348be5f1e44c0 (Share data_blob implementation) breaks Samba3

Jeremy Allison jra at samba.org
Sun Oct 12 22:14:15 GMT 2008


On Mon, Oct 13, 2008 at 08:59:36AM +1100, Andrew Bartlett wrote:
> 
> Or we can find out which bits rely on the different behaviours.  Why
> does Samba3's smbd rely on a length-limited string being also NULL
> terminated?

At a quick look it looks like the memcache code, which is
quite recent. But there are also uses in rpc_server/srv_samr_nt.c
which are RPC related.

The problem is the patch isn't complete as applied. It broke
previously working code. 

I'm sure you wouldn't be pleased at changes that break
the underlying S4 semantics without testing. This is the
same. I think it needs reverting until such testing is
done and the patch can be made complete.

Jeremy.


More information about the samba-technical mailing list