[brussels-dev] [networking-discuss] Brussels Phase II aka ipadm design review [20-31 March 2009]

[James Carlson]

Peter Memishian writes:
> 
>  > > If there are no addresses, then there is no point in "plumbing" the
>  > > associated interface under the hood.  The administrator's intent is
>  > > implicit in the addresses that are configured over the interface.
>  > > 
>  > 
>  > what if I want to plumb the interface just to snoop the traffic
>  > on it? (I've done this in the past). 
> 
> FWIW, there used to be a bug in snoop where it would do a SIOCGLIFCONF if
> it was invoked without an explicit datalink name.  I changed this several
> builds back to do a dlpi_walk() instead.  Perhaps that's what you're
> recalling.

Could be.  The SIOCGLIFCONF thing is so abhorrent that I _always_
invoke snoop with "-d", even on systems that have only one interface.
If it weren't for folks who'd be confused, and the lack of general
interest in doing anything more with snoop, I'd argue to have -d
become mandatory.

-- 
James Carlson, Solaris Networking              <james.d.carlson at sun.com>
Sun Microsystems / 35 Network Drive        71.232W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.496N   Fax +1 781 442 1677