From: ldg@ulysium.net
Date: Fri Mar 28 2003 - 20:19:04 EST
I tried a few more things to find out why the networking isn't working on
this tru64 5.1b, so I have a few more detailed info to post here, which
perhaps may help me get a few more pointers.
First, I brought the configuration down to a minimum and committed this to
the config files so this would be the config after a reboot:
the /etc/hosts file only has the 127.0.0.1 localhost line and one more with
10.0.1.2 pointing to the hostname of that machine, then /etc/routes only has
default 10.0.1.1 and /etc/networks only has loop 127 loopback and nothing
else. The rc.config has the hostname defined there, NETDEV="tu0",
IFCONFIG_0="10.0.1.2 netmask 255.255.255.0", the other NETDEV and IFCONFIG
are left blank and NUM_NETCONFIG="1".
I'm not running any routed, gated, rwhod or anything else of that nature.
I made a change to test the assumption that perhaps the auto-negotiation on
the interface wasn't working right, by forcing it to use fast ethernet at
the srm console, which has the same result as before with 10baset.
I'm testing this with a mac on the other end of the switch, it reports a
fast ethernet half duplex, which is what I forced with the srm console, that
made no change. (the switch is also fast ethernet capable)
My last test was to start tcpdump on tu0 such as:
tcpdump -i tu0
and watch what happens. If I bring on a local config on the mac that puts it
on 10.0.1.1, I can see tcpdump saying something like this:
arp who-has 10.0.1.1 tell 0.0.0.0
and
arp who-has 10.0.1.1 tell 10.0.1.1
(I'm omitting the numbers at the start of those lines)
So, the interfaces are actually talking to each other, but a ping doesn't
cause anything like that. If I issue a:
ifconfig tu0 10.0.1.1
on the alpha, then the mac tells me that 10.0.1.1 is in use, which is
correct.
The alpha can ping itself on 10.0.1.2 but can't ping 10.0.1.1 (the mac) and
the mac can't ping 10.0.1.2 (the alpha).
I hope this gives a bit better picture of what's going on.
What else can I try that could point to where the real problem is?
-- Didier Godefroy mailto:dg@ulysium.net
This archive was generated by hypermail 2.1.7 : Sat Apr 12 2008 - 10:49:14 EDT