Fix incorrect msDS-IntID handling

[Andrew Bartlett]

On Wed, 2016-01-06 at 00:06 +0100, Stefan Metzmacher wrote:
> Hi Andrew,
> 
> > > > A final version of these patches is in that repo. 
> > > 
> > > I'm ignoring the python related patches, which are currently on
> > > top...
> > > 
> > > > Please review/push!
> > > 
> > > in err_replmetadata_incorrect_attid()
> > > 
> > > +                if o.local_usn > hash_att[att].local_usn:
> > > +                    # This is always what we would have sent
> > > over
> > > DRS,
> > > +                    # because the DRS server will have sent the
> > > +                    # msDS-IntID, but with the values from both
> > > +                    # attribute entries.
> > > +                    hash_att[att].version = o.version
> > > +                    hash_att[att].originating_change_time =
> > > o.originating_change_time
> > > +                    hash_att[att].originating_invocation_id =
> > > o.originating_invocation_id
> > > +                    hash_att[att].originating_usn =
> > > o.originating_usn
> > > +                    hash_att[att].local_usn = o.local_usn
> > > 
> > > Should  we have a "continue" here
> > > otherwise it makes no sense if we always do "hash_att[att] = o"
> > > 
> > > +            hash_att[att] = o
> > > +            set_att.add(att.lower())
> > 
> > Well spotted.  Thanks
> 
> In the branch you added the "continue" only within the
> "if o.local_usn > hash_att[att].local_usn:" statement.
> I guess we should remove 4 whitespaces and do it always
> if hash_att[att] already exists.

OK.  I'll take another look at that.

> > > In general are the source4/selftest/provisions/ changes required
> > > to
> > > be
> > > in the "dbcheck: Fix incorrect/duplicate attrid in
> > > replPropertMetaData"
> > > commit?
> > 
> > It is the test for that change. 
> > 
> > > Or can they be part of the "selftest: Update release-4-1-0rc3
> > > with
> > > more
> > > test records"
> > > commit?
> > 
> > I don't want anything extra in that commit, it is too large as it
> > is
> > and essentially impossible to review.  Like we do ABI updates in
> > their
> > own commit, this belongs in it's own commit.
> 
> Ok.
> 
> metze

Thanks.  I also have had a failure that indicates our python reference
counting issue may be biting us again.  I'm looking into that.

Andrew Bartlett

-- 
Andrew Bartlett
https://samba.org/~abartlet/
Authentication Developer, Samba Team         https://samba.org
Samba Development and Support, Catalyst IT   
https://catalyst.net.nz/services/samba





-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: This is a digitally signed message part
URL: <http://lists.samba.org/pipermail/samba-technical/attachments/20160106/b5a1e386/signature.sig>