jfz
-
Posts
5 -
Joined
-
Last visited
Content Type
Profiles
Forums
Events
Posts posted by jfz
-
-
I did some googling around and turned up this bug report:
http://www.ussg.iu.edu/hypermail/linux/ker...402.2/1118.html
and this one:
http://www.ussg.iu.edu/hypermail/linux/ker...403.0/0234.html
It looks like some recent patching of the 2.6.3 kernel reintroduced a bug into the e1000 driver that had been removed at one point. Kernel 2.6.4, here I come.
-
I tried the "acpi=ht" option, which seemed to have no effect at all. I looked for the source .config file with no success, until I realized this last install I didn't bother to select kernel source. I'll try that when I get home tonight, but it's starting to look like that PCI NIC is the best hope. And maybe submitting a bug to Mandrake.
-
I'm using the 2.6.3-4mdk kernel, or trying to. Here's my lilo.conf:
# File generated by DrakX/drakboot
# WARNING: do not forget to run lilo after modifying this file
boot=/dev/hda
map=/boot/map
default="linux-enterprise"
keytable=/boot/us.klt
prompt
nowarn
timeout=100
message=/boot/message
menu-scheme=wb:bw:wb:bw
image=/boot/vmlinuz
label="linux"
root=/dev/hda1
initrd=/boot/initrd.img
append="devfs=mount splash=silent"
vga=788
read-only
image=/boot/vmlinuz
label="linux-nonfb"
root=/dev/hda1
initrd=/boot/initrd.img
append="devfs=mount"
read-only
image=/boot/vmlinuz-enterprise
label="linux-enterprise"
root=/dev/hda1
initrd=/boot/initrd-enterprise.img
append="devfs=mount splash=silent"
read-only
image=/boot/vmlinuz-smp
label="linux-smp"
root=/dev/hda1
initrd=/boot/initrd-smp.img
append="devfs=mount splash=silent"
read-only
image=/boot/vmlinuz
label="failsafe"
root=/dev/hda1
initrd=/boot/initrd.img
append="failsafe devfs=nomount"
read-only
other=/dev/fd0
label="floppy"
unsafe
-
I am unable to get Mandrake 10.0 to boot successfully with the enterprise kernel, unless I disable the onboard gigabit NIC on my Intel motherboard. Enabled, the kernel hangs during boot bringing up eth0. Same story with the SMP kernel. The "plain vanilla" linux kernel DOES successfully configure and boot with the NIC enabled, so it can apparently be done (which is how I'm posting this now). Unfortunately, this plain kernel has no idea what to make of my P4's hyperthreading (only one virtual CPU listed under /proc/cpuinfo), and sees less than half of my 2 GB of RAM.
I'm just hoping somebody has also seen this behavior and can provide some ideas; if need be, I may have access to a PCI NIC I can try and swap in, but I'm hoping that won't be necessary. System details:
3.0 GHz P4
Intel D875PBZLK mobo
2 GB DDR400 memory
MSI GF4 MX440 video
SB Live 5.1 audio
Liteon burner/DVD ROM drives
NB: This same system had been running fine for months under MDK 9.1, only requiring a manual build and install of the NIC driver from a tarball on Intel's website. I already checked, there's nothing there for the 2.6 kernel, unfortunately.
eth0 Problems
in Hardware
Posted
As you've figured out, there's a problem with CE and the Intel gigabit NIC. I ran into this exact problem myself, as detailed here:
http://www.mandrakeusers.org/index.php?sho...t=0entry99278
Basically, the 2.6.3 kernel that CE installs has a bug in its e1000 driver. You can try booting into the non-enterprise, non-smp kernel, and see if your onboard network resumes functioning. I ended up installing another NIC and disabling the onboard for the time being, don't know if that's an option for you. I've since been playing with a couple of other distros, while waiting for 10.0 Official .iso's to drop.