Disable SMB2 for 3.6?

[Volker Lendecke]

On Fri, Jul 08, 2011 at 10:45:02PM +1000, ronnie sahlberg wrote:
> +1 for what Volker says.
> 
> If it has been in there for months and no one has noticed. It means
> virtually no real world exposure.
> Who knows what else will show up after a few more kilo/mega-hours of real use.
> 
> No one benefits from a "uncertain quality release".
> Better drop SMB2 from the release until such stage it is both well
> tested, mature and stable.
> 
> 
> Volkers data suggests it is neither well tested nor mature so
> therefore I think it should be dropped.
> 
> Drop smb2 for now, and add it back later.

Made up my mind, see my latest message. Sorry for the
confusion.

The main argument that makes a non-table based approach
appropriate is the vastly reduced number of SMB2 requests
that we have to handle compared to SMB1. With metze's patch
the code is completely equivalent to what the SMB1 server
does, so the same security assumptions that have served us
very well for almost two decades hold true again. It would
be nice to have that code together in one file, but this is
not strictly necessary from a pure functional security
perspective.

Volker

-- 
SerNet GmbH, Bahnhofsallee 1b, 37081 Göttingen
phone: +49-551-370000-0, fax: +49-551-370000-9
AG Göttingen, HRB 2816, GF: Dr. Johannes Loxen