Broke dynamic linker! /lib/libc.so.6 symbol _dl_out_of_memory GLIBC
 

I installed an RPM which I though was glibc2.4 but it was actually a patch for glibc2.4 and now the system hangs at runlevel 5. I'm hoping someone can help me out to resolve this. The distro is SUSE 9 SP4 and it's a remote server that we can only access through KVM. Here is a screen shot of the problem during boot up...

http://www.mtprogramming.com/screenshot.jpg

Some suggestions from the internet are to replace the /lib and /usr/lib directories with the same directories from another machine running SUSE 9 SP4? Is this a good idea? Or is there a simpler way? I can get to the machine and reboot it manually but if it hangs at runlevel 5 how would I get around this so that I can login and fix the problem?

Your help is greatly appreciated!

If you can't reboot, there is only one way : switch off
the computer.
Extract the 'glibc' package, you want to use, (glibc-2.3.2 ?)
on another machine : 'rpm2cpio <package.rpm> | cpio -idmv'
> > in a new empty directory.
And now you have the 4 directories etc, lib, sbin, usr,
containing the glibc files.
'su' .. # 'chmod -R 0755 *' .. 'chown -R root:root *'
Next : # 'tar -cf glibc-2.x.x.tar etc lib sbin usr' ..
..# 'gzip glibc-2.x.x.tar' and copy the result to a usb - key.

Boot into a live cd Linux on the computer to be "repaired",
cp the glibc-2.x.x.tar.gz package to the "live Linux" ...
.. 'cd /media/<Suse9-location> .. 'su' ..
'tar xvf <path-to-glibc-package>/glibc-2.x.x.tar.gz' .

And then ... just hope, that the Suse 9 will be usable again.
......

I tried to boot up in fail safe and still no luck...

Please read post # 2 again, I changed a few details.

And you will not be able to boot into a system with
a faulty 'glibc'. Failsafe or no failsafe.

.....

OK thanks alot... I'll try that.