eth0: Unknown host

Hello folks!

My linux box (Debian GNU/Linux 3.0) runs samba as primary domain
controller and some other services (jabber, DNS, TeamSpeak etc.).

Recently I had some problems with my NIC which was configured as eth0
device (Realtek 10Mpbs PCI lan card). The problem was that my linux
box just stopped communicating with other hosts on network. At first I
thought it's some samba-related problem but then I realised no traffic
is going to nor from my linux box! Not even ICMP echo (ping) packets.
Only thing it was able to ping was itself (127.0.0.1 via intefrace lo
and 192.168.1.2 via interface eth0). All other machines are on same
subnet (192.168.1.0) but noone can ping this linux machine nor it can
ping them!

I also noticed that I have strange system messages every now and then
saying "eth0: Unknown host" (e.g. on system boot sequence or when I'm
running some network tools like ifup/ifdown/etherconf). Haven't looked
at system log files for details.

After spending some time trying to figure out what happened and why
isn't it working any more I decided to replace LAN card and I did so.
I replaced it with similar Realtek based card and everything seems to
work fine. Ping works fine now, samba is working as PDC as usual and
other installed other network services are operational) but I still
(!?!) get those messages "eth0: Unknown host" What does that mean?

My new ethernet card is configured as eth0, I had a look at linuxconf
and my host is still configured with IP address 192.168.1.2 on
interface eth0. ipconfig reports it as eth0.

Sorry for long post, can somebody tell me what does that mean "eth0:
Unknown host"? Is my old card really broken or is it something with my
debian?

Note: I don't use firewall rules so it's not anyting related to
firewall.

Thanks,

Tomislav

Nesquik schrieb am Sonntag, 29. Februar 2004 00:49 in
comp.os.linux.networking:

> [...]
> but then I realised no traffic is going to nor from my linux
> box! Not even ICMP echo (ping) packets. Only thing it was able
> to ping was itself (127.0.0.1 via intefrace lo and 192.168.1.2
> via interface eth0). All other machines are on same subnet
> (192.168.1.0) but noone can ping this linux machine nor it can
> ping them!
Did you check the LED's on the NIC?
Did you try another port on your switch?

In our networt we have an rater old switch whitch has some defect
ports. Their LED's schow a good physical link but I get no
traffic over this ports...

Doing this two checks will at least asure that your switch ist
ok...

>
> [...]Haven't looked at system log files for
> details.
But this may contain importand information we may need to help
you!

> [...] but I still (!?!) get those messages
> "eth0: Unknown host" What does that mean?
Usually this message means, that the hostname "eth0" could not be
resolved by your DNS. But that may not a problem of your DNS
because "eth0" most likly not a computername.
I guess there is any software (maybe a script) on your Linux-box
that wrongly tries to access "eth0" as computer name.

Try a 'grep eth0 /etc/* -R' and look into the files reported to
check, if the occurence of "eth0" is in the right context.

> [...]
> Is my old card really broken
At this point I'd say: "no".

> [...]

bye
TT

--
PGP: 0x273c213E

On Sun, 29 Feb 2004 11:49:17 +0100, Timothy Truckle
<will.not.read.this@web.de> wrote:
>Did you check the LED's on the NIC?
>Did you try another port on your switch?

Yes I did, I double checked everything. But when I replaced ethernet
card in linux box everything was working perfectly so the problem is
not in cable/switch.

>> "eth0: Unknown host" What does that mean?
>Usually this message means, that the hostname "eth0" could not be
>resolved by your DNS. But that may not a problem of your DNS
>because "eth0" most likly not a computername.
>I guess there is any software (maybe a script) on your Linux-box
>that wrongly tries to access "eth0" as computer name.
>
>Try a 'grep eth0 /etc/* -R' and look into the files reported to
>check, if the occurence of "eth0" is in the right context.

I ran grep on my etc folder and string "eth0" showed up in many files
there. However, file /etc/network/interfaces was most suspicious
becouse it is used by ifup/ifdown so I checked it out. This is how it
looked like:
--------------------------------------------------------------------------
auto lo eth0

iface lo inet loopback

iface eth0 inet static
address 192.168.1.2
netmask 255.255.255.0
gateway
--------------------------------------------------------------------------

I also found an older version named interfaces.pre-etherconf so I
opened it to see what's the difference, here is that old version:
---------------------------------------------------------------------------
# /etc/network/interfaces -- configuration file for ifup(8), ifdown(8)

# The loopback interface
auto lo
iface lo inet loopback

# The first network card - this entry was created during the Debian
installation
# (network, broadcast and gateway are optional)
auto eth0
iface eth0 inet static
address 192.168.0.1
netmask 255.255.255.0
network 192.168.0.0
broadcast 192.168.0.255
----------------------------------------------------------------------------

Analysing the last one I realised every network interface should begin
with a line like "auto <interfacename>" and then some lines defining
interface itself. So I modified the infertaces file and now it looks
like this:
-----------------------------------------------------------------------------
auto lo
iface lo inet loopback

auto eth0
iface eth0 inet static
address 192.168.1.2
netmask 255.255.255.0
network 192.168.1.0
broaadcast 192.168.1.255
-----------------------------------------------------------------------------

Everything works fine now, no more "eth0: Unknown host" messages on my
linux box! I even tried to put in my old ethernet card (just in case)
and it seems to be malfunctioning. I couldn't ping anything and then
several ICMP echos returned from remote machines but it was terrible
response time (> 1000 ms) and packet loss was more than 90%

I will take that old ethernet card to test it on one Windows machine,
but I don't think I'll ever put it back in my linux box.

Thank you very much for helping me and I hope you all don't mind I've
posted all this back on the group. I did it for all other people who
might be reading this and have similar problems.

Greetz from Croatia

Tomislav Knapic

Nesquik schrieb am Sonntag, 29. Februar 2004 14:47 in
comp.os.linux.networking:

> Thank you very much for helping me
That's what this group is good for... ;o)

> and I hope you all don't
> mind I've posted all this back on the group. I did it for all
> other people who might be reading this and have similar
> problems.
>
This was a good idea!
I whish all users would post what finally solved their problem.

bye
TT

--
PGP: 0x273c213E