| View previous topic :: View next topic |
| Author |
Message |
todd93 Apprentice

Joined: 30 Oct 2006 Posts: 248
|
Posted: Sat Feb 06, 2010 5:58 pm Post subject: |
|
|
| DONAHUE wrote: | @ bobspencer
he has a grub which is loading a kernel
therefore grub is not a player in this problem.
he has now revealed an ati chipset in his lspci -kvm
he earlier showed no drivers for the hard drive controllers on that chipset in his posted config |
I have recompiled the kernel with the drivers you suggested, still no luck.
Thanks
Todd |
|
| Back to top |
|
 |
DONAHUE Veteran


Joined: 09 Dec 2006 Posts: 1565 Location: Goose Creek SC
|
Posted: Sat Feb 06, 2010 8:07 pm Post subject: |
|
|
You are looking like?
| Quote: |
# <partition> <mountpoint> <type> <opts> <dump/pass>
/dev/sdb1 /boot ext2 defaults,noatime 1 2
/dev/sdb2 none swap sw 0 0
/dev/sdb3 / ext3 noatime 0 1
#/dev/sr1 /mnt/cdrom auto noauto,user 0 0 #comment out to allow automount
/dev/sda1 /mnt/windows ntfs-3g defaults,rw,user,umask=0 0 0 |
| Quote: | default 0
timeout 30
splashimage=(hd1,0)/boot/grub/splash.xpm.gz
title Gentoo Linux 2.6.31-r6
root (hd1,0)
kernel /boot/kernel-2.6.31-gentoo-r6 root=/dev/sdb3
# vim:ft=conf:
title Windows 7 Ultimate
rootnoverify (hd0,0)
makeactive
chainloader +1 |
and you have hard drive controller drivers in kernel?
in chroot run:
| Code: | grub
root (hd1,0)
setup (hd0)
quit
exit
reboot |
|
|
| Back to top |
|
 |
Jaglover Veteran


Joined: 28 May 2005 Posts: 1571 Location: Saint Amant, Acadiana
|
Posted: Sat Feb 06, 2010 8:15 pm Post subject: |
|
|
In fact, if you feel confused with multi-HDD multi-boot setup here's the foolproof recipe:
Disconnect all hard drives but the one for Gentoo;
install Gentoo and GRUB;
add your secondary drives, make sure your Gentoo drive is the drive BIOS boots;
configure GRUB to boot other operating systems. |
|
| Back to top |
|
 |
todd93 Apprentice

Joined: 30 Oct 2006 Posts: 248
|
Posted: Sun Feb 07, 2010 8:36 am Post subject: |
|
|
| DONAHUE wrote: | You are looking like?
| Quote: |
# <partition> <mountpoint> <type> <opts> <dump/pass>
/dev/sdb1 /boot ext2 defaults,noatime 1 2
/dev/sdb2 none swap sw 0 0
/dev/sdb3 / ext3 noatime 0 1
#/dev/sr1 /mnt/cdrom auto noauto,user 0 0 #comment out to allow automount
/dev/sda1 /mnt/windows ntfs-3g defaults,rw,user,umask=0 0 0 |
| Quote: | default 0
timeout 30
splashimage=(hd1,0)/boot/grub/splash.xpm.gz
title Gentoo Linux 2.6.31-r6
root (hd1,0)
kernel /boot/kernel-2.6.31-gentoo-r6 root=/dev/sdb3
# vim:ft=conf:
title Windows 7 Ultimate
rootnoverify (hd0,0)
makeactive
chainloader +1 |
and you have hard drive controller drivers in kernel? |
Yes, that is correct.
| DONAHUE wrote: | in chroot run:
| Code: | grub
root (hd1,0)
setup (hd0)
quit
exit
reboot |
|
Here's where I am hitting the wall. after completing the steps that you have provided, I reboot, and the kernel panics, and tells me that the only available drives are: sda1, 2, and 3. I get a string of some numbers in brackets, followed by f's in brackets with particular numbers in them. The system then does not respond to keystrokes or anything, and the only way to get it out is the power button.
I am getting really confused here about what to do, it's looking like I'm going to have to follow jaglovers advice and disonnect all drives but the one for linux, and go from there.
I really appreciate all the hlp
Thanks
Todd |
|
| Back to top |
|
 |
whiteghost Apprentice


Joined: 26 Jul 2009 Posts: 285 Location: nj xt 145 whiteghost inda hood
|
Posted: Sun Feb 07, 2010 9:04 am Post subject: |
|
|
| Quote: | | I get a string of some numbers in brackets, followed by f's in brackets with particular numbers in them. The system then does not respond to keystrokes or anything, and the only way to get it out is the power button. |
from linux sea: are you seeing these "The digits "0,0" or "8,3"
| Quote: | VFS: Cannot open root device "sda3" or unknown-block(8,3)
Please append a correct "root=" boot option; here are the available partitions:
sda driver: sd
sda1 sda2
The digits "0,0" or "8,3" can be different in your case - it refers to the device that the kernel tries to access (and which fails). Generally speaking one can say that, if the first digit is 0, then the kernel is unable to identify the hardware. If it is another digit (like 8 ), it is unable to identify the file system (but is able to access the hardware).
The problem here is that the kernel that you are booting cannot translate the "root=/dev/..." parameter you gave it (inside the boot loader configuration) into a real, accessible file system. Several reasons can result in such a failure: |
_________________ www.informationclearinghouse.info |
|
| Back to top |
|
 |
todd93 Apprentice

Joined: 30 Oct 2006 Posts: 248
|
Posted: Sun Feb 07, 2010 9:47 am Post subject: |
|
|
| whiteghost wrote: | | Quote: | | I get a string of some numbers in brackets, followed by f's in brackets with particular numbers in them. The system then does not respond to keystrokes or anything, and the only way to get it out is the power button. |
from linux sea: are you seeing these "The digits "0,0" or "8,3"
| Quote: | VFS: Cannot open root device "sda3" or unknown-block(8,3)
Please append a correct "root=" boot option; here are the available partitions:
sda driver: sd
sda1 sda2
The digits "0,0" or "8,3" can be different in your case - it refers to the device that the kernel tries to access (and which fails). Generally speaking one can say that, if the first digit is 0, then the kernel is unable to identify the hardware. If it is another digit (like 8 ), it is unable to identify the file system (but is able to access the hardware).
The problem here is that the kernel that you are booting cannot translate the "root=/dev/..." parameter you gave it (inside the boot loader configuration) into a real, accessible file system. Several reasons can result in such a failure: |
|
YES!! that is it, it is says unknown-block(2,0).
It then says precisely what you have posted. Your post seems to have cut off the possible reasons for my failure.
Thanks
Todd |
|
| Back to top |
|
 |
whiteghost Apprentice


Joined: 26 Jul 2009 Posts: 285 Location: nj xt 145 whiteghost inda hood
|
Posted: Sun Feb 07, 2010 9:55 am Post subject: |
|
|
| Quote: | Fri Feb 05, 2010 3:10 pm Post subject: Report this post Reply with quote Edit/Delete this post
you could disconnect your windows drive, edit files to use sda, just to get your gentoo running and find out what is working / not working.
linux sea chapter 7 for kernel error codes.
http://swift.siphos.be/linux_sea/ch07.html#id3002568 |
read down to bottom of linked page. or from top of page for kernel howto.
or the book for much more. _________________ www.informationclearinghouse.info |
|
| Back to top |
|
 |
whiteghost Apprentice


Joined: 26 Jul 2009 Posts: 285 Location: nj xt 145 whiteghost inda hood
|
Posted: Sun Feb 07, 2010 10:15 am Post subject: |
|
|
| Quote: | | unknown-block(2,0) |
since first digit is not 0 and you have ext2 and ext3 compiled into kernel i think your fstab or grub.conf is wrong. _________________ www.informationclearinghouse.info |
|
| Back to top |
|
 |
Mike Hunt Advocate


Joined: 19 Jul 2009 Posts: 4671
|
Posted: Sun Feb 07, 2010 10:25 am Post subject: |
|
|
You can try this:
Boot your installCD, mount the partitions, /dev and /proc, chroot, env-update && source /etc/profile like before.
Run grub to find where the root is.: | Code: | ~ # grub
GNU GRUB version 0.97 (640K lower / 3072K upper memory)
[ Minimal BASH-like line editing is supported. For the first word, TAB
lists possible command completions. Anywhere else TAB lists the possible
completions of a device/filename. ]
grub> find /grub/stage1 <---------Try this one first-------------------<
grub> find /boot/grub/stage1 <---------Try this if the first fails------------<
(hd1,0) <---------------------USE the output here as your root-----------<
grub> root (hd1,0)
Filesystem type is ext2, partition type 0x83
grub> setup (hd0)
... succeeded
grub> quit
~ #
|
Make sure your /boot/grub/grub/conf has the corresponding values, recheck /etc/fstab.
exit the chroot, unmount partitions.
reboot. |
|
| Back to top |
|
 |
DONAHUE Veteran


Joined: 09 Dec 2006 Posts: 1565 Location: Goose Creek SC
|
Posted: Sun Feb 07, 2010 11:02 am Post subject: |
|
|
in grub.conf change | Quote: | | kernel /boot/kernel-2.6.31-gentoo-r6 root=/dev/sdb3 | to | Quote: | | kernel /boot/kernel-2.6.31-gentoo-r6 root=/dev/sda3 |
kernel is picking different device names than expected.
You do not have a grub problem; you have a kernel problem; specifically the kernel option root=
You can change to | Quote: | | kernel /boot/kernel-2.6.31-gentoo-r6 | and let the kernel find root on its own.
The good news is the kernel now has HDD controller drivers.
Is windows booting? |
|
| Back to top |
|
 |
todd93 Apprentice

Joined: 30 Oct 2006 Posts: 248
|
Posted: Sun Feb 07, 2010 4:50 pm Post subject: |
|
|
| DONAHUE wrote: | in grub.conf change | Quote: | | kernel /boot/kernel-2.6.31-gentoo-r6 root=/dev/sdb3 | to | Quote: | | kernel /boot/kernel-2.6.31-gentoo-r6 root=/dev/sda3 |
kernel is picking different device names than expected.
You do not have a grub problem; you have a kernel problem; specifically the kernel option root=
You can change to | Quote: | | kernel /boot/kernel-2.6.31-gentoo-r6 | and let the kernel find root on its own.
The good news is the kernel now has HDD controller drivers.
Is windows booting? |
Yes, windows boots fine.
Ok, I have altered /etc/fstab and changed the linux disks all to sda instead of sdb. I commented out the windows drive for now. Altered grub.conf to boot root=/dev/sda3. I am now in the kernel with only one problem: eth0 refuses to start. I have built the same module into the kernel that the livedvd was using for eth0, however no luck, without eth0 up and running, I'm really crippled, and am having trouble understanding what exactly the problem is. I think it's obviously something I have missed.
Thanks
Todd |
|
| Back to top |
|
 |
DONAHUE Veteran


Joined: 09 Dec 2006 Posts: 1565 Location: Goose Creek SC
|
Posted: Sun Feb 07, 2010 5:07 pm Post subject: |
|
|
| Quote: | | # CONFIG_E1000 is not set | appears in your config
lspci -kvm shows:
| Quote: | Device: 03:00.0
Class: Ethernet controller
Vendor: Realtek Semiconductor Co., Ltd.
Device: RTL8111/8168B PCI Express Gigabit Ethernet controller
SVendor: Giga-byte Technology
SDevice: GA-EP45-DS5 Motherboard
Rev: 02
Driver: r8169
Module: r8169
Device: 04:06.0
Class: Ethernet controller
Vendor: Intel Corporation
Device: 82541PI Gigabit Ethernet Controller
SVendor: Intel Corporation
SDevice: PRO/1000 GT Desktop Adapter
Rev: 05
Driver: e1000
Module: e1000 |
Suspect you are plugged into the intel nic which has no driver.
Edit menuconfig to add e1000 and e1000e as well as r8169 as modules, then
| Code: | make modules && make modules_install;
modprobe -r r8169;
modprobe r8169;
modprobe e1000;
ifconfig -a |
|
|
| Back to top |
|
 |
todd93 Apprentice

Joined: 30 Oct 2006 Posts: 248
|
Posted: Sun Feb 07, 2010 6:36 pm Post subject: |
|
|
| DONAHUE wrote: | | Quote: | | # CONFIG_E1000 is not set | appears in your config
lspci -kvm shows:
| Quote: | Device: 03:00.0
Class: Ethernet controller
Vendor: Realtek Semiconductor Co., Ltd.
Device: RTL8111/8168B PCI Express Gigabit Ethernet controller
SVendor: Giga-byte Technology
SDevice: GA-EP45-DS5 Motherboard
Rev: 02
Driver: r8169
Module: r8169
Device: 04:06.0
Class: Ethernet controller
Vendor: Intel Corporation
Device: 82541PI Gigabit Ethernet Controller
SVendor: Intel Corporation
SDevice: PRO/1000 GT Desktop Adapter
Rev: 05
Driver: e1000
Module: e1000 |
Suspect you are plugged into the intel nic which has no driver.
Edit menuconfig to add e1000 and e1000e as well as r8169 as modules, then
| Code: | make modules && make modules_install;
modprobe -r r8169;
modprobe r8169;
modprobe e1000;
ifconfig -a |
|
Alright, we're getting closer, and I'm starting to get anxious! followed the steps you have laid out and am receiving a fatal error after entering the command:
The error is as follows: | Code: | | FATAL: Error inserting r8169 (/lib/modules/2.6.31-gentoo-r6/kernel/drivers/net/r8169.ko): Invalid module format |
I appreciate all the help I'm getting.
Thanks
Todd |
|
| Back to top |
|
 |
Mike Hunt Advocate


Joined: 19 Jul 2009 Posts: 4671
|
Posted: Sun Feb 07, 2010 6:46 pm Post subject: |
|
|
Is the kernel you booted, the same one which you did: make modules_install ?
Was /boot mounted when you copied the new kernel build ? |
|
| Back to top |
|
 |
DONAHUE Veteran


Joined: 09 Dec 2006 Posts: 1565 Location: Goose Creek SC
|
Posted: Sun Feb 07, 2010 7:29 pm Post subject: |
|
|
i now see that r8169 was a built in.
a whole kernel recompile and recopy is probably needed. Sorry, I knew this until I forgot r8169 was builtin halfway into my last.
you should be able to live with r8169 builtin ( don't modprobe it until the recompile is complete)
what is ifconfig -a saying? |
|
| Back to top |
|
 |
todd93 Apprentice

Joined: 30 Oct 2006 Posts: 248
|
Posted: Sun Feb 07, 2010 8:43 pm Post subject: |
|
|
| Mike Hunt wrote: | Is the kernel you booted, the same one which you did: make modules_install ?
Was /boot mounted when you copied the new kernel build ? |
Yes, and yes. |
|
| Back to top |
|
 |
todd93 Apprentice

Joined: 30 Oct 2006 Posts: 248
|
Posted: Sun Feb 07, 2010 8:48 pm Post subject: |
|
|
| DONAHUE wrote: | i now see that r8169 was a built in.
a whole kernel recompile and recopy is probably needed. Sorry, I knew this until I forgot r8169 was builtin halfway into my last.
you should be able to live with r8169 builtin ( don't modprobe it until the recompile is complete)
what is ifconfig -a saying? |
I'm not sure I understand what is meant by recopy. I have done a whole kernel rebuild.
ifconfig -a is outputting the following: | Code: | lo (followed by a paragraph of routine information)
sit0 (also followed by a small paragraph of imformation) |
Both of the paragraphs can be posted if necessary, also, when I followed the original commands that you posted (modprobe -r r8169, etc, etc.) I actually had r8169 compiled as a module, and not built in.
Thanks
Todd |
|
| Back to top |
|
 |
DONAHUE Veteran


Joined: 09 Dec 2006 Posts: 1565 Location: Goose Creek SC
|
Posted: Sun Feb 07, 2010 10:15 pm Post subject: |
|
|
Something odd is happening.
As a check:
boot the cd
mount the gentoo partitions
run to verify both mounted
run | Code: | | ls -l /mnt/gentoo/boot |
verify there is only one kernel file listed and that it shows a date/time within the last couple of hours
if that checks out reboot into your install and run to verify the realtek and the intel cards are listed; if so look to see if they have drivers listed.
run and see if r8169 and e1000 are listed.
If you included:
| Quote: | General setup --->
<*> Kernel .config support
[*] Enable access to .config through /proc/config.gz |
in your kernel, you can run:
| Code: | zgrep -i r8169 /proc/config.gz
zgrep -i e1000 /proc/config.gz |
to verify that the running kernel used a config file that included r8169 and e1000
Does running
| Code: | | find /lib/modules//lib/modules/2.6.30-gentoo-r6/ -type f -iname '*.o' -or -iname '*.ko' | less |
show r8169 and e1000 in the modules library? |
|
| Back to top |
|
 |
DONAHUE Veteran


Joined: 09 Dec 2006 Posts: 1565 Location: Goose Creek SC
|
Posted: Sun Feb 07, 2010 10:46 pm Post subject: |
|
|
does the kernel have:
| Code: | [*] Enable loadable module support --->
--- Enable loadable module support
[ ] Forced module loading
[*] Module unloading
[*] Forced module unloading
[ ] Module versioning support
[ ] Source checksum for all modules |
|
|
| Back to top |
|
 |
todd93 Apprentice

Joined: 30 Oct 2006 Posts: 248
|
Posted: Mon Feb 08, 2010 12:20 am Post subject: |
|
|
| DONAHUE wrote: | does the kernel have:
| Code: | [*] Enable loadable module support --->
--- Enable loadable module support
[ ] Forced module loading
[*] Module unloading
[*] Forced module unloading
[ ] Module versioning support
[ ] Source checksum for all modules |
|
I appreciate your patience with me here, something odd is happening, I will complete the steps you laid out in the previous post tomorrow, however, "Forced module unloading" is not built into the kernel, should it be?
Thanks
Todd |
|
| Back to top |
|
 |
DONAHUE Veteran


Joined: 09 Dec 2006 Posts: 1565 Location: Goose Creek SC
|
Posted: Mon Feb 08, 2010 12:28 am Post subject: |
|
|
| I think I have forced unloading because it used to be default. I've had it for a long time with no harm resulting. After reading the help I'll probably remove it next compile. |
|
| Back to top |
|
 |
todd93 Apprentice

Joined: 30 Oct 2006 Posts: 248
|
Posted: Mon Feb 08, 2010 12:41 pm Post subject: |
|
|
| DONAHUE wrote: | Something odd is happening.
As a check:
boot the cd
mount the gentoo partitions
run to verify both mounted
run | Code: | | ls -l /mnt/gentoo/boot |
verify there is only one kernel file listed and that it shows a date/time within the last couple of hours
if that checks out reboot into your install |
This is all now verified to be correct.
| DONAHUE wrote: | run to verify the realtek and the intel cards are listed; if so look to see if they have drivers listed.
run and see if r8169 and e1000 are listed. |
OK, here's a red flag (I think) nothing intel is listed here. The only ethernet controller it lists is: | Code: | Realtek Semiconductor Co., LTD. RTL8111/8168b PCI Express Gigabit Ethernet controller (rev 03)
Subsystem: Giga-byte Technology GA-EP45-DS5 Motherboard
Kernel Modules: r8169 |
| DONAHUE wrote: | If you included:
| Quote: | General setup --->
<*> Kernel .config support
[*] Enable access to .config through /proc/config.gz |
in your kernel, you can run:
| Code: | zgrep -i r8169 /proc/config.gz
zgrep -i e1000 /proc/config.gz |
|
I have the items that you indicated built in to my kernel, however, when I enter the command:
| Code: | | zgrep -i r8169 /proc/config.gz |
I get the following error:
| Code: | | gzip: /proc/config.gz: No such file or directory |
This i assume is a problem?
| DONAHUE wrote: | to verify that the running kernel used a config file that included r8169 and e1000
Does running
| Code: | | find /lib/modules//lib/modules/2.6.30-gentoo-r6/ -type f -iname '*.o' -or -iname '*.ko' | less |
show r8169 and e1000 in the modules library? |
Yes, they are both listed.
I hope this information helps!!
Thanks!
Todd |
|
| Back to top |
|
 |
DONAHUE Veteran


Joined: 09 Dec 2006 Posts: 1565 Location: Goose Creek SC
|
Posted: Mon Feb 08, 2010 2:47 pm Post subject: |
|
|
Embarrassing confession time:
I've been looking at whiteghost lspci -kvm thinking it was yours run from the cd.
you don't have an Intel nic only a realtek 8111/8168.
you don't need e1000, remove when convenient
googling suggests that the r8169 module is not right for a realtek 8111/8168. Edit: aha!! Bugzilla Bug 298741
the correct r8168 has vanished from the kernel.
the r8168 apparently was supposed to be incorporated in r8169 but wasn't.
For awhile when both r8168 and r8169 were around the auto load would pick r8169 which would not work.
The fix then was to remove 8169 and force load r8168.
Now that r8168 has vanished, my semi-educated suggestion is:
go to:
LINUX driver for kernel 2.6.x and 2.4.x (Support x86 and x64)
and download "LINUX driver for kernel 2.6.x and 2.4.x (Support x86 and x64)" to your gentoo system (or transport the downloaded file to your gentoo install by whatever means available).
running the gentoo system, cd to the directory that contains the downloaded file (r8168-8.016.00.tar.bz2), run:
| Code: | tar xvjpf r8168-8.016.00.tar.bz2
cd r8168-8.016.00 |
read and follow the readme document
run to verify the realtek nic has driver r8168.
will address other anomalies in a bit.
Last edited by DONAHUE on Mon Feb 08, 2010 4:37 pm; edited 1 time in total |
|
| Back to top |
|
 |
DONAHUE Veteran


Joined: 09 Dec 2006 Posts: 1565 Location: Goose Creek SC
|
Posted: Mon Feb 08, 2010 3:07 pm Post subject: |
|
|
| Quote: | | nothing intel is listed here |
<-- whiteghost has, you don't, sorry about red herring
| Quote: | Realtek Semiconductor Co., LTD. RTL8111/8168b PCI Express Gigabit Ethernet controller (rev 03)
Subsystem: Giga-byte Technology GA-EP45-DS5 Motherboard
Kernel Modules: r8169 |
<-- r8169 was selected but not installed as driver, you got an incompatible message because r8169 is currently incompatible (I think)
| Quote: | I get the following error:
gzip: /proc/config.gz: No such file or directory |
<--does show a config file? if not, is strange but harmless. |
|
| Back to top |
|
 |
whiteghost Apprentice


Joined: 26 Jul 2009 Posts: 285 Location: nj xt 145 whiteghost inda hood
|
Posted: Mon Feb 08, 2010 4:41 pm Post subject: |
|
|
i have an intel pci card besides my onboard realtek.
i usually do not have intel pci plugged in.
i have noticed using various livecds the e1000 driver works for the realtek.
and we do have same hardware, r8169 works here.
mine | Quote: |
Device: 03:00.0
Class: Ethernet controller
Vendor: Realtek Semiconductor Co., Ltd.
Device: RTL8111/8168B PCI Express Gigabit Ethernet controller
SVendor: Giga-byte Technology
SDevice: GA-EP45-DS5 Motherboard
|
todd93 | Quote: | Device: 05:00.0
Class: Ethernet controller
Vendor: Realtek Semiconductor Co., Ltd.
Device: RTL8111/8168B PCI Express Gigabit Ethernet controller
SVendor: Giga-byte Technology
SDevice: GA-EP45-DS5 Motherboard
Rev: 03 |
edit: just read the bug report. interesting.
just built e1000e module and it does not work on my realtek
sorry if i'm getting off track _________________ www.informationclearinghouse.info
Last edited by whiteghost on Mon Feb 08, 2010 5:26 pm; edited 1 time in total |
|
| Back to top |
|
 |
|