samba or not.

--
Regards,
Richard Sharpe
(何以解憂？唯有杜康。--曹操)




-- 
Regards,
Richard Sharpe
(何以解憂？唯有杜康。--曹操)