determining ip information for eth0...failed

robalo · 07-19-2004, 04:21 PM

were you guys able to fix this problem?

OldDogT · 07-20-2004, 12:11 AM

Yes, robalo, I just got this problem fixed on my new Linux box. I'm running Fedora Core 2 with a 3Com 3c905 100BaseTX [Boomerang]. I kept getting errors on startup regarding eth0. I found in a number of places, including here, that kudzu might be the issue.

Anyway, I disabled it and did a restart (shutdown -r now) and am fine.

The way I disabled it on Fedora Core 2 was
1) run the "setup" command
2) arrow down to System services and select it
3) arrow down to kudzu and deselect it
4) select OK
5) restart

That did it for me.

-OldDogT

robalo · 07-20-2004, 12:22 PM

OldDog,

I tried that and it didn't fix the problem.

/bin/bash · 07-20-2004, 11:16 PM

Quote:

were you guys able to fix this problem?

Well both flamah and ScooterB have not posted back since I suggested they try another module back in Post #17, so that may have been the problem. It's always nice when people post back if they get a problem solved, so others can benefit from their experience.

Quote:

I tried that and it didn't fix the problem.

What exactly is the problem?
And post any information you can including config files requested earlier in this thread.

ScooterB · 07-25-2004, 11:57 AM

I did fix my problem. It seems that for some reason RH9 didn't recognize the card. Changed to one that I knew would work (had used it before successfully) and everything worked fine. I think many of these manufacturers push equipment out the door in an effort to have the "biggest, fastest, most powerful widget" and don't do the proper amount of pre-release testing to insure that it works with all platforms (even though they say it will). Sometimes the tried and true is the best; at least to start with.

chadwen · 04-13-2006, 09:22 AM

Ok it seems I'm getting a simular problem, but I'm really not doing a new install. I shut my server down for about 4 months and just recently decided to restart it.
Before everything was working fine. Now I get Determining IP information for eth0.... failed when starting redhat 9.

My card is 3com etherlink XL(3c900b-tpo)

ifdown gives nothing
ifup gives info on eth0 and lo

I have disabled kudzu(read that it can cause problems), but still eth0 won't activate. The RH system is connection to a D-link wireless router and the D-link is connection to my ISP's router. I have linux using dhcp to my D-link and the dhcp server active on the D-link.(its been that way for years.)

my windows machine works fine. also I noticed when I log into my D-link router under DHCP section it shows 2 dhcp clients. one is my windows machine other is my linux box(can tell by mac address)That makes me think that the connection is good(network cable is good)
Linux just won't alow the network card to activate. I'm thinking its a driver issue, but am unsure what to do with the driver on
3coms site. when I download it and copied it to a disk. linux said it wasn't a valid .tar file. So I had to open it on my windows machine, but when I did that all that was in it was a file named [content] and thats it. in the readme and install files on the site they mention a file called 3c90x.c and compiling it. no such file is there.
I'm stuck at that point. I have look and noticed that I have been using the 3c59x.o file for my card. not sure what changed?

suggestions?

fatvalet · 01-31-2008, 10:18 PM

Hi, I have just installed Fedora 8 and am getting exactly the same error message as many people who posted here: "Determining IP information for ethx... failed;". Things I've tried

(1) I disabled Kuzdu and the firewall: no effect;
(2) Inserted another PCI NIC: no effect, same error message;

Fedora recognizes both NICs (one is on-board).

Code:

[root@localhost ~]# cat /etc/modprobe.conf | grep eth
alias eth0 e100
alias eth1 tulip

Both NICs clearly attempt to communicate with the router because both quickly return "Check cable?" if I indeed disconnect the cable. When the cable is connected, Fedora sits for about a minute the returns the dreaded message. This is not just a start up problem: service network restart returns the same error.

If anybody out there has the solution for this, please respond. So far I have run out of options and my next step is, as always in these cases, to downgrade the OS

okos · 01-31-2008, 11:39 PM

Quote:

Originally Posted by JuanKerr

Mine will never start on boot, something that I'm working on at the moment (hence came across this thread), however if you manually initialize it ('# ifconfig eth0 up') then activate it with 'neat' it should work. fealnx.o is the correct module for it but lord alone knows why redhat won't start it properly on its own.

Now I don't have red hat. However, I think I can help.
If I understand you correctly, you have internet connection but it does not start at boot. You have to manually connect each time after boot.
Two ways I solved that problem for I had the same problem:
1. Find your rc.local file. That file initializes programs at boot. Mine is located at /etc/rc.d/rc.local but yours might just be /etc/rc.local.

In there add ifconfig eth0 up and dhcpcd eth0. Now I dont have 'neat' and I am not sure what that is. You should be able to figure it out from there.

2. My initialization script for connecting the internet is /etc/rc.d/rc.inet1.conf and all I had to do was set dhcpcd to yes for eth1.
See link
http://www.linuxquestions.org/questi...t-boot-584059/

fatvalet · 02-11-2008, 10:30 PM

Quote:

Originally Posted by fatvalet

Hi, I have just installed Fedora 8 and am getting exactly the same error message as many people who posted here: "Determining IP information for ethx... failed;". Things I've tried

(1) I disabled Kuzdu and the firewall: no effect;
(2) Inserted another PCI NIC: no effect, same error message;

Fedora recognizes both NICs (one is on-board).

Code:

[root@localhost ~]# cat /etc/modprobe.conf | grep eth
alias eth0 e100
alias eth1 tulip

Both NICs clearly attempt to communicate with the router because both quickly return "Check cable?" if I indeed disconnect the cable. When the cable is connected, Fedora sits for about a minute the returns the dreaded message. This is not just a start up problem: service network restart returns the same error.

If anybody out there has the solution for this, please respond. So far I have run out of options and my next step is, as always in these cases, to downgrade the OS

I figured it out... well, I found something on another forum, which resembled my problem: start linux with the following additional option

Code:

acpi=off