public inbox for linux-fsdevel@vger.kernel.org
 help / color / mirror / Atom feed
From: Jeremy Allison <jra-eUNUBHrolfbYtjvyW6yDsg@public.gmane.org>
To: Steve French <smfrench-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Cc: Anton Altaparmakov
	<aia21-KWPb1pKIrIJaa/9Udqfwiw@public.gmane.org>,
	"linux-cifs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org"
	<linux-cifs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
	samba-technical
	<samba-technical-w/Ol4Ecudpl8XjKLYN78aQ@public.gmane.org>,
	linux-fsdevel
	<linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org>,
	Shirish Pargaonkar
	<shirishpargaonkar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Subject: Re: How mode bits are stored in NFS/NTFS/CIFS/SMB3 ACLs
Date: Thu, 25 Sep 2014 17:09:43 -0700	[thread overview]
Message-ID: <20140926000943.GE17111@samba2> (raw)
In-Reply-To: <CAH2r5mu0xCTXh401C0ujpyJuTQw5AuG7-+DOi8umVTukB6egGQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>

On Thu, Sep 25, 2014 at 09:02:50AM -0700, Steve French wrote:
> On Thu, Sep 25, 2014 at 3:29 AM, Anton Altaparmakov <aia21-KWPb1pKIrIJaa/9Udqfwiw@public.gmane.org> wrote:
> > Hi Steve,
> >
> > On 25 Sep 2014, at 07:04, Steve French <smfrench-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org> wrote:
> >> Did some experiments today to see how mode bits are stored by the
> >> Windows NFS server in the RichACL (CIFS or NFS ACL).   mounted nfsv4.1
> >> to Windows from Linux then created a bunch of files and did chmod of
> >> various combinations of 07777 bits (including sticky, setuid etc.)
> >>
> >> Windows NFS server is storing the user owner bits with SID
> >> S-1-5-88-1 and using SID S-15-88-2 for group owner and S-1-5-88-4 for
> >> the ACE for "other" (this is easy to spot over CIFS/SMB3 etc because
> >> user owner and group owner map to these SIDs in the security
> >> descriptor returned over the wire).
> >>
> >> As expected, for each of the 3 ACEs, it is setting "GENERIC_READ" in
> >> the ACE for '4' (read) and GENERIC_WRITE for '2' (write) and
> >> GENERIC_EXECUTE for '1' (execute).  What is puzzling is where it
> >> stores the setuid and sticky bits (bits 07000) because they are not
> >> visible in the CIFS/NTFS ACL.
> >
> > As far as I know the Windows NFS server user "Services For Unix (SFU)" and those special bits are stored on NTFS in an Extended Attribute (EA) (note this is the $EA attribute not a named stream/named $DATA attribute on NTFS).  I wrote about this 9 years ago on linux-ntfs-dev mailing list.  Archive post is here (read my point "2" in that post for the details):
> >
> >         http://marc.info/?l=linux-ntfs-dev&m=112965244715312
> >
> > This means that those bits only take effect / have any significance for applications using the Windows POSIX subsystem (e.g. NFS server and Cygwin), i.e. normal Win32 based apps will not be affected by them at all.
> >
> I did a getfattr to list all the windows (os/2) exstended attributes
> (over cifs) and didn't see it, perhaps it is hidden - but I can query
> for SETFILEBITS directly

Try using smbclient's "geteas <filename>" command.

      parent reply	other threads:[~2014-09-26  0:09 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-09-25  6:04 How mode bits are stored in NFS/NTFS/CIFS/SMB3 ACLs Steve French
     [not found] ` <CAH2r5mt70hDes3tHMuij3YkDY9N2aG+bL77mhkg7k2W1y6W-jQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-09-25 10:29   ` Anton Altaparmakov
     [not found]     ` <5154DE0E-C6D9-4051-AE5C-3AC38FB1B7BA-KWPb1pKIrIJaa/9Udqfwiw@public.gmane.org>
2014-09-25 16:02       ` Steve French
     [not found]         ` <CAH2r5mu0xCTXh401C0ujpyJuTQw5AuG7-+DOi8umVTukB6egGQ-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2014-09-26  0:09           ` Jeremy Allison [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20140926000943.GE17111@samba2 \
    --to=jra-eunubhrolfbytjvyw6ydsg@public.gmane.org \
    --cc=aia21-KWPb1pKIrIJaa/9Udqfwiw@public.gmane.org \
    --cc=linux-cifs-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=linux-fsdevel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
    --cc=samba-technical-w/Ol4Ecudpl8XjKLYN78aQ@public.gmane.org \
    --cc=shirishpargaonkar-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
    --cc=smfrench-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox