[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

(usagi-users 00665) Re: SO_BINDTODEVICE functionality removed?



On Sun, Aug 05, 2001 at 12:19:56AM +0900, YOSHIFUJI Hideaki / ?$B5HF#1QL@?(B wrote:
> In article <Pine.LNX.4.33.0107281127300.1467-100000@xxxxxxxxxxxxxxxxx> (at Sat, 28 Jul 2001 11:35:23 -0400 (EDT)), Nathan Lutchansky <lutchann@xxxxxxxxxx> says:
> 
> > Why was this change made?  I suspect it was done to simplify the new bind
> > semantics, but it is a critical piece of functionality for a critical
> > piece of software.  Would it be possible to add this back, maybe as a
> > compile-time option, so that I can run a USAGI kernel again?  Right now
> > I'm using vanilla 2.4.7.
> 
> Because bind(2) for specific device after bind(2) for any device causes
> the "binding-closer attack" problem.  You can steal packets by
> bind(2)ing each interface attached on your box.  BAD.

But this sockopt is restricted to users with raw socket capability!  That
seems safe to me.  -Nathan

-- 
+-------------------+---------------------+------------------------+
| Nathan Lutchansky | lutchann@xxxxxxxxxx |  Lithium Technologies  |
+------------------------------------------------------------------+
|  I dread success.  To have succeeded is to have finished one's   |
|  business on earth...  I like a state of continual becoming,     |
|  with a goal in front and not behind. - George Bernard Shaw      |
+------------------------------------------------------------------+

Attachment: pgpT0kPIPfM7p.pgp
Description: PGP signature