Russell wrote:
Others have it, too. But they are not using the same files etc. I just checked a RedHat machine, it shows the lo route. But if you can see lo in ifconfig, and it works, you should be ok. At least my debian machine works fine without the lo entry shown in 'route -nv'. You might find some info on this defference between distros in the debian docs.Stephen Gran wrote:This one time, at band camp, Russell said:Current books on linux networking don't mention things on implicit loopbacks. It's really annoying. I'd like to get an answer: (A)the books are now wrong, or (B)ifupdown isn't adding a loopback (127.0.0.0) route when it should.If you see an entry for lo come up from ifconfig, it's being added at boot time. To make sure it's working, try telnet localhost 25 without an external network connection. If it succeeds, it's routing successfully over lo. I think the problem is in route's reading of the kernel table, rather than in the actual kernel table.I got: Trying ::1... Trying 127.0.0.1... Connected to localhost. Escape character is '^]'. ... I guess it's working:) I would have done a simple test like this myself if i had more experience, but i've only had debian installed for a few days. I last installed and used mandrake a couple of years ago, which was set up using 'older' methods. Is ifupdown specific to debian, or do other distros have it too?
-- Johan Ehnberg johan@ehnberg.net "Windows? No... I don't think so."