David Thompson wrote:
Hi all:
I'm trying to rebuild the anaconda installer for Centos 4.0 (i386) to add some
RPMs and do some other things, and I've run into a problem. I've done lots of
anaconda rebuilds for Tao Linux 1.0, and I'm bringing those bits forward into Centos 4.0. The all-knowing web hasn't turned up anything useful. It looks like the PXE-loaded kernel unpacked and mounted the initrd OK, but then things
went awry (quickly). Anyone seen this on a PXE boot of an anaconda build?
<snip> RAMDISK: Compressed image found at block 0 VFS: Mounted root (ext2 filesystem). VFS: Cannot open root device "<NULL>" or unknown-block(3,2) Please append a correct "root=" boot option Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(3,2)
Well, since no one replied to this I'll follow up my own post.
It turns out that the 2.6.9-5.0.3.EL series of kernels has a bug in loop back mounting file systems:
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=105595
(the bug is filed against fc3 but affects the original centos 4.0 kernels as well.) The bug prevents the initrd.img in anaconda from being populated, and everything goes kaplooey as soon as the PXE boot tries to do anything with it's initrd. It would also affect rebuilt ISO images.
The fix is to upgrade the host running buildinstall to a 2.6.9-5.0.5.EL kernel, which has a fix for bug 105595. Then everything works just fine.
-- Dave Thompson ( thomas(at)cs(dot)wisc(dot)edu )
Associate Researcher Department of Computer Science University of Wisconsin-Madison http://www.cs.wisc.edu/(twiddle)thomas 1210 West Dayton Street Madison, WI 53706-1685 --
On Wed, 2005-04-27 at 14:24 -0500, David Thompson wrote:
David Thompson wrote:
Hi all:
I'm trying to rebuild the anaconda installer for Centos 4.0 (i386) to add some
RPMs and do some other things, and I've run into a problem. I've done lots of
anaconda rebuilds for Tao Linux 1.0, and I'm bringing those bits forward into Centos 4.0. The all-knowing web hasn't turned up anything useful. It looks like the PXE-loaded kernel unpacked and mounted the initrd OK, but then things
went awry (quickly). Anyone seen this on a PXE boot of an anaconda build?
<snip> RAMDISK: Compressed image found at block 0 VFS: Mounted root (ext2 filesystem). VFS: Cannot open root device "<NULL>" or unknown-block(3,2) Please append a correct "root=" boot option Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(3,2)
Well, since no one replied to this I'll follow up my own post.
It turns out that the 2.6.9-5.0.3.EL series of kernels has a bug in loop back mounting file systems:
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=105595
(the bug is filed against fc3 but affects the original centos 4.0 kernels as well.) The bug prevents the initrd.img in anaconda from being populated, and everything goes kaplooey as soon as the PXE boot tries to do anything with it's initrd. It would also affect rebuilt ISO images.
The fix is to upgrade the host running buildinstall to a 2.6.9-5.0.5.EL kernel, which has a fix for bug 105595. Then everything works just fine.
-- Dave Thompson ( thomas(at)cs(dot)wisc(dot)edu )
Associate Researcher Department of Computer Science University of Wisconsin-Madison http://www.cs.wisc.edu/(twiddle)thomas 1210 West Dayton Street Madison, WI 53706-1685 --
Thank you very much for the follow up post.
On Wed, Apr 27, 2005 at 05:02:58PM -0500, Johnny Hughes wrote:
On Wed, 2005-04-27 at 14:24 -0500, David Thompson wrote:
David Thompson wrote:
Hi all:
I'm trying to rebuild the anaconda installer for Centos 4.0 (i386) to add some
RPMs and do some other things, and I've run into a problem. I've done lots of
anaconda rebuilds for Tao Linux 1.0, and I'm bringing those bits forward into Centos 4.0. The all-knowing web hasn't turned up anything useful. It looks like the PXE-loaded kernel unpacked and mounted the initrd OK, but then things
went awry (quickly). Anyone seen this on a PXE boot of an anaconda build?
<snip> RAMDISK: Compressed image found at block 0 VFS: Mounted root (ext2 filesystem). VFS: Cannot open root device "<NULL>" or unknown-block(3,2) Please append a correct "root=" boot option Kernel panic - not syncing: VFS: Unable to mount root fs on unknown-block(3,2)
Well, since no one replied to this I'll follow up my own post.
It turns out that the 2.6.9-5.0.3.EL series of kernels has a bug in loop back mounting file systems:
https://bugzilla.redhat.com/bugzilla/show_bug.cgi?id=105595
(the bug is filed against fc3 but affects the original centos 4.0 kernels as well.) The bug prevents the initrd.img in anaconda from being populated, and everything goes kaplooey as soon as the PXE boot tries to do anything with it's initrd. It would also affect rebuilt ISO images.
The fix is to upgrade the host running buildinstall to a 2.6.9-5.0.5.EL kernel, which has a fix for bug 105595. Then everything works just fine.
-- Dave Thompson ( thomas(at)cs(dot)wisc(dot)edu )
Associate Researcher Department of Computer Science University of Wisconsin-Madison http://www.cs.wisc.edu/(twiddle)thomas 1210 West Dayton Street Madison, WI 53706-1685 --
Thank you very much for the follow up post.
Thank you very much for your thank you very much post. Btw, check out this link: http://www.river.com/users/share/etiquette/
/Morten
-
David Thompson -
Johnny Hughes -
Morten Kjeldgaard