[PATCH] Fix wbinfo --sids-to-unix_ids with some unmapped sids

[Volker Lendecke]

On Mon, Jan 07, 2019 at 12:36:44PM +0100, Stefan Metzmacher wrote:
> Hi Volker,
> 
> >>> We have to reply all we have if some SIDs fail to map.
> >>>
> >>> Review appreciated!
> >>
> >> I'd like to have this, but somethings are not clear to me.
> >>
> >> Is this an wbinfo only thing?
> > 
> > That is not wbinfo only.
> > 
> >> Or is smbd also affected?
> > 
> > smbd is also affected.
> 
> Ok, thanks!
> 
> >> Do we need backports?
> > 
> > This only popped up at a customer site. Nobody public has complained
> > yet. So it depends if you think that many people are affected.
> 
> Ok, we may not need to backport this.
> 
> >> Please don't use S-1-5-1000 for the test, it's really
> >> a bug (a different one) that lookup_wellknown_sid() doesn't handle
> >> it. I'll see if I can come up with a fix for that... (we already handle
> >> it in dom_sid_lookup_predefined_name() and we need to unify the code)
> > 
> > Which one do you know for 100% sure will never be mapped? At my
> > customer site this is the one that caused trouble. I thought that in
> > principle idmap tdb maps everything and I was happy that I found one
> > that is not mappable.
> 
> I updated the test a bit to use S-1-5-123456789 and made it a bit more
> strict, so that it demonstrates the bug with a knownfail entry, which
> gets revert together with the fix.

I did not think of using just another arbitrary value like 123456789.

> What about the attached patchset?

Looks good, pushed! Thanks for taking care!

Volker

-- 
SerNet GmbH, Bahnhofsallee 1b, 37081 Göttingen
phone: 0551-370000-0, mailto:kontakt at sernet.de
Gesch.F.: Dr. Johannes Loxen und Reinhild Jung
AG Göttingen: HR-B 2816 - http://www.sernet.de