Date: Tue, 17 Oct 1995 15:23:45 -0600 From: Nate Williams <nate@rocky.sri.MT.net> To: "Garrett A. Wollman" <wollman@lcs.mit.edu> Cc: Nate Williams <nate@rocky.sri.MT.net>, freebsd-bugs@freefall.freebsd.org Subject: Re: kern/785: 2.0.5-950622-SNAP Various ifconfig alias problems Message-ID: <199510172123.PAA08091@rocky.sri.MT.net> In-Reply-To: <9510172113.AA04070@halloran-eldar.lcs.mit.edu> References: <199510171950.MAA04955@freefall.freebsd.org> <199510172048.OAA07972@rocky.sri.MT.net> <9510172113.AA04070@halloran-eldar.lcs.mit.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
> > Is there anyway that ifconfig could be hacked to always use a netmask of > > 0xffffffff when alias was specified to avoid all of these kind of error > > reports? Is there *any* time when an alias is used when the netmask is > > not all 1's? > > Yes. Believe it or not, the code was not originally written to > support `fake hosts'. It was originally written to support `multiple > logical IP subnets on a single wire'. I can understand why this would be a 'good thing', but isn't this same functionality available via the alias keyword? Can you give an example of when you would use the alias keyword without an all 1's netmask? It appears from all of your responses that the netmask must always be all ones to have multiple IP addresses on a single network device. What is the limitation of using 'multiple logical IP subnets' on a single interface vs. using multiple 'fake hosts' on a single interface? Nate
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199510172123.PAA08091>