Dear LQ users

I am trying to compile a kernel for my server, being this a complex task since there shouldn't be many errors

My question is, were can I find information about block devices?
I remember that I did found a usefull documentation in the kernel source, which listed all the devices. However I didn't take note and I really want my server to speed up with a custom kernel.

Here's the error:

Kmod: Failed to exec /sbn/modprobe -s -k block-major-33. errno =2
VFS: Cannot open root device "hde3" or 21:03
please append a correct "root=" boot partition
Kernel Panic: VFS: Unable to mount root fs on 21:03

Here's lilo info:
image=/boot/vmlinuz-2.4.26
label="TESTING_KERNEL"
initrd=/boot/initrd-2.4.26.img
read-only
#append="root=/dev/hde3"
root=/dev/hde3

Any ideas?

Hello Xavier!!

That LILO config should get you started, unless of course the root filesystem does NOT reside inside hde3.

Only by this "hde" drive, I'm bold enough as to guess you are using SATA disks, if this is the case and if you are using Libata instead of IDE/ATA/MRL, etc support you drive will no longer be hde, but sda, you'll just have to boot into rescue mode (which will still see the drive as hde) and re-map the devices (in lilo.conf and fstab) to the SCSI bus.

Hope this helps.

What I did actually (and I don't think it's listed in other parts of Linuxquestions.org) is that for ALL the:

block-major-xx

problems, you could go to your kernel source (ex: /usr/src/linux-2.x.xx) and read the

Documentation/devices.txt

In my case, here is what I found

33 = /dev/st1l Second SCSI tape, mode 1

Therefore, I compiled all of the SCSI as modules (exepct 'scsi support', hope it works now)

Regards

Good luck!

Dear LQ users:

After I install kernel 2.4.26 apache doesn't starts

I receive this error:

[Sun Jul 04 23:01:54 2004] [notice] suEXEC mechanism enabled (wrapper: /usr/local/apache-2.0.49/bin/suexec)
[Sun Jul 04 23:01:54 2004] [crit] (38)Function not implemented: mod_rewrite: could not create rewrite_log_lock
Configuration Failed

Any ideas?

It's all solved

I had to reconfigure, make, and make install apache

I suppose this apache will now work with my previous kernel as well

Hi,
I meet a kernel problem after a compile the 2.6.7, when I reboot the pc with the new kernel, I recieve this message:
Kernel panic: VFS: Unable to mount root vfs on unknow- block(0,0)

my origrinal version is 2.4.0 now it still can work, and during my configurition, except some warning, everyting seems to be ok. Here is my grub.conf script:
some friends say delete hdc-idc-scsi will be ok, but it doesn't work for me :cry:

Hope someone could help me :rolleyes: thx any way for every help

Dear LQ users

I noticed that my kernel is not loading the modules listed in /boot/initrd-2.4.6

I did created the file manually with "mkinitrd -v" (verbbose) and I was able to see how, jfs, ext3, aic7xxx, sr_mod... were loaded in the initrd

However, when In configured GRUB to use other initrd file (which does loads the aic7xxx module for kernel 2.4.20), I noticed that the kernel simply doesn't loads any module in the initrd file

Why does that happens?

Dear LQ Users

I successfully compiled the 2.4.26 kernel for my RedHat 9 system, however I noticed that the aic7xxx module was not loading at startup.

checking at the config, I noticed that the initial Ram disk support was disabled.

I made all the neccesary changes, recompiled the kernel, run the 'mkinitrd -v' command (in which I saw the ext3, jbd, aic7xxx... modules built into /boot/initrd-2.4.6), restarted the machine and this is what I see

Loading aic7xxx module
unresolved symbol bread-RSS74607
unresolved symbol scsi-unblock_requests_R&ae42de
unresolved symbol xxxx
unresolved symbol yyyy
unresolved symbol zzzz

For every module compressed into the initrd-2.4.6.img, file, I see tons of those errors

Finally, I receive a Kernel Panic.

If I boot the same kernel without the initrd-2.4.6.img file, then it boots correctly. However I have to manually load the aic7xxx module if I want to use my CD-R

Changing subs, Thetargos are you sponsored by LQ in some way?
I can't imagine how valueble is your help for this forum

morriset
change hdc-ide-scsi to hdc=ide-scsi and check your .config to see if you have your filesystem supported

Kernel Panic
 

I am trying to install a 2.6.8.1 kernel alongside the 2.4.20-8 kernel that came with my Red Hat 9 installation. Everything went well until I tried to boot the system with the new kernel, where I got these error messages:

Here is a copy of my /boot/grub/menu.lst:

And here are the results of a ls -l in /boot:

Thanks in advance for you help :)

Hey this is great!
I've got my first linux kernel upgrade by myself and my first kernel rpm !! :D:D:D:D:D

I have seen a lot of people here with lot of panics and fails.
I think that there is something missing for rh9. It's everybody copying their old config file in their new kernel source root with the name ".config"?? I think that the impatients got the commands and executed them.

Hello and sorry for the late reply...

Are you building your kernel image with the filesystem (ext2, ext3, etc) support into the image (i.e not as a module)? And you may need to substitute the root=LABEL=/ statement from your grub.conf for the actual partition that holds your root partition (/dev/hda2 in my case, check your partitions).

Congratulations!!

Hello Guys,

Can someone please help me take a look on this thread ?

http://www.linuxquestions.org/questi...61#post1135761