I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00. I'm looking through xconfig to see if I don't have devfs setup. I used the .config included with fedora core kernel source. It doesn't surprise me that the .config isn't accurate - I couldn't get the fedora core source to compile acpi, so i think it's not an accurate source tree.
Any pointers the devfs issue is appreciated.
thanks
On Sun, 2003-10-05 at 10:45, W. Chris Shank wrote:
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00. I'm looking through xconfig to see if I don't have devfs setup. I used the .config included with fedora core kernel source. It doesn't surprise me that the .config isn't accurate - I couldn't get the fedora core source to compile acpi, so i think it's not an accurate source tree.
You probably want to use the config files found in /boot that you get from installing a kernel image.
Any pointers the devfs issue is appreciated.
thanks
On Sun, 2003-10-05 at 08:03, Sean Middleditch wrote:
On Sun, 2003-10-05 at 10:45, W. Chris Shank wrote:
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00. I'm looking through xconfig to see if I don't have devfs setup. I used the .config included with fedora core kernel source. It doesn't surprise me that the .config isn't accurate - I couldn't get the fedora core source to compile acpi, so i think it's not an accurate source tree.
You probably want to use the config files found in /boot that you get from installing a kernel image.
actually, I've had to edit /boot/grub/grub.conf (or /etc/grub.conf) to reflect the actual partiion instead of the device label for the new kernel's entry. Never figured out another way around it.
-josh
Tried that too. It gets farther along - the screen does blank and it seems to stall. This is on a laptop - ASUS M3000N, so fixing the ACPI and installing a custom kernel is important for me.
On Sun, 2003-10-05 at 11:26, Joshua Legbandt wrote:
On Sun, 2003-10-05 at 08:03, Sean Middleditch wrote:
On Sun, 2003-10-05 at 10:45, W. Chris Shank wrote:
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00. I'm looking through xconfig to see if I don't have devfs setup. I used the .config included with fedora core kernel source. It doesn't surprise me that the .config isn't accurate - I couldn't get the fedora core source to compile acpi, so i think it's not an accurate source tree.
You probably want to use the config files found in /boot that you get from installing a kernel image.
actually, I've had to edit /boot/grub/grub.conf (or /etc/grub.conf) to reflect the actual partiion instead of the device label for the new kernel's entry. Never figured out another way around it.
-josh
Still won't compile:
kernel/kernel.o(.text+0xf9f): In function `schedule': : undefined reference to `active_load_balance' make[1]: *** [kallsyms] Error 1 make[1]: Leaving directory `/usr/src/linux-2.4.22-1.2061.nptl' make: *** [vmlinux] Error 2
Me thinks the source tree is out of sync with the kernel delivered.
On Sun, 2003-10-05 at 11:03, Sean Middleditch wrote:
On Sun, 2003-10-05 at 10:45, W. Chris Shank wrote:
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00. I'm looking through xconfig to see if I don't have devfs setup. I used the .config included with fedora core kernel source. It doesn't surprise me that the .config isn't accurate - I couldn't get the fedora core source to compile acpi, so i think it's not an accurate source tree.
You probably want to use the config files found in /boot that you get from installing a kernel image.
Any pointers the devfs issue is appreciated.
thanks
On Sun, 2003-10-05 at 11:35, W. Chris Shank wrote:
Still won't compile:
kernel/kernel.o(.text+0xf9f): In function `schedule': : undefined reference to `active_load_balance' make[1]: *** [kallsyms] Error 1 make[1]: Leaving directory `/usr/src/linux-2.4.22-1.2061.nptl' make: *** [vmlinux] Error 2
Me thinks the source tree is out of sync with the kernel delivered.
Which patches did you install? Perhaps some patch you used didn't work quite well with those already in the RH kernel.
On Sun, 2003-10-05 at 11:03, Sean Middleditch wrote:
On Sun, 2003-10-05 at 10:45, W. Chris Shank wrote:
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00. I'm looking through xconfig to see if I don't have devfs setup. I used the .config included with fedora core kernel source. It doesn't surprise me that the .config isn't accurate - I couldn't get the fedora core source to compile acpi, so i think it's not an accurate source tree.
You probably want to use the config files found in /boot that you get from installing a kernel image.
Any pointers the devfs issue is appreciated.
thanks
I didn't apply any patches to the stock fedora kernel: linux-2.4.22-1.2061.nptl
I applied the acpi patches to the kernel.org 2.4.22 kernel. I'm attempting to compile the vanilla kernel with the fedora config.
On Sun, 2003-10-05 at 11:45, Sean Middleditch wrote:
On Sun, 2003-10-05 at 11:35, W. Chris Shank wrote:
Still won't compile:
kernel/kernel.o(.text+0xf9f): In function `schedule': : undefined reference to `active_load_balance' make[1]: *** [kallsyms] Error 1 make[1]: Leaving directory `/usr/src/linux-2.4.22-1.2061.nptl' make: *** [vmlinux] Error 2
Me thinks the source tree is out of sync with the kernel delivered.
Which patches did you install? Perhaps some patch you used didn't work quite well with those already in the RH kernel.
On Sun, 2003-10-05 at 11:03, Sean Middleditch wrote:
On Sun, 2003-10-05 at 10:45, W. Chris Shank wrote:
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00. I'm looking through xconfig to see if I don't have devfs setup. I used the .config included with fedora core kernel source. It doesn't surprise me that the .config isn't accurate - I couldn't get the fedora core source to compile acpi, so i think it's not an accurate source tree.
You probably want to use the config files found in /boot that you get from installing a kernel image.
Any pointers the devfs issue is appreciated.
thanks
On Sun, 05 Oct 2003 11:35:22 -0400, W. Chris Shank wrote:
Still won't compile:
kernel/kernel.o(.text+0xf9f): In function `schedule': : undefined reference to `active_load_balance' make[1]: *** [kallsyms] Error 1 make[1]: Leaving directory `/usr/src/linux-2.4.22-1.2061.nptl' make: *** [vmlinux] Error 2
Me thinks the source tree is out of sync with the kernel delivered.
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=105978
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=105979
--Kai
This looks promising - thanks
On Sun, 2003-10-05 at 11:57, Kai Thomsen wrote:
On Sun, 05 Oct 2003 11:35:22 -0400, W. Chris Shank wrote:
Still won't compile:
kernel/kernel.o(.text+0xf9f): In function `schedule': : undefined reference to `active_load_balance' make[1]: *** [kallsyms] Error 1 make[1]: Leaving directory `/usr/src/linux-2.4.22-1.2061.nptl' make: *** [vmlinux] Error 2
Me thinks the source tree is out of sync with the kernel delivered.
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=105978
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=105979
--Kai
-- fedora-list mailing list fedora-list@redhat.com http://www.redhat.com/mailman/listinfo/fedora-list
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Sunday 05 October 2003 09:45 am, W. Chris Shank wrote:
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00. I'm looking through xconfig to see if I don't have devfs setup. I used the .config included with fedora core kernel source. It doesn't surprise me that the .config isn't accurate - I couldn't get the fedora core source to compile acpi, so i think it's not an accurate source tree.
Any pointers the devfs issue is appreciated.
thanks
I have this same problem with 2.4.20 on RH8. Is there anyway to remove a FS label? That appears to be the problem.
Mike W - -- Registered Linux - 256979 NRA Life ARS: W0TMW
From: "Mike Watson" mikew@crucis.net
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Sunday 05 October 2003 09:45 am, W. Chris Shank wrote:
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00. I'm looking through xconfig to see if I don't have devfs setup. I used the .config included with fedora core kernel source. It doesn't surprise me that the .config isn't accurate - I couldn't get the fedora core source to compile acpi, so i think it's not an accurate source tree.
Any pointers the devfs issue is appreciated.
thanks
I have this same problem with 2.4.20 on RH8. Is there anyway to remove a FS label? That appears to be the problem.
Have you tried editing /etc/fstab to change the format from "LABEL=/" to "/dev/hda2" or the like? {^_^}
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Sunday 05 October 2003 06:46 pm, jdow wrote:
From: "Mike Watson" mikew@crucis.net
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Sunday 05 October 2003 09:45 am, W. Chris Shank wrote:
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00. I'm looking through xconfig to see if I don't have devfs setup. I used the .config included with fedora core kernel source. It doesn't surprise me that the .config isn't accurate - I couldn't get the fedora core source to compile acpi, so i think it's not an accurate source tree.
Any pointers the devfs issue is appreciated.
thanks
I have this same problem with 2.4.20 on RH8. Is there anyway to remove a FS label? That appears to be the problem.
Have you tried editing /etc/fstab to change the format from "LABEL=/" to "/dev/hda2" or the like? {^_^}
I'd like to say I did, but I don't remember exactly....
Will check it out.
Mike W - -- Registered Linux - 256979 NRA Life ARS: W0TMW
Yes. It didn't help
On Sun, 2003-10-05 at 19:46, jdow wrote:
From: "Mike Watson" mikew@crucis.net
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Sunday 05 October 2003 09:45 am, W. Chris Shank wrote:
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00. I'm looking through xconfig to see if I don't have devfs setup. I used the .config included with fedora core kernel source. It doesn't surprise me that the .config isn't accurate - I couldn't get the fedora core source to compile acpi, so i think it's not an accurate source tree.
Any pointers the devfs issue is appreciated.
thanks
I have this same problem with 2.4.20 on RH8. Is there anyway to remove a FS label? That appears to be the problem.
Have you tried editing /etc/fstab to change the format from "LABEL=/" to "/dev/hda2" or the like? {^_^}
-- fedora-list mailing list fedora-list@redhat.com http://www.redhat.com/mailman/listinfo/fedora-list
Um, I just happened to think that if you are using devfs you probably need to change the "/dev/hda2" format to something devfs likes. I've carefully avoided stepping into that pile of doodoo so beyond this I cannot help. {^_^} ----- Original Message ----- From: "W. Chris Shank" chris.shank@acetechgroup.com
Yes. It didn't help
On Sun, 2003-10-05 at 19:46, jdow wrote:
Have you tried editing /etc/fstab to change the format from "LABEL=/" to "/dev/hda2" or the like? {^_^}
On Mon, 2003-10-06 at 07:46, jdow wrote:
From: "Mike Watson" mikew@crucis.net
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00.
[snip other problem desc]
Have you tried editing /etc/fstab to change the format from "LABEL=/" to "/dev/hda2" or the like?
On the above solution I wanted to ask one question. On RedHat systems if I download a vanilla kernel and compile and install it successfully why do I have to change the fstab format to /dev/hdX? Why can't I let it remain in the LABEL=/** format. I personally don't have any issues with the LABEL format but everytime I install a vanilla kernel either I have to pass the root=/dev/hdX note it does not work if I pass root=LABEL=/ like redhat does. I just wanted to know why? It has just a minor irritation which I could not understand.
On Tue, 2003-10-07 at 09:02, Arindam Dey wrote:
On Mon, 2003-10-06 at 07:46, jdow wrote:
From: "Mike Watson" mikew@crucis.net
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/ or 00:00.
[snip other problem desc]
Have you tried editing /etc/fstab to change the format from "LABEL=/" to "/dev/hda2" or the like?
On the above solution I wanted to ask one question. On RedHat systems if I download a vanilla kernel and compile and install it successfully why do I have to change the fstab format to /dev/hdX? Why can't I let it remain in the LABEL=/** format. I personally don't have any issues with the LABEL format but everytime I install a vanilla kernel either I have to pass the root=/dev/hdX note it does not work if I pass root=LABEL=/ like redhat does. I just wanted to know why? It has just a minor irritation which I could not understand.
sct already answered that one:
Sounds like you need an initrd. The kernel can't parse mount labels, but if you use an initrd the real root fs gets mounted by user space instead.
--Stephen
Nils
On Fri, 2003-10-10 at 03:15, Nils Philippsen thus hollered from the rooftop:
sct already answered that one:
Ah thank you, my mistake should have paid closer attention.
I have this same problem with 2.4.20 on RH8. Is there anyway to remove a FS label? That appears to be the problem.
The program 'e2label' is just what you're looking for.
Cheers, e.
I have this same problem with 2.4.20 on RH8. Is there anyway to remove a FS label? That appears to be the problem.
The program 'e2label' is just what you're looking for.
Or go old skool:
tune2fs -L label-name /dev/hdX
-Chuck
-- Quantum Linux Laboratories - ACCELERATING Business with Open Technology * Education | -=^ Ad Astra Per Aspera ^=- * Integration | http://www.quantumlinux.com * Support | chuckw@quantumlinux.com
A: Because we read from top to bottom, left to right. Q: Why should i start my reply below the quoted text?
From: "Chuck Wolber" chuckw@quantumlinux.com
I have this same problem with 2.4.20 on RH8. Is there anyway to remove a FS label? That appears to be the problem.
The program 'e2label' is just what you're looking for.
Or go old skool:
tune2fs -L label-name /dev/hdX
"Logically" if the fstab "LABEL=/" does not work with devfs then some translation process is not taking place correctly. That means the disk label MAY mean nothing. If its presence guarantees a mistranslation then the only help is "e2label <path to device>" to create a blank or empty label. If you figure out what the form is for <path to device> simply use that in fstab and be done with it.
I wonder if this is somehow related to having the wrong mod-utils with the kernel package.
{^_^} I also admit, I don't guarantee the logic above. There could be a fatal issue with devfs that means you just aren't going to boot using devfs the way you want.
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Sunday 05 October 2003 10:47 pm, jdow wrote:
From: "Chuck Wolber" chuckw@quantumlinux.com
I have this same problem with 2.4.20 on RH8. Is there anyway to remove a FS label? That appears to be the problem.
The program 'e2label' is just what you're looking for.
Or go old skool:
tune2fs -L label-name /dev/hdX
"Logically" if the fstab "LABEL=/" does not work with devfs then some translation process is not taking place correctly. That means the disk label MAY mean nothing. If its presence guarantees a mistranslation then the only help is "e2label <path to device>" to create a blank or empty label. If you figure out what the form is for <path to device> simply use that in fstab and be done with it.
I wonder if this is somehow related to having the wrong mod-utils with the kernel package.
{^_^} I also admit, I don't guarantee the logic above. There could be a fatal issue with devfs that means you just aren't going to boot using devfs the way you want.
The parameters for e2lable is e2label <device> <label>
If no 2nd parameter is furnished, e2label returns the value of the existing label. So how do you specific a blank label?
Mike W - -- Registered Linux - 256979 NRA Life ARS: W0TMW
Try this:
[erik@woody ~erik]$ e2label /dev/hda2 /usr [erik@woody ~erik]$ e2label /dev/hda2 "" [erik@woody ~erik]$ e2label /dev/hda2
[erik@woody ~erik]$
Hope this helps, Erik.
Mike Watson wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Sunday 05 October 2003 10:47 pm, jdow wrote:
From: "Chuck Wolber" chuckw@quantumlinux.com
I have this same problem with 2.4.20 on RH8. Is there anyway to remove a FS label? That appears to be the problem.
The program 'e2label' is just what you're looking for.
Or go old skool:
tune2fs -L label-name /dev/hdX
"Logically" if the fstab "LABEL=/" does not work with devfs then some translation process is not taking place correctly. That means the disk label MAY mean nothing. If its presence guarantees a mistranslation then the only help is "e2label <path to device>" to create a blank or empty label. If you figure out what the form is for <path to device> simply use that in fstab and be done with it.
I wonder if this is somehow related to having the wrong mod-utils with the kernel package.
{^_^} I also admit, I don't guarantee the logic above. There could be a fatal issue with devfs that means you just aren't going to boot using devfs the way you want.
The parameters for e2lable is e2label <device> <label>
If no 2nd parameter is furnished, e2label returns the value of the existing label. So how do you specific a blank label?
Mike W
Registered Linux - 256979 NRA Life ARS: W0TMW
-----BEGIN PGP SIGNATURE----- Version: GnuPG v1.0.7 (GNU/Linux)
iD8DBQE/gf9H5fq6h2uDDlQRAq+wAKCZsdR6pPInrEUIgnfDrvUHLnVRFACeIA44 WvWwiiHlyK8Xh+gx0MQb2YI= =UYdk -----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Monday 06 October 2003 07:16 pm, Erik Williamson wrote:
Try this:
[erik@woody ~erik]$ e2label /dev/hda2 /usr [erik@woody ~erik]$ e2label /dev/hda2 "" [erik@woody ~erik]$ e2label /dev/hda2
[erik@woody ~erik]$
Hope this helps, Erik.
Thanks! I'll try'em.
Mike W - -- Registered Linux - 256979 NRA Life ARS: W0TMW
From: "Mike Watson" mikew@crucis.net
On Sunday 05 October 2003 10:47 pm, jdow wrote:
From: "Chuck Wolber" chuckw@quantumlinux.com
I have this same problem with 2.4.20 on RH8. Is there anyway to remove a FS label? That appears to be the problem.
The program 'e2label' is just what you're looking for.
Or go old skool:
tune2fs -L label-name /dev/hdX
"Logically" if the fstab "LABEL=/" does not work with devfs then some translation process is not taking place correctly. That means the disk label MAY mean nothing. If its presence guarantees a mistranslation then the only help is "e2label <path to device>" to create a blank or empty label. If you figure out what the form is for <path to device> simply use that in fstab and be done with it.
I wonder if this is somehow related to having the wrong mod-utils with the kernel package.
{^_^} I also admit, I don't guarantee the logic above. There could be a fatal issue with devfs that means you just aren't going to boot using devfs the way you want.
The parameters for e2lable is e2label <device> <label>
If no 2nd parameter is furnished, e2label returns the value of the existing label. So how do you specific a blank label?
Normally that is an exercise left to the newby. But I'll be nice.
e2label <device> ""
That is an empty label which should be the same as no label.
{^_^}
Hi,
On Sun, 2003-10-05 at 15:45, W. Chris Shank wrote:
I tried to compile a custom 2.4.22 kernel - when I boot it I get a VFS kernel panic error saying it cannot open root device LABEL=/
Sounds like you need an initrd. The kernel can't parse mount labels, but if you use an initrd the real root fs gets mounted by user space instead.
--Stephen
-
Arindam Dey -
Chuck Wolber -
Erik Williamson -
jdow -
Joshua Legbandt -
Kai Thomsen -
Mike - W0TMW -
Nils Philippsen -
Sean Middleditch -
Stephen C. Tweedie -
W. Chris Shank