I have a Speedtouch 510iv4 which has entries in its IP route table for 169.254.0.0 which has been fed through to the Kernal IP routing table table as follows

route -n
Kernel IP routing table
Destination Gateway Genmask Flags Metric Ref Use Iface
169.254.0.0 0.0.0.0 255.255.0.0 U 0 0 0 eth0
10.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 eth0
127.0.0.0 0.0.0.0 255.0.0.0 U 0 0 0 lo
0.0.0.0 10.0.0.138 0.0.0.0 UG 0 0 0 eth0


What is 169.254.0.0? Do I need it?

I've removed it from the Kernal IP routing table and from the IP route table of the Speedtouch, start/stop eth0 but it just reappears!

It looks like its the network your medem is connected to. It will be their while your modem is plugged in and connected.

It's not a network that I recognise. Where would the Kernal IP routing table get this address from?

Your modem will have an IP address and when it is connected a route to the network is usually added.

If you bring the eth0 interface down then the route should disappear.

You can also check the modems IP address. If it begins 169.254 then then the route is your modem.

I have the same entry, but on eth1 which is connected to my local LAN, not the adsl modem, so where does it come from.

The 169.254 address range is used for some kind of intermediary step to the internet. The Speedtouch modem may always tell the computer that its ip address is in the 169.254 range because that range isn't used for the internet or for LANs, so it shouldn't conflict with anything.

That's just a guess. I don't know anything about those modems or much about the 169.254 addresses.

the 169.254 addresses popup usually when you are using DHCP but have not been able to obtain an address. They are associated with DHCP and are not routable.

Found this thread on the net:

http://cert.uni-stuttgart.de/archive.../msg00309.html

The range of 169.254.x.x has been reserved for this purpose. The idea is
that when DHCP fails, it still gives clients a way to communicate with
TCP/IP. It is, off course, not intended to production use, but it will give
the client a chance to use TCP/IP in order to recover. For example, on a
widely switched Ethernet backbone, it could enable the user to at least
e-mail MIS saying that their network connection is flaky.

But I'm not using DHCP, all my IP's are static.
One more thing, I cant delete it from the routing table.
NET not found is replied when i try to "route del 169.254.0.0", maybe I'm doing something wrong, but....

To delete the route need to supply more information -

route del -net 169.254.0.0 netmask 254.254.0.0 dev eth0

I did this to remove the entry from the table and also removed the corresponding entries from the modem. Recycled eth0 and 169.254 reappeared - spooky. Must also be added somewhere from within Linux.

Sorry should be of course -

route del -net 169.254.0.0 netmask 255.255.0.0 dev eth0

Tnx aikempshall, it worked.
The big question is where does this route come from?

Well I'm not sure about the DHCP causing the route entry and subsequent address, but I have encountered it many times when I would configure an adapter for DHCP, but it failed to get an address so instead it would use the 169.254.0.0

Glad you were able to remove the entry anyhow.

for delete that route only add NOZEROCONF=yes to /etc/sysconfig/network file and restart the network (/sbin/service network restart)


Salut !

APIPA is what the 169.254 address scheme is. Automatic Private IP Addressing is what it stands for. Micro$oft uses it to address a NIC when DHCP fails. It's kind of becoming a standard though not and RFC, at least not one I'm aware of.

There is no problem with it being in your routing table, it will not affect your IP routing.

Hope it helps.

MrKnisely