Kernel Panic - Unable to Boot RHEL4 *******glibc*******8
 

When iam booting into my system it is giving the following message

at Boot after GRUB.....
XXXXXXXXXXXxxx
XXXXXXXXXXXXXXxx
XXXXXXXXXXXXXXXXXXXxxx

...........................blah
...........................blah
Uncompressing Linux...Ok, booting the kernel.
Red hat nash version ....... blah blah
Reading all Physical volumes.........blah blah
2 logical volume(s) in volume group "VolGroup00" now active
/sbin/init: /lib/tls/libc.so.6: version GLIBC_2.3.4 not found (required by /lib/libselinux.so.1)
kernel panic - not syncing: Attempted to kill init!

and my keyboard lights of Caps lock and Scroll lock are glowing on and off

please give the solution how i should get back my system booting.

(Note: Previously i tried updating oldpackage the old Glibc-2.2.xx rpm without deps and ...so how i can get back install the original even i unable to boot into my system)

After this i tried to mount this hard disk on another system of mine which is installed Fedora core 5 but iam unable to mount my hard disks.

please help me out to get rid of this

iam very thank full for you

lot of important data is there

this is the lesson for me ..........................

You changed the glibc rpm packages on the system??? This is one good way to destroy a working installation. The system cannot boot because you did this, you have the wrong glibc version installed for selinux to work. Not only is selinux messed up but a lot of other packages as well.

Recovery may be possible but you need to first restore the glibc rpm packages to what should be installed on the system. You can try booting the system with the installation media into rescue mode answer a few questions and use chroot as directed by the on-screen messages. To gain access to the installation media to hopefully restore the glibc rpm packages try the following;

# mknod /tmp/tmpcdrom b 22 0
# mount -t iso9660 /tmp/tmpcdrom /cdrom

thanks
 

Thanks for the reply

before your reply i tried to use the command

I run the Rescue mode from there iam able to see the all the partitions at sysimage

and tried to issue this command

chroot /mnt/sysimage


saying that it needs again the same GLIBC .... and coming back

so when i dont get the solution and iam in hurry to recover my data just i invoke the ftp to another system
then i tar my files and folders from there all these ftp to another my system

after this i formatted the drive and installed new

why i done this is when i tried to mount or issuing the chroot /mnt/sysimage it is asking the glibc ........


another i tried is i have the puppy live cd so i run this on that and copied all the glibc rpm's to the boot partition of my system and then coming to rescue mode

i typed the following commands

i.e first i got the prompt then

cd mnt
cd sysimage
cd boot


here i can see all those rpm's from here i issued the commands

rpm -ivh glib*

but it is giving the errors

then i tried

rpm -Uvh --nodeps glib*

even now also i got the error

-------------------------------------------

why i compiled the glib rpm because

to install Oracle 9i in Redhat 9 we have to update the old glibc-2.2.5 ....... so i done the same on RHEL 4 this made me this big trouble so take care friends before applying trick of one to another

bye


this makes me to ftp my files and formatting the system

thanks for the reply

i hope some more information which may help for us who are having this problem

Well you cannot install the glibc rpm packages with a broken glibc installation because rpm relies on glibc (which is broken). If you can access the system from another good RHEL4 installation you might be able to ftp the correct glibc files to the /lib and /lib64 (if 64-bit) directory location(s). You might be able to use cpio to install the correct glibc files to the proper locations also. I'm not 100% sure about either method.

Check this out for Oracle 9i installation help, CentOS4 is a 100% binary rpm compatible 'clone' of RHEL4;

http://www.centos.org/modules/newbb/...m.php?forum=35

I know this is a bit old but I found this post while researching a similar issue and want to add the following for the record...

When using the mknod command from above, 22 may not be the major number for your CDROM. Check on a similar system to see:

$ ls -l /dev/cdrom
lrwxrwxrwx 1 root root 3 Aug 2 15:54 /dev/cdrom -> hda
$ ls -l /dev/hda
brw-rw---- 1 root disk 3, 0 Aug 2 15:54 /dev/hda

In the case of my system, the major number is 3 so my command would read:

# mknod /tmp/tmpcdrom b 3 0

Once booted into rescue mode you do not have to chroot to install packages. In the case of glibc, you can use the following to install it once the media is mounted:

rpm -ivh --force --root /mnt/sysimage /tmp/tmpcdrom/RedHat/RPMS/glibc-x.x.x-whatever.rpm

and (if needed):

rpm -ivh --force --root /mnt/sysimage /tmp/tmpcdrom/RedHat/RPMS/glibc-common-x.x.x-whatever.rpm

Hope this helps someone in the future.