I inadvertently missed using the list...here are my recent messages.
On 3/8/2011 8:39 PM, Nico Kadel-Garcia wrote:
On Tue, Mar 8, 2011 at 11:31 PM, Todd Carytodd@aristesoftware.com wrote:
reinstall is not an option for yum. I ran "yum install kernel" and it completed without errors however there are no links created.
Oh, dear. Can you grab the RPM and do "rpm -U -replacepkgs [kernel-whatver].rpm"? You should be able to use "yum remove" on the old kernel packages, consistent with freeing up the space, and now install your new kernel with yum.
Would this be the correct ln command for vmlinuz-2.6.9-89.35.1
# /boot/vmlinuz-2.6.9-89.35.1 /boot/vmlinuz
Todd
On 3/8/2011 7:04 PM, Nico Kadel-Garcia wrote:
On Tue, Mar 8, 2011 at 9:58 PM, Todd Carytodd@aristesoftware.com wrote:
I started a new thread since the original one is getting rather long.
I have retrieved the files I deleted in /boot and /boot/grub, however I need to make links for
/boot/System.map (System.map -> System.map-2.6.9-89.35.1) /boot/vmlinuz (vmlinuz -> vmlinuz-2.6.9-89.35.1) /boot/grub/menu.lst (menu.lst -> ./grub.conf)
Instead, re-install your kernel. "yum reinstall kernel". This should regenerate your symlinks correctly, except possibly the grub.conf.
If it was not so important to get it correct, I would appreciate the syntax for the command. Usually I would figure it out.
Since I have restored the files (I will double check to make sure they are all there), do I need to run grub-install?
i think yes. The old location of the boot loader is listed in /boot/grub/grub.conf, and should be used as the argument to that command. grub is much smarter than LILO used to be, but I think the bootstrap procedure relies on knowing details of where the fiddly bits of grub live on the relevant ex2 compatible filesytem.
My apologies for bothering everyone with such a dumb error on my part.
Todd
-- Ariste Software Petaluma, CA 94952
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- Ariste Software Petaluma, CA 94952
I inadvertently missed using the list...here are my recent messages.
As Nico suggested, download the kernel but also grub and redhat-logos, like so wget http://mirrors.kernel.org/centos/4.9/updates/i386/RPMS/kernel-2.6.9-100.EL.i... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/redhat-logos-1.1.26... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/grub-0.95-3.8.i386....
Then do a
rpm -Uvh --replacepkgs --replacefiles kernel-2.6.9-100.EL.i686.rpm redhat-logos-1.1.26-1.centos4.4.noarch.rpm grub-0.95-3.8.i386.rpm
And the show us the contents of 'ls -laR /boot' and 'cat /etc/grub.conf'
Simon
On 3/8/2011 8:39 PM, Nico Kadel-Garcia wrote:
On Tue, Mar 8, 2011 at 11:31 PM, Todd Carytodd@aristesoftware.com wrote:
reinstall is not an option for yum. I ran "yum install kernel" and it completed without errors however there are no links created.
Oh, dear. Can you grab the RPM and do "rpm -U -replacepkgs [kernel-whatver].rpm"? You should be able to use "yum remove" on the old kernel packages, consistent with freeing up the space, and now install your new kernel with yum.
Would this be the correct ln command for vmlinuz-2.6.9-89.35.1
# /boot/vmlinuz-2.6.9-89.35.1 /boot/vmlinuz
Todd
On 3/8/2011 7:04 PM, Nico Kadel-Garcia wrote:
On Tue, Mar 8, 2011 at 9:58 PM, Todd Carytodd@aristesoftware.com wrote:
I started a new thread since the original one is getting rather long.
I have retrieved the files I deleted in /boot and /boot/grub, however I need to make links for
/boot/System.map (System.map -> System.map-2.6.9-89.35.1) /boot/vmlinuz (vmlinuz -> vmlinuz-2.6.9-89.35.1) /boot/grub/menu.lst (menu.lst -> ./grub.conf)
Instead, re-install your kernel. "yum reinstall kernel". This should regenerate your symlinks correctly, except possibly the grub.conf.
If it was not so important to get it correct, I would appreciate the syntax for the command. Usually I would figure it out.
Since I have restored the files (I will double check to make sure they are all there), do I need to run grub-install?
i think yes. The old location of the boot loader is listed in /boot/grub/grub.conf, and should be used as the argument to that command. grub is much smarter than LILO used to be, but I think the bootstrap procedure relies on knowing details of where the fiddly bits of grub live on the relevant ex2 compatible filesytem.
My apologies for bothering everyone with such a dumb error on my part.
Todd
-- Ariste Software Petaluma, CA 94952
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- Ariste Software Petaluma, CA 94952
-- Ariste Software Petaluma, CA 94952
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Simon -
I performed the tasks as outlined and here are the contents of /boot:
/boot/: total 25121 drwxr-xr-x 5 root root 9216 Mar 9 10:31 . drwxr-xr-x 24 root root 4096 Jan 24 08:34 .. -rw-r--r-- 1 root root 51676 Feb 17 22:41 config-2.6.9-100.EL drwxr-xr-x 2 root root 1024 Mar 9 10:31 grub -rw-r--r-- 1 root root 444812 May 5 2007 grub-0.95-3.8.i386.rpm -rw-r--r-- 1 root root 1343054 Mar 9 10:29 initrd-2.6.9-100.EL.img -rw-r--r-- 1 root root 13409764 Feb 18 06:25 kernel-2.6.9-100.EL.i686.rpm drwx------ 2 root root 12288 Jan 12 2007 lost+found -rw-r--r-- 1 root root 9371 Aug 12 2006 message -rw-r--r-- 1 root root 9371 Aug 12 2006 message.ja -rw-r--r-- 1 root root 7919724 Aug 13 2006 redhat-logos-1.1.26-1.centos4.4.noarch.rpm -rw-r--r-- 1 root root 67797 Feb 17 22:41 symvers-2.6.9-100.EL.gz -rw-r--r-- 1 root root 770652 Feb 17 22:41 System.map-2.6.9-100.EL drwx------ 2 root root 9216 Mar 9 10:03 .Trash-root -rw-r--r-- 1 root root 1538264 Feb 17 22:41 vmlinuz-2.6.9-100.EL
/boot/grub: total 345 drwxr-xr-x 2 root root 1024 Mar 9 10:31 . drwxr-xr-x 5 root root 9216 Mar 9 10:31 .. -rw-r--r-- 1 root root 82 Mar 8 17:39 device.map -rw-r--r-- 1 root root 7956 Mar 8 17:40 e2fs_stage1_5 -rw-r--r-- 1 root root 7684 Mar 8 17:40 fat_stage1_5 -rw-r--r-- 1 root root 6996 Mar 8 17:40 ffs_stage1_5 -rw------- 1 root root 599 Mar 9 10:31 grub.conf -rw-r--r-- 1 root root 7028 Mar 8 17:40 iso9660_stage1_5 -rw-r--r-- 1 root root 8448 Mar 8 17:40 jfs_stage1_5 -rw------- 1 root root 4240 Mar 8 20:24 menu.lst -rw-r--r-- 1 root root 7188 Mar 8 17:40 minix_stage1_5 -rw-r--r-- 1 root root 9396 Mar 8 17:40 reiserfs_stage1_5 -rw-r--r-- 1 root root 3605 Aug 12 2006 splash.xpm.gz -rw-r--r-- 1 root root 512 Mar 8 17:45 stage1 -rw-r--r-- 1 root root 103688 Mar 8 17:45 stage2 -rw-r--r-- 1 root root 67701 Mar 8 18:30 symvers-2.6.9-89.35.1.EL.gz -rw-r--r-- 1 root root 68477 Mar 8 18:30 symvers-2.6.9-89.35.1.ELsmp.gz -rw-r--r-- 1 root root 7272 Mar 8 17:40 ufs2_stage1_5 -rw-r--r-- 1 root root 6612 Mar 8 17:40 vstafs_stage1_5 -rw-r--r-- 1 root root 9308 Mar 8 17:40 xfs_stage1_5
/boot/lost+found: total 22 drwx------ 2 root root 12288 Jan 12 2007 . drwxr-xr-x 5 root root 9216 Mar 9 10:31 ..
/boot/.Trash-root: total 19 drwx------ 2 root root 9216 Mar 9 10:03 . drwxr-xr-x 5 root root 9216 Mar 9 10:31 ..
On 3/9/2011 12:23 AM, Simon Matter wrote:
I inadvertently missed using the list...here are my recent messages.
As Nico suggested, download the kernel but also grub and redhat-logos, like so wget http://mirrors.kernel.org/centos/4.9/updates/i386/RPMS/kernel-2.6.9-100.EL.i... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/redhat-logos-1.1.26... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/grub-0.95-3.8.i386....
Then do a
rpm -Uvh --replacepkgs --replacefiles kernel-2.6.9-100.EL.i686.rpm redhat-logos-1.1.26-1.centos4.4.noarch.rpm grub-0.95-3.8.i386.rpm
And the show us the contents of 'ls -laR /boot' and 'cat /etc/grub.conf'
Simon
On 3/8/2011 8:39 PM, Nico Kadel-Garcia wrote:
On Tue, Mar 8, 2011 at 11:31 PM, Todd Carytodd@aristesoftware.com wrote:
reinstall is not an option for yum. I ran "yum install kernel" and it completed without errors however there are no links created.
Oh, dear. Can you grab the RPM and do "rpm -U -replacepkgs [kernel-whatver].rpm"? You should be able to use "yum remove" on the old kernel packages, consistent with freeing up the space, and now install your new kernel with yum.
Would this be the correct ln command for vmlinuz-2.6.9-89.35.1
# /boot/vmlinuz-2.6.9-89.35.1 /boot/vmlinuz
Todd
On 3/8/2011 7:04 PM, Nico Kadel-Garcia wrote:
On Tue, Mar 8, 2011 at 9:58 PM, Todd Carytodd@aristesoftware.com wrote:
I started a new thread since the original one is getting rather long.
I have retrieved the files I deleted in /boot and /boot/grub, however I need to make links for
/boot/System.map (System.map -> System.map-2.6.9-89.35.1) /boot/vmlinuz (vmlinuz -> vmlinuz-2.6.9-89.35.1) /boot/grub/menu.lst (menu.lst -> ./grub.conf)
Instead, re-install your kernel. "yum reinstall kernel". This should regenerate your symlinks correctly, except possibly the grub.conf.
If it was not so important to get it correct, I would appreciate the syntax for the command. Usually I would figure it out.
Since I have restored the files (I will double check to make sure they are all there), do I need to run grub-install?
i think yes. The old location of the boot loader is listed in /boot/grub/grub.conf, and should be used as the argument to that command. grub is much smarter than LILO used to be, but I think the bootstrap procedure relies on knowing details of where the fiddly bits of grub live on the relevant ex2 compatible filesytem.
My apologies for bothering everyone with such a dumb error on my part.
Todd
-- Ariste Software Petaluma, CA 94952
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- Ariste Software Petaluma, CA 94952
-- Ariste Software Petaluma, CA 94952
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
On 3/9/2011 12:35 PM, Todd Cary wrote:
Simon -
I performed the tasks as outlined and here are the contents of /boot:
It might be a good time to review what you can do with an install disk after booting with "linux rescue" at the prompt. If you've made a mistake in the setup you can get back in to fix it that way - including reinstalling grub and doing a kernel update.
And here are the contents of grub.conf:
# grub.conf generated by anaconda # # Note that you do not have to rerun grub after making changes to this file # NOTICE: You have a /boot partition. This means that # all kernel and initrd paths are relative to /boot/, eg. # root (hd0,0) # kernel /vmlinuz-version ro root=/dev/VolGroup00/LogVol00 # initrd /initrd-version.img #boot=/dev/hdc default=0 timeout=5 splashimage=(hd0,0)/grub/splash.xpm.gz hiddenmenu title CentOS (2.6.9-100.EL) root (hd0,0) kernel /vmlinuz-2.6.9-100.EL ro root=/dev/VolGroup00/LogVol00 rhgb quiet initrd /initrd-2.6.9-100.EL.img
Todd
On 3/9/2011 12:23 AM, Simon Matter wrote:
I inadvertently missed using the list...here are my recent messages.
As Nico suggested, download the kernel but also grub and redhat-logos, like so wget http://mirrors.kernel.org/centos/4.9/updates/i386/RPMS/kernel-2.6.9-100.EL.i... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/redhat-logos-1.1.26... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/grub-0.95-3.8.i386....
Then do a
rpm -Uvh --replacepkgs --replacefiles kernel-2.6.9-100.EL.i686.rpm redhat-logos-1.1.26-1.centos4.4.noarch.rpm grub-0.95-3.8.i386.rpm
And the show us the contents of 'ls -laR /boot' and 'cat /etc/grub.conf'
Simon
On 3/8/2011 8:39 PM, Nico Kadel-Garcia wrote:
On Tue, Mar 8, 2011 at 11:31 PM, Todd Carytodd@aristesoftware.com wrote:
reinstall is not an option for yum. I ran "yum install kernel" and it completed without errors however there are no links created.
Oh, dear. Can you grab the RPM and do "rpm -U -replacepkgs [kernel-whatver].rpm"? You should be able to use "yum remove" on the old kernel packages, consistent with freeing up the space, and now install your new kernel with yum.
Would this be the correct ln command for vmlinuz-2.6.9-89.35.1
# /boot/vmlinuz-2.6.9-89.35.1 /boot/vmlinuz
Todd
On 3/8/2011 7:04 PM, Nico Kadel-Garcia wrote:
On Tue, Mar 8, 2011 at 9:58 PM, Todd Carytodd@aristesoftware.com wrote:
I started a new thread since the original one is getting rather long.
I have retrieved the files I deleted in /boot and /boot/grub, however I need to make links for
/boot/System.map (System.map -> System.map-2.6.9-89.35.1) /boot/vmlinuz (vmlinuz -> vmlinuz-2.6.9-89.35.1) /boot/grub/menu.lst (menu.lst -> ./grub.conf)
Instead, re-install your kernel. "yum reinstall kernel". This should regenerate your symlinks correctly, except possibly the grub.conf.
If it was not so important to get it correct, I would appreciate the syntax for the command. Usually I would figure it out.
Since I have restored the files (I will double check to make sure they are all there), do I need to run grub-install?
i think yes. The old location of the boot loader is listed in /boot/grub/grub.conf, and should be used as the argument to that command. grub is much smarter than LILO used to be, but I think the bootstrap procedure relies on knowing details of where the fiddly bits of grub live on the relevant ex2 compatible filesytem.
My apologies for bothering everyone with such a dumb error on my part.
Todd
-- Ariste Software Petaluma, CA 94952
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- Ariste Software Petaluma, CA 94952
-- Ariste Software Petaluma, CA 94952
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
How goes the repair? Got it all worked out?
And here are the contents of grub.conf:
# grub.conf generated by anaconda # # Note that you do not have to rerun grub after making changes to this file # NOTICE: You have a /boot partition. This means that # all kernel and initrd paths are relative to /boot/, eg. # root (hd0,0) # kernel /vmlinuz-version ro root=/dev/VolGroup00/LogVol00 # initrd /initrd-version.img #boot=/dev/hdc default=0 timeout=5 splashimage=(hd0,0)/grub/splash.xpm.gz hiddenmenu title CentOS (2.6.9-100.EL) root (hd0,0) kernel /vmlinuz-2.6.9-100.EL ro root=/dev/VolGroup00/LogVol00 rhgb quiet initrd /initrd-2.6.9-100.EL.img
OK, the file listing of /boot from your last mail and now grub.conf, they look quite good. grub.conf has been updated by the kernel update, and also a new initrd-2.6.9-100.EL.img has been created, so that doesn't look bad.
The only thing I'm not really sure is if grub is installed correctly now. Maybe you have to run grub-install again to be sure but I'm just not so sure about grubs internals. Maybe someone can tell you more about this.
As someone else mentioned, it's a very good idea to have a current CentOS 4.8 disk at hand so you could boot into rescue mode with 'linux rescue' at the boot prompt if somethings goes wrong.
Simon
Todd
On 3/9/2011 12:23 AM, Simon Matter wrote:
I inadvertently missed using the list...here are my recent messages.
As Nico suggested, download the kernel but also grub and redhat-logos, like so wget http://mirrors.kernel.org/centos/4.9/updates/i386/RPMS/kernel-2.6.9-100.EL.i... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/redhat-logos-1.1.26... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/grub-0.95-3.8.i386....
Then do a
rpm -Uvh --replacepkgs --replacefiles kernel-2.6.9-100.EL.i686.rpm redhat-logos-1.1.26-1.centos4.4.noarch.rpm grub-0.95-3.8.i386.rpm
And the show us the contents of 'ls -laR /boot' and 'cat /etc/grub.conf'
Simon
On 3/8/2011 8:39 PM, Nico Kadel-Garcia wrote:
On Tue, Mar 8, 2011 at 11:31 PM, Todd Carytodd@aristesoftware.com wrote:
reinstall is not an option for yum. I ran "yum install kernel" and it completed without errors however there are no links created.
Oh, dear. Can you grab the RPM and do "rpm -U -replacepkgs [kernel-whatver].rpm"? You should be able to use "yum remove" on the old kernel packages, consistent with freeing up the space, and now install your new kernel with yum.
Would this be the correct ln command for vmlinuz-2.6.9-89.35.1
# /boot/vmlinuz-2.6.9-89.35.1 /boot/vmlinuz
Todd
On 3/8/2011 7:04 PM, Nico Kadel-Garcia wrote:
On Tue, Mar 8, 2011 at 9:58 PM, Todd Carytodd@aristesoftware.com wrote: > I started a new thread since the original one is getting rather > long. > > I have retrieved the files I deleted in /boot and /boot/grub, > however I need to make links for > > /boot/System.map (System.map -> System.map-2.6.9-89.35.1) > /boot/vmlinuz (vmlinuz -> vmlinuz-2.6.9-89.35.1) > /boot/grub/menu.lst (menu.lst -> ./grub.conf) Instead, re-install your kernel. "yum reinstall kernel". This should regenerate your symlinks correctly, except possibly the grub.conf.
> If it was not so important to get it correct, I would appreciate > the syntax for the command. Usually I would figure it out. > > Since I have restored the files (I will double check to make sure > they are all there), do I need to run grub-install? i think yes. The old location of the boot loader is listed in /boot/grub/grub.conf, and should be used as the argument to that command. grub is much smarter than LILO used to be, but I think the bootstrap procedure relies on knowing details of where the fiddly bits of grub live on the relevant ex2 compatible filesytem.
> My apologies for bothering everyone with such a dumb error on my > part. > > Todd > > -- > Ariste Software > Petaluma, CA 94952 > > http://www.aristesoftware.com > > _______________________________________________ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos >
-- Ariste Software Petaluma, CA 94952
-- Ariste Software Petaluma, CA 94952
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- Ariste Software Petaluma, CA 94952
Unfortunately, I live out with the cows, so I am using DSL to download the latest - it will take awhile. It has been awhile since I downloaded the four disks, however I assume disk 1 contains all that I need to do a "rescue".
Once I get that down, I will use torrent to get all four disks.
Hey, guys, many thanks. Any of you live in the SF Bay Area? Love to treat you to a beer.
Todd
On 3/9/2011 1:03 PM, Simon Matter wrote:
And here are the contents of grub.conf:
# grub.conf generated by anaconda # # Note that you do not have to rerun grub after making changes to this file # NOTICE: You have a /boot partition. This means that # all kernel and initrd paths are relative to /boot/, eg. # root (hd0,0) # kernel /vmlinuz-version ro root=/dev/VolGroup00/LogVol00 # initrd /initrd-version.img #boot=/dev/hdc default=0 timeout=5 splashimage=(hd0,0)/grub/splash.xpm.gz hiddenmenu title CentOS (2.6.9-100.EL) root (hd0,0) kernel /vmlinuz-2.6.9-100.EL ro root=/dev/VolGroup00/LogVol00 rhgb quiet initrd /initrd-2.6.9-100.EL.img
OK, the file listing of /boot from your last mail and now grub.conf, they look quite good. grub.conf has been updated by the kernel update, and also a new initrd-2.6.9-100.EL.img has been created, so that doesn't look bad.
The only thing I'm not really sure is if grub is installed correctly now. Maybe you have to run grub-install again to be sure but I'm just not so sure about grubs internals. Maybe someone can tell you more about this.
As someone else mentioned, it's a very good idea to have a current CentOS 4.8 disk at hand so you could boot into rescue mode with 'linux rescue' at the boot prompt if somethings goes wrong.
Simon
Todd
On 3/9/2011 12:23 AM, Simon Matter wrote:
I inadvertently missed using the list...here are my recent messages.
As Nico suggested, download the kernel but also grub and redhat-logos, like so wget http://mirrors.kernel.org/centos/4.9/updates/i386/RPMS/kernel-2.6.9-100.EL.i... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/redhat-logos-1.1.26... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/grub-0.95-3.8.i386....
Then do a
rpm -Uvh --replacepkgs --replacefiles kernel-2.6.9-100.EL.i686.rpm redhat-logos-1.1.26-1.centos4.4.noarch.rpm grub-0.95-3.8.i386.rpm
And the show us the contents of 'ls -laR /boot' and 'cat /etc/grub.conf'
Simon
On 3/8/2011 8:39 PM, Nico Kadel-Garcia wrote:
On Tue, Mar 8, 2011 at 11:31 PM, Todd Carytodd@aristesoftware.com wrote:
reinstall is not an option for yum. I ran "yum install kernel" and it completed without errors however there are no links created.
Oh, dear. Can you grab the RPM and do "rpm -U -replacepkgs [kernel-whatver].rpm"? You should be able to use "yum remove" on the old kernel packages, consistent with freeing up the space, and now install your new kernel with yum.
Would this be the correct ln command for vmlinuz-2.6.9-89.35.1
# /boot/vmlinuz-2.6.9-89.35.1 /boot/vmlinuz
Todd
On 3/8/2011 7:04 PM, Nico Kadel-Garcia wrote: > On Tue, Mar 8, 2011 at 9:58 PM, Todd Carytodd@aristesoftware.com > wrote: >> I started a new thread since the original one is getting rather >> long. >> >> I have retrieved the files I deleted in /boot and /boot/grub, >> however I need to make links for >> >> /boot/System.map (System.map -> System.map-2.6.9-89.35.1) >> /boot/vmlinuz (vmlinuz -> vmlinuz-2.6.9-89.35.1) >> /boot/grub/menu.lst (menu.lst -> ./grub.conf) > Instead, re-install your kernel. "yum reinstall kernel". This should > regenerate your symlinks correctly, except possibly the grub.conf. > >> If it was not so important to get it correct, I would appreciate >> the syntax for the command. Usually I would figure it out. >> >> Since I have restored the files (I will double check to make sure >> they are all there), do I need to run grub-install? > i think yes. The old location of the boot loader is listed in > /boot/grub/grub.conf, and should be used as the argument to that > command. grub is much smarter than LILO used to be, but I think the > bootstrap procedure relies on knowing details of where the fiddly > bits > of grub live on the relevant ex2 compatible filesytem. > >> My apologies for bothering everyone with such a dumb error on my >> part. >> >> Todd >> >> -- >> Ariste Software >> Petaluma, CA 94952 >> >> http://www.aristesoftware.com >> >> _______________________________________________ >> CentOS mailing list >> CentOS@centos.org >> http://lists.centos.org/mailman/listinfo/centos
>>
Ariste Software Petaluma, CA 94952
-- Ariste Software Petaluma, CA 94952
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- Ariste Software Petaluma, CA 94952
Unfortunately, I live out with the cows, so I am using DSL to download the latest - it will take awhile. It has been awhile since I downloaded the four disks, however I assume disk 1 contains all that I need to do a "rescue".
Yes that's correct, you need to download only disk 1.
Once I get that down, I will use torrent to get all four disks.
Hey, guys, many thanks. Any of you live in the SF Bay Area? Love to treat you to a beer.
Thanks, but it may be a bit difficult. Just let us know if you have been able to boot successfully.
Simon
Todd
On 3/9/2011 1:03 PM, Simon Matter wrote:
And here are the contents of grub.conf:
# grub.conf generated by anaconda # # Note that you do not have to rerun grub after making changes to this file # NOTICE: You have a /boot partition. This means that # all kernel and initrd paths are relative to /boot/, eg. # root (hd0,0) # kernel /vmlinuz-version ro root=/dev/VolGroup00/LogVol00 # initrd /initrd-version.img #boot=/dev/hdc default=0 timeout=5 splashimage=(hd0,0)/grub/splash.xpm.gz hiddenmenu title CentOS (2.6.9-100.EL) root (hd0,0) kernel /vmlinuz-2.6.9-100.EL ro root=/dev/VolGroup00/LogVol00 rhgb quiet initrd /initrd-2.6.9-100.EL.img
OK, the file listing of /boot from your last mail and now grub.conf, they look quite good. grub.conf has been updated by the kernel update, and also a new initrd-2.6.9-100.EL.img has been created, so that doesn't look bad.
The only thing I'm not really sure is if grub is installed correctly now. Maybe you have to run grub-install again to be sure but I'm just not so sure about grubs internals. Maybe someone can tell you more about this.
As someone else mentioned, it's a very good idea to have a current CentOS 4.8 disk at hand so you could boot into rescue mode with 'linux rescue' at the boot prompt if somethings goes wrong.
Simon
Todd
On 3/9/2011 12:23 AM, Simon Matter wrote:
I inadvertently missed using the list...here are my recent messages.
As Nico suggested, download the kernel but also grub and redhat-logos, like so wget http://mirrors.kernel.org/centos/4.9/updates/i386/RPMS/kernel-2.6.9-100.EL.i... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/redhat-logos-1.1.26... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/grub-0.95-3.8.i386....
Then do a
rpm -Uvh --replacepkgs --replacefiles kernel-2.6.9-100.EL.i686.rpm redhat-logos-1.1.26-1.centos4.4.noarch.rpm grub-0.95-3.8.i386.rpm
And the show us the contents of 'ls -laR /boot' and 'cat /etc/grub.conf'
Simon
On 3/8/2011 8:39 PM, Nico Kadel-Garcia wrote:
On Tue, Mar 8, 2011 at 11:31 PM, Todd Carytodd@aristesoftware.com wrote: > reinstall is not an option for yum. I ran "yum install kernel" and > it > completed without errors however there are no links created. Oh, dear. Can you grab the RPM and do "rpm -U -replacepkgs [kernel-whatver].rpm"? You should be able to use "yum remove" on the old kernel packages, consistent with freeing up the space, and now install your new kernel with yum.
> Would this be the correct ln command for vmlinuz-2.6.9-89.35.1 > > # /boot/vmlinuz-2.6.9-89.35.1 /boot/vmlinuz > > Todd > > On 3/8/2011 7:04 PM, Nico Kadel-Garcia wrote: >> On Tue, Mar 8, 2011 at 9:58 PM, Todd Carytodd@aristesoftware.com >> wrote: >>> I started a new thread since the original one is getting rather >>> long. >>> >>> I have retrieved the files I deleted in /boot and /boot/grub, >>> however I need to make links for >>> >>> /boot/System.map (System.map -> System.map-2.6.9-89.35.1) >>> /boot/vmlinuz (vmlinuz -> vmlinuz-2.6.9-89.35.1) >>> /boot/grub/menu.lst (menu.lst -> ./grub.conf) >> Instead, re-install your kernel. "yum reinstall kernel". This >> should >> regenerate your symlinks correctly, except possibly the grub.conf. >> >>> If it was not so important to get it correct, I would appreciate >>> the syntax for the command. Usually I would figure it out. >>> >>> Since I have restored the files (I will double check to make sure >>> they are all there), do I need to run grub-install? >> i think yes. The old location of the boot loader is listed in >> /boot/grub/grub.conf, and should be used as the argument to that >> command. grub is much smarter than LILO used to be, but I think >> the >> bootstrap procedure relies on knowing details of where the fiddly >> bits >> of grub live on the relevant ex2 compatible filesytem. >> >>> My apologies for bothering everyone with such a dumb error on my >>> part. >>> >>> Todd >>> >>> -- >>> Ariste Software >>> Petaluma, CA 94952 >>> >>> http://www.aristesoftware.com >>> >>> _______________________________________________ >>> CentOS mailing list >>> CentOS@centos.org >>> http://lists.centos.org/mailman/listinfo/centos >>> > -- > Ariste Software > Petaluma, CA 94952 > > http://www.aristesoftware.com > >
-- Ariste Software Petaluma, CA 94952
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- Ariste Software Petaluma, CA 94952
-- Ariste Software Petaluma, CA 94952
Well, I have the disks in hand - all 4, but there is the overriding level of apprehension. Is there a reference to what I should do *if* I cannot reboot that I should read?
Also, after reading the responses to my query about using FAT32 to store data, I decided to follow the suggestions to use Samba to copy the data that needs to be shared with others to my Windows connected NTFS external. So, I am in the process of doing that first even though I have an ext3 backup of the same data.
Preparing to jump as I look for my spare parachute....
Todd
On 3/10/2011 2:15 AM, Simon Matter wrote:
Unfortunately, I live out with the cows, so I am using DSL to download the latest - it will take awhile. It has been awhile since I downloaded the four disks, however I assume disk 1 contains all that I need to do a "rescue".
Yes that's correct, you need to download only disk 1.
Once I get that down, I will use torrent to get all four disks.
Hey, guys, many thanks. Any of you live in the SF Bay Area? Love to treat you to a beer.
Thanks, but it may be a bit difficult. Just let us know if you have been able to boot successfully.
Simon
Todd
On 3/9/2011 1:03 PM, Simon Matter wrote:
And here are the contents of grub.conf:
# grub.conf generated by anaconda # # Note that you do not have to rerun grub after making changes to this file # NOTICE: You have a /boot partition. This means that # all kernel and initrd paths are relative to /boot/, eg. # root (hd0,0) # kernel /vmlinuz-version ro root=/dev/VolGroup00/LogVol00 # initrd /initrd-version.img #boot=/dev/hdc default=0 timeout=5 splashimage=(hd0,0)/grub/splash.xpm.gz hiddenmenu title CentOS (2.6.9-100.EL) root (hd0,0) kernel /vmlinuz-2.6.9-100.EL ro root=/dev/VolGroup00/LogVol00 rhgb quiet initrd /initrd-2.6.9-100.EL.img
OK, the file listing of /boot from your last mail and now grub.conf, they look quite good. grub.conf has been updated by the kernel update, and also a new initrd-2.6.9-100.EL.img has been created, so that doesn't look bad.
The only thing I'm not really sure is if grub is installed correctly now. Maybe you have to run grub-install again to be sure but I'm just not so sure about grubs internals. Maybe someone can tell you more about this.
As someone else mentioned, it's a very good idea to have a current CentOS 4.8 disk at hand so you could boot into rescue mode with 'linux rescue' at the boot prompt if somethings goes wrong.
Simon
Todd
On 3/9/2011 12:23 AM, Simon Matter wrote:
I inadvertently missed using the list...here are my recent messages.
As Nico suggested, download the kernel but also grub and redhat-logos, like so wget http://mirrors.kernel.org/centos/4.9/updates/i386/RPMS/kernel-2.6.9-100.EL.i... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/redhat-logos-1.1.26... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/grub-0.95-3.8.i386....
Then do a
rpm -Uvh --replacepkgs --replacefiles kernel-2.6.9-100.EL.i686.rpm redhat-logos-1.1.26-1.centos4.4.noarch.rpm grub-0.95-3.8.i386.rpm
And the show us the contents of 'ls -laR /boot' and 'cat /etc/grub.conf'
Simon
On 3/8/2011 8:39 PM, Nico Kadel-Garcia wrote: > On Tue, Mar 8, 2011 at 11:31 PM, Todd Carytodd@aristesoftware.com > wrote: >> reinstall is not an option for yum. I ran "yum install kernel" and >> it >> completed without errors however there are no links created. > Oh, dear. Can you grab the RPM and do "rpm -U -replacepkgs > [kernel-whatver].rpm"? You should be able to use "yum remove" on the > old kernel packages, consistent with freeing up the space, and now > install your new kernel with yum. > >> Would this be the correct ln command for vmlinuz-2.6.9-89.35.1 >> >> # /boot/vmlinuz-2.6.9-89.35.1 /boot/vmlinuz >> >> Todd >> >> On 3/8/2011 7:04 PM, Nico Kadel-Garcia wrote: >>> On Tue, Mar 8, 2011 at 9:58 PM, Todd Carytodd@aristesoftware.com >>> wrote: >>>> I started a new thread since the original one is getting rather >>>> long. >>>> >>>> I have retrieved the files I deleted in /boot and /boot/grub, >>>> however I need to make links for >>>> >>>> /boot/System.map (System.map -> System.map-2.6.9-89.35.1) >>>> /boot/vmlinuz (vmlinuz -> vmlinuz-2.6.9-89.35.1) >>>> /boot/grub/menu.lst (menu.lst -> ./grub.conf) >>> Instead, re-install your kernel. "yum reinstall kernel". This >>> should >>> regenerate your symlinks correctly, except possibly the grub.conf. >>> >>>> If it was not so important to get it correct, I would appreciate >>>> the syntax for the command. Usually I would figure it out. >>>> >>>> Since I have restored the files (I will double check to make sure >>>> they are all there), do I need to run grub-install? >>> i think yes. The old location of the boot loader is listed in >>> /boot/grub/grub.conf, and should be used as the argument to that >>> command. grub is much smarter than LILO used to be, but I think >>> the >>> bootstrap procedure relies on knowing details of where the fiddly >>> bits >>> of grub live on the relevant ex2 compatible filesytem. >>> >>>> My apologies for bothering everyone with such a dumb error on my >>>> part. >>>> >>>> Todd >>>> >>>> -- >>>> Ariste Software >>>> Petaluma, CA 94952 >>>> >>>> http://www.aristesoftware.com >>>> >>>> _______________________________________________ >>>> CentOS mailing list >>>> CentOS@centos.org >>>> http://lists.centos.org/mailman/listinfo/centos >>>> >> -- >> Ariste Software >> Petaluma, CA 94952 >> >> http://www.aristesoftware.com >>
>>
Ariste Software Petaluma, CA 94952
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- Ariste Software Petaluma, CA 94952
-- Ariste Software Petaluma, CA 94952
Well, I have the disks in hand - all 4, but there is the overriding level of apprehension. Is there a reference to what I should do *if* I cannot reboot that I should read?
As I said before, you may need to run grub-install, but I don't know for sure. And then, you have to know where to install grub, and I don't know where you have installed it. In fact I don't know how we could know because it really depends on how your BIOS boots the box. It can be that it's installed into the MBR of /dev/hdc, then you should be able to install it using 'grub-install /dev/hdc'. But, since the disk is named /dev/hdc, it's most likely that there is also /dev/hda and /dev/hdb, and then it's also likely that grub has been installed into the MBR of /dev/hda. Who knows?
That said, check disk 1 by putting it into another computer, and chose 'linux rescue' at the boot prompt. Then it will boot using a root filesystem in ram, and configure network if you want and then tries to find any CentOS installation in the disks, and mount them if it finds one. Maybe it wont find one but it should find it on your server. Then it will mount it as something like /mnt/sysimage. You can then 'chroot /mnt/sysimage' and fix things.
Good luck! Simon
Also, after reading the responses to my query about using FAT32 to store data, I decided to follow the suggestions to use Samba to copy the data that needs to be shared with others to my Windows connected NTFS external. So, I am in the process of doing that first even though I have an ext3 backup of the same data.
Preparing to jump as I look for my spare parachute....
Todd
On 3/10/2011 2:15 AM, Simon Matter wrote:
Unfortunately, I live out with the cows, so I am using DSL to download the latest - it will take awhile. It has been awhile since I downloaded the four disks, however I assume disk 1 contains all that I need to do a "rescue".
Yes that's correct, you need to download only disk 1.
Once I get that down, I will use torrent to get all four disks.
Hey, guys, many thanks. Any of you live in the SF Bay Area? Love to treat you to a beer.
Thanks, but it may be a bit difficult. Just let us know if you have been able to boot successfully.
Simon
Todd
On 3/9/2011 1:03 PM, Simon Matter wrote:
And here are the contents of grub.conf:
# grub.conf generated by anaconda # # Note that you do not have to rerun grub after making changes to this file # NOTICE: You have a /boot partition. This means that # all kernel and initrd paths are relative to /boot/, eg. # root (hd0,0) # kernel /vmlinuz-version ro root=/dev/VolGroup00/LogVol00 # initrd /initrd-version.img #boot=/dev/hdc default=0 timeout=5 splashimage=(hd0,0)/grub/splash.xpm.gz hiddenmenu title CentOS (2.6.9-100.EL) root (hd0,0) kernel /vmlinuz-2.6.9-100.EL ro root=/dev/VolGroup00/LogVol00 rhgb quiet initrd /initrd-2.6.9-100.EL.img
OK, the file listing of /boot from your last mail and now grub.conf, they look quite good. grub.conf has been updated by the kernel update, and also a new initrd-2.6.9-100.EL.img has been created, so that doesn't look bad.
The only thing I'm not really sure is if grub is installed correctly now. Maybe you have to run grub-install again to be sure but I'm just not so sure about grubs internals. Maybe someone can tell you more about this.
As someone else mentioned, it's a very good idea to have a current CentOS 4.8 disk at hand so you could boot into rescue mode with 'linux rescue' at the boot prompt if somethings goes wrong.
Simon
Todd
On 3/9/2011 12:23 AM, Simon Matter wrote:
> I inadvertently missed using the list...here are my recent > messages. As Nico suggested, download the kernel but also grub and redhat-logos, like so wget http://mirrors.kernel.org/centos/4.9/updates/i386/RPMS/kernel-2.6.9-100.EL.i... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/redhat-logos-1.1.26... wget http://mirrors.kernel.org/centos/4.9/os/i386/CentOS/RPMS/grub-0.95-3.8.i386....
Then do a
rpm -Uvh --replacepkgs --replacefiles kernel-2.6.9-100.EL.i686.rpm redhat-logos-1.1.26-1.centos4.4.noarch.rpm grub-0.95-3.8.i386.rpm
And the show us the contents of 'ls -laR /boot' and 'cat /etc/grub.conf'
Simon
> On 3/8/2011 8:39 PM, Nico Kadel-Garcia wrote: >> On Tue, Mar 8, 2011 at 11:31 PM, Todd >> Carytodd@aristesoftware.com >> wrote: >>> reinstall is not an option for yum. I ran "yum install kernel" >>> and >>> it >>> completed without errors however there are no links created. >> Oh, dear. Can you grab the RPM and do "rpm -U -replacepkgs >> [kernel-whatver].rpm"? You should be able to use "yum remove" on >> the >> old kernel packages, consistent with freeing up the space, and now >> install your new kernel with yum. >> >>> Would this be the correct ln command for vmlinuz-2.6.9-89.35.1 >>> >>> # /boot/vmlinuz-2.6.9-89.35.1 /boot/vmlinuz >>> >>> Todd >>> >>> On 3/8/2011 7:04 PM, Nico Kadel-Garcia wrote: >>>> On Tue, Mar 8, 2011 at 9:58 PM, Todd >>>> Carytodd@aristesoftware.com >>>> wrote: >>>>> I started a new thread since the original one is getting rather >>>>> long. >>>>> >>>>> I have retrieved the files I deleted in /boot and /boot/grub, >>>>> however I need to make links for >>>>> >>>>> /boot/System.map (System.map -> >>>>> System.map-2.6.9-89.35.1) >>>>> /boot/vmlinuz (vmlinuz -> vmlinuz-2.6.9-89.35.1) >>>>> /boot/grub/menu.lst (menu.lst -> ./grub.conf) >>>> Instead, re-install your kernel. "yum reinstall kernel". This >>>> should >>>> regenerate your symlinks correctly, except possibly the >>>> grub.conf. >>>> >>>>> If it was not so important to get it correct, I would >>>>> appreciate >>>>> the syntax for the command. Usually I would figure it out. >>>>> >>>>> Since I have restored the files (I will double check to make >>>>> sure >>>>> they are all there), do I need to run grub-install? >>>> i think yes. The old location of the boot loader is listed in >>>> /boot/grub/grub.conf, and should be used as the argument to that >>>> command. grub is much smarter than LILO used to be, but I think >>>> the >>>> bootstrap procedure relies on knowing details of where the >>>> fiddly >>>> bits >>>> of grub live on the relevant ex2 compatible filesytem. >>>> >>>>> My apologies for bothering everyone with such a dumb error on >>>>> my >>>>> part. >>>>> >>>>> Todd >>>>> >>>>> -- >>>>> Ariste Software >>>>> Petaluma, CA 94952 >>>>> >>>>> http://www.aristesoftware.com >>>>> >>>>> _______________________________________________ >>>>> CentOS mailing list >>>>> CentOS@centos.org >>>>> http://lists.centos.org/mailman/listinfo/centos >>>>> >>> -- >>> Ariste Software >>> Petaluma, CA 94952 >>> >>> http://www.aristesoftware.com >>> >>> > -- > Ariste Software > Petaluma, CA 94952 > > http://www.aristesoftware.com > > _______________________________________________ > CentOS mailing list > CentOS@centos.org > http://lists.centos.org/mailman/listinfo/centos > _______________________________________________ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
-- Ariste Software Petaluma, CA 94952
-- Ariste Software Petaluma, CA 94952
-- Ariste Software Petaluma, CA 94952
On Thu, Mar 10, 2011 at 10:43 AM, Simon Matter simon.matter@invoca.ch wrote:
Well, I have the disks in hand - all 4, but there is the overriding level of apprehension. Is there a reference to what I should do *if* I cannot reboot that I should read?
As I said before, you may need to run grub-install, but I don't know for sure. And then, you have to know where to install grub, and I don't know where you have installed it. In fact I don't know how we could know
This information is normally stored, commented out, in /boot/grub/grub.conf.
because it really depends on how your BIOS boots the box. It can be that it's installed into the MBR of /dev/hdc, then you should be able to install it using 'grub-install /dev/hdc'. But, since the disk is named /dev/hdc, it's most likely that there is also /dev/hda and /dev/hdb, and then it's also likely that grub has been installed into the MBR of /dev/hda. Who knows?
Don't assume this. If the PATA cable is plugged into a second PATA controller port, and nothing or a CD drive on the first controller port, it would explain how he wound up with /dev/hdc has his hard drive.
That said, check disk 1 by putting it into another computer, and chose 'linux rescue' at the boot prompt. Then it will boot using a root filesystem in ram, and configure network if you want and then tries to find any CentOS installation in the disks, and mount them if it finds one. Maybe it wont find one but it should find it on your server. Then it will mount it as something like /mnt/sysimage. You can then 'chroot /mnt/sysimage' and fix things.
Yeah, I love the live CD's for this as well.
Good luck! Simon
-
compdoc -
Les Mikesell -
Nico Kadel-Garcia -
Simon Matter -
Todd Cary