On Thu, Feb 26, 2015 at 1:57 PM, Walter Dnes <waltd...@waltdnes.org> wrote: > > Let's forget about additional routes with different metrics, and try > to get the simplest case working, and move forward from there. If I have > /etc/conf.d/net as... > > modules=( "!iproute2" ) > config_eth0=" > 192.168.123.251/29 broadcast 192.168.123.255 > 169.254.1.1/16 broadcast 169.254.255.255" > routes_eth0=" > default via 192.168.123.254" > > ...and run "/etc/init.d/net.eth0 restart", I can access the device on > 169.254.0.0/16. If I comment out the "modules=( "!iproute2" )" line and > run "/etc/init.d/net.eth0 restart", I can't access the device. The > above is the minimal "net" file require for ifconfig. I want to be able > to work this down the road when ifconfig goes away, and iproute2 becomes > mandatory.
No idea. I set up a test VM. I don't have a 169.254 device to try to access but I have the same "ip a" and "ip r" output for the net-tools and iproute2 setups (except for the "eth0:1" label in the net-tools setup). gentest conf.d # cat net modules="!iproute2" config_eth0=" 10.0.2.15/24 169.254.1.1/16" routes_eth0="default via 10.0.2.2" dns_servers_eth0="8.8.8.8" gentest conf.d # ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 brd 127.255.255.255 scope host lo valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:20:20:20 brd ff:ff:ff:ff:ff:ff inet 10.0.2.15/24 brd 10.0.2.255 scope global eth0 valid_lft forever preferred_lft forever inet 169.254.1.1/16 brd 169.254.255.255 scope global eth0:1 valid_lft forever preferred_lft forever gentest conf.d # ip r default via 10.0.2.2 dev eth0 metric 2 10.0.2.0/24 dev eth0 proto kernel scope link src 10.0.2.15 127.0.0.0/8 dev lo scope host 169.254.0.0/16 dev eth0 proto kernel scope link src 169.254.1.1 gentest conf.d # vi net gentest conf.d # cat net config_eth0=" 10.0.2.15/24 169.254.1.1/16" routes_eth0="default via 10.0.2.2" dns_servers_eth0="8.8.8.8" gentest conf.d # rc-service net.eth0 restart gentest conf.d # ip a 1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN group default link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 brd 127.255.255.255 scope host lo valid_lft forever preferred_lft forever 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast state UP group default qlen 1000 link/ether 52:54:00:20:20:20 brd ff:ff:ff:ff:ff:ff inet 10.0.2.15/24 brd 10.0.2.255 scope global eth0 valid_lft forever preferred_lft forever inet 169.254.1.1/16 brd 169.254.255.255 scope global eth0 valid_lft forever preferred_lft forever gentest conf.d # ip r default via 10.0.2.2 dev eth0 metric 2 10.0.2.0/24 dev eth0 proto kernel scope link src 10.0.2.15 127.0.0.0/8 dev lo scope host 169.254.0.0/16 dev eth0 proto kernel scope link src 169.254.1.1 > I'm still trying to wrap my brain around how the ADSL router decides > whether to look for 169.254.0.0/16 on the ethernet cable to the TV tuner > rather than on the phone line going to my ISP, and then the world. 169.254 isn't supposed to be routable so the router might be set up not to route it out to the world. You'd be better off setting up a 192.168 address on the tuner, if possible.