Bizzare routing table entry.
Ian Smith
smithi at nimnet.asn.au
Wed Aug 8 06:01:44 PDT 2007
On Tue, 7 Aug 2007 15:37:50 -0700 Chuck Swiger <cswiger at mac.com> wrote:
> On Aug 7, 2007, at 3:13 PM, Adam J Richardson wrote:
> > Modulok wrote:
> >> 0&0xc0a80132 link#1 UCS 0 0 bge0
> > <snip>
> >> 1. The first entry, it's not IPv4, IPv6 or a MAC address that I've
> >> ever seen, what format is it?
> >
> > Hi Modulok,
> >
> > It's possible to represent IPv4 addresses as a single number. I
> > don't recall the algorithm for converting that four byte dot-
> > delimited group into an integer, though, so I can't tell you what
> > number it is. Perhaps you can Google the algorithm and do the math
> > to figure out what it is.
>
> aaa.bbb.ccc.ddd => 0xAABBCCDD, where AA = hex(aaa), BB = hex(bbb), etc.
> In particular, 0xc0a80132 is the hex equivalent of 192.168.1.50.
>
> An IP address + netmask can normally be represented in the routing
> table via the slash notation-- say 192.168.1.50/24 meaning a
> 255.255.255.0 (or 0xffffff00) netmask. Non-contiguous netmasks are
> represented by "address & netmask", but since no normal network ever
> uses such a netmask, they almost always represent a
> misconfiguration-- someone confused the arguments such that the route
> command interpreted the gateway IP as a netmask instead.
Been there; in my case it was a rogue route added by an ifconfig with an
incorrect - as you say, non-contiguous - netmask. In this case it might
have been specified/interpreted as 0.0.0.0 netmask 192.168.1.50 ?
Cheers, Ian
More information about the freebsd-questions
mailing list