[PATCH] nfs4acl: Fix owner mapping with ID_TYPE_BOTH

[Christof Schmitt]

On Wed, Sep 14, 2016 at 11:43:54AM -0700, Christof Schmitt wrote:
> On Wed, Sep 14, 2016 at 09:00:55AM -0700, Jeremy Allison wrote:
> > On Tue, Sep 13, 2016 at 04:26:03PM -0700, Jeremy Allison wrote:
> > > On Mon, Sep 12, 2016 at 05:34:21PM -0700, Christof Schmitt wrote:
> > > > From f1883adf6ed027a03be2a3d4f1631d8bdb283e38 Mon Sep 17 00:00:00 2001
> > > > From: Christof Schmitt <cs at samba.org>
> > > > Date: Mon, 12 Sep 2016 16:22:16 -0700
> > > > Subject: [PATCH] nfs4acl: Fix owner mapping with ID_TYPE_BOTH
> > > > 
> > > > This fixes a corner case when using NFS4 ACLs with ID_TYPE_BOTH.  Before
> > > > this patch, the owner entry in the ACL would be mapped to a gid entry in
> > > > the NFSv4 ACL, and not the expected special owner entry. This is caused
> > > > by the id mapping returning a valid gid and the nfs4 mapping assumed
> > > > that this was actually a group.
> > > > 
> > > > Fix this by asking for the uid first, and explicitly checking if the
> > > > mapped uid matches the owner. That creates a uid entry in the NFSv4 ACL
> > > > that can be changed later in smbacl4_substitute_{simple,special} to the
> > > > expected special owner entry.
> > > 
> > > OK, went through the logic here very carefully (w.r.t chown
> > > interactions), and I think this is the correct thing to do.
> > > 
> > > Pushed.
> > 
> > The push failed (unrelated issue :-) and in the shower
> > this morning I had a thought (something about this patch
> > must have been bothering me :-).
> > 
> > This patch adds logic to check the uid first - if the
> > mapped uid matches the owner, which is correct, but
> > it also removes the fallback mapping of:
> > 
> >  -           } else if (sid_to_uid(&ace_nt->trustee, &uid)) {
> >  -                   ace_v4->who.uid = uid;
> > 
> > I don't think that's right. I think the logic should be:
> > 
> > check the uid first - if the mapped uid matches the owner
> > check if mapped to group
> > check if mapped to user.
> > 
> > (i.e. keep the final fallback code). Given that, I think
> > the following patch is correct. Christof, can you confirm ?
> 
> Yes, correct. This is required for id mappings that return ID_TYPE_UID
> and an ACL entry that does not match the owner. I missed that, since i
> was thinking along the implications of ID_TYPE_BOTH.

Tested again, and pushed to autobuild.

Christof