James Morris <jmorrisat_private> writes: > On Tue, 1 Oct 2002, Olaf Dietsche wrote: > >> Well, we'll never know until we try :-). Besides that, sys_bind() and >> inet_bind() are on an entirely different level. > > Sorry, but I'm not in favour of this hook. > > Firstly, as far as I can tell, what you're trying to do in accessfs is > provide fine grained control over access to ports with otherwise normal > Unix user/group/other file permissions, and the purpose of the hook is to > determine the range of ports which are protected by this scheme. This is > unnecessarily overloading the existing kernel logic relating to reserved > ports as part of a quite different access control model. > > Secondly, what accessfs (and this hook) is trying to do is essentially > authoritative+permissive, a model not explicitly supported by LSM at this > point. Well, I must admit, I didn't bother about design philosophies. I just thought it would be foolish to not use this existing framework and reinvent the wheel on my own. What I do care about, however, is the goal of having a more secure system than before and I thought that's what LSM is all about. Seems like I have to invent my own make of security_ops. Anyway, thanks for listening to my weird ideas. ;-) > Please don't get me wrong: I think the general idea of accessfs is pretty > cool, but it seems to be out of scope for LSM as a restrictive framework. I'm glad you like it nonetheless. Regards, Olaf. _______________________________________________ linux-security-module mailing list linux-security-moduleat_private http://mail.wirex.com/mailman/listinfo/linux-security-module
This archive was generated by hypermail 2b30 : Thu Oct 03 2002 - 05:03:15 PDT