On Mon, Apr 10, 2000 at 03:26:31AM -0700, H. Peter Anvin wrote:
> Followup to: <20000410170638.A7359@saw.sw.com.sg>
> By author: Andrey Savochkin <saw@saw.sw.com.sg>
> In newsgroup: linux.dev.kernel
> > >
> > > This implies that a gateway address, being a regular IP address used
> > > by a router, may not have the value 0 for "Host-number" (the part
> > > after the network mask). Nothing to argue at all.
> >
> > RFC isn't Bible. I don't know any good reasoning for such restrictions.
> >
>
> Actually, it *IS* the Bible for these kinds of stuff, and the good
> reasoning is interoperability.
:-)
Have you ever tried to count how many times network communication standards
are broken in Linux (or any other OS)?
Certainly, casual breaks are prohibited, and if we spot any we fix it as a
bug. However, well-considered breaks (including considering interoperability
consequences) and, more important, fixes for RFC bugs have always been ok.
I connect two computers by a TP. I do not have anything like Internet in a
mile around. What's wrong with assigning arbitrary IP addresses for this
link? Nothing.
Certainly, people playing such games in a mixed environment (I'm not
even speaking about Internet) should understand very well what they do and
consider all the consequences. The kernel should also provide a kind of
fool-tolerance for administrators. For this case it's a special route flag:
without it such routes are refused to be created.
Best regards
Andrey V.
Savochkin
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Apr 15 2000 - 21:00:13 EST