|
eth0 won't recognize link upon bootup
Hi, having a very strange problem with Fedora.
This system was originally running RH7.3 and I did not have these problems, I decided to wipe and install Fedora and now this very weird problem occurs: I have two NICs installed, eth0 and eth1. Upon bootup, eth0 will not recognize link. The nic itself indicates link on its external LED. NO amount of restarting network will get this interface up. I have to physically unplug the network cable and then plug back in for it to work. I have tried all the standard testing for cabling or hardware problems. I have swapped cables, I have swapped nics, I changed the order of the nics physicaly on the Mainboard. Always, it is eth0 that will not recognize link, no matter which network card is used. I can even take the network cable from the eth0 and use in eth1 and move eth1's network cable to eth0 and eth0 still refuses to work, and eth1 is just fine. So I'm thinking this is a bug of some sort in Fedora. Has anybody had this same or similar problem and have a solution? |
I am having the exact same problem in RHEL 3, and I have never seen this before.
|
http://www.linuxquestions.org/questi...367#post751367 That is my solution for the same problem with Mandy10 (2.6 kernel). Try putting your eth module in the /etc/modules file if you run 2.4 or what I did if you run 2.6
|
I've looked for the modprobe config file, must be something different in redhat.
[root@duly etc]# find / -name *modprobe* /proc/sys/kernel/modprobe /etc/log.d/conf/services/modprobe.conf /etc/log.d/scripts/services/modprobe /usr/share/man/man8/modprobe.8.gz /sbin/modprobe [root@duly etc]# |
There is a file called /etc/modules.conf
Mine has nothing in it except for: alias eth0 e100 |
Could use post the file /etc/modules.conf and /etc/sysconfig/network-scripts/ifcfg-eth0.
Alex |
This is my /etc/sysconfig/network-scripts/ifcfg-eth0
DEVICE=eth0 ONBOOT=yes BOOTPROTO=none IPADDR=192.168.1.8 NETMASK=255.255.255.0 GATEWAY=192.168.1.2 TYPE=Ethernet USERCTL=no PEERDNS=no NETWORK=192.168.1.0 BROADCAST=192.168.1.255 As mentioned, there is nothing in my modules.conf file except for: alias eth0 e100 |
Hi,
Just do me a favour, post /etc/sysconfig/network and the log messages in /var/log/messages from the module e100. Alex |
This is my /etc/sysconfig/network:
NETWORKING=yes HOSTNAME=duly Here are my messages related to eth0 during a reboot: Feb 10 23:35:44 duly kernel: e100: eth0 NIC Link is Down Feb 10 23:36:00 duly kernel: e100: eth0 NIC Link is Up 100 Mbps Full duplex Feb 10 23:39:48 duly kernel: e100: eth0: Intel(R) PRO/100 Network Connection Feb 10 23:39:48 duly kernel: e100: eth0: Intel(R) PRO/100 Network Connection Feb 10 23:39:48 duly kernel: e100: eth0 NIC Link is Up 100 Mbps Full duplex Feb 10 23:40:26 duly kernel: e100: eth0 NIC Link is Down Feb 10 23:40:34 duly kernel: e100: eth0 NIC Link is Up 100 Mbps Full duplex |
maybe I will just disable the second onboard nic in the bios, and rebuild the kernel. I could later let kudzu pick it up.
|
Hi,
It looks like the module has a problem with the hardware. These are onboard nics. I got the same problem with broadcom gigabit onboard nics. Set the speed in your module.conf, maybe this helps. option e100 e100_speed_duplex=4,4 DESCRIPTION: e100_speed_duplex Valid Range: 0-4 (1=10half;2=10full;3=100half;4=100full) Default Value: 0 The default value of 0 sets the adapter to auto-negotiate. Other values set the adapter to forced speed and duplex. Example usage: insmod e100.o e100_speed_duplex=4,4 (for two adapters) Alex |
So I would make my modules.conf file look like this:
alias eth0 e100 option e100 e100_speed_duplex=4,4 Just two lines and you think that will do it? |
I added that to my modules.conf and rebooted. It didn't work.
|
Quote:
|
I am experiencing the exact same issues with the same hardware and drivers. I have tried numerous different switches and settings.
Has anyone solved this problem yet? |
| All times are GMT -5. The time now is 02:03 PM. |