Ozzie, Jan is correct here... The OS and switch will be confused with this.
A simple solution without bonding is just: * Configure eth0 as dhcp client. In your example, 172.16.17.15 is the dynamic ip address. * create a virtual ip on eth0. In your example, it should be 172.16.17.95 as the static ip address. Both ip is up at the same time, and no confusion for the OS and switch. Plus, the routes always go to eth0. And you don't need to manually remove/add the rj45 cables. If you want to use both nics for failover, then use bonding, then configure the logical nic as dhcp client, and create static ip as well. regards, Andre | http://www.varon.ca On Mon, Jan 26, 2009 at 4:47 AM, Ozzie de Leon <[email protected]> wrote: > On Sat, Jan 24, 2009 at 2:31 AM, jan gestre <[email protected]> wrote: >> >> Both eth0 and eth1 belong to the same network that is why the latter is >> happening, if what you want is to route traffic to other subnets (Is this >> what you're trying to do?), I suggest you configure the either eth0 or eth1 >> e.g., 172.16.18.95. >> HTH. >> > > sorry... didn't quite get that. both NICs do need to stay in the same > subnet, but eth0 is dynamic (172.16.18.15 was given by the DHCP server) > while eth1 is static. in a way, it's kinda like a fall back so when the unit > gets deployed and client's DHCP server fails, i can always connect a cable > to eth1 and connect to it (it's one of those things without a keyboard so > making changes or troubleshooting remotely is the best way to go). > > it's really cosmetic in the sense that it does work since i can connect to > it. i just don't want to find any "gotchas" once it gets deployed since eth0 > seems to be making both 172.16.18.15 and 172.16.18.95 active/accessible. _________________________________________________ Philippine Linux Users' Group (PLUG) Mailing List http://lists.linux.org.ph/mailman/listinfo/plug Searchable Archives: http://archives.free.net.ph
