I have an Abit NFS-8 m/b. It has a built-in Nvidia nForce MCP network adaptor. I also put in a D-Link NIC. During the Redhat 9 installation, it did not pick up the built-in Nvidia network adaptor. However, the D-Link one was configured and worked afterwards. Does any one know why Redhat 9 or, in general Linux, does not configure both NICs? And how does the installation determine which NIC to install?

I later went to Nvidia web site and found out that they offer an all-in-one Linux RPM for the nForce m/b that presumably includes all Linux drivers for the m/b. How do I find out in that RPM file it has the nForce MCP network adaptor driver for Linux and how do I install only the NIC driver?

- Dhsieh

RedHat (or any OS for that matter) does not have drivers for every NIC ever made, paritcularly if the NIC is either new or uncommon. Just install the RPM per nVidia's intructions. (Probably something like "rpm -ih filename.rpm". It probably contains a collection of drivers that will be installed in /lib/modules/<kernel version>/. It may also modify /etc/modules.conf with a line similar to "alias ethN nic-module" where N is a 0 or 1 and nic-module is the name of the driver.

Hi gamu829,

After installing the Nvidia binary RPM file, my Redhat 9 picks up the built-in NIC from Nvidia as "eth1". I assigned a new IP to the device. Unfortunately, it did not work. Is there any way to further debug this device?

I gorot to include the error msg from initializing the NIC from Nvidia:

Command failed: /sbin/modprobe nvnet

Output:
modprobe: Can't locate module nvnet