RODC and the samba_kcc code

[Andrew Bartlett]

On Tue, 2014-07-29 at 12:23 +0200, Stefan (metze) Metzmacher wrote:
> Am 29.07.2014 12:22, schrieb Andrew Bartlett:
> > On Tue, 2014-07-29 at 12:19 +0200, Stefan (metze) Metzmacher wrote:
> >> Am 29.07.2014 12:16, schrieb Andrew Bartlett:
> > 
> >>> I've been working with Garming to get wintest up and going again, and a
> >>> successful wintest run would be part of my criteria, so we know we
> >>> didn't break something subtle.  We already found other subtle and
> >>> unexpected breakages, such as the python samba_kcc breaking RODC
> >>> support, for example.
> >>
> >> Is it related to https://bugzilla.samba.org/show_bug.cgi?id=10697 ?
> > 
> > It seems it is up to the KCC to set the SPECIAL_SECRETS_PROCESSING mode
> > in the repsFrom, and the samba_kcc has no such code, while the old
> > implementation does check an is_rodc flag.  
> > 
> > That said, the other bugs probably don't help either.
> 
> So what should we do with samba_kcc for 4.2? I'd revert to the old kcc.

Yes, we were going to write it up and propose exactly that.

Andrew Bartlett

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