Making, drinking tea and reading an opus magnum from Srebrenko Sehic:
> On Sun, Mar 30, 2003 at 06:29:09PM +0200, Henning Brauer wrote:
>
> > yes, that is known. I don't see a real world problem with this; the effect
> > is zero as skip steps solve that nicely.
>
> Hurray for skip-steps then ;) However, imagine having tons of
> aliases; resulting filter rules might get a bit ugly.
>
> Without checking, can't this be fixed rather easily? Check on the
> netmask; if same as non-aliased mask, skip it.
w/ "lots" of aliases set on the interface
and being of the same cider block one can
just utilize the link1 flag for lo(4) and
use static route on the gateway.
i dunno how antispoof will handle it though (;
cu
--
paranoic mickey (my employers have changed but, the name has remained)
