Hi all,
I am a newbie to linux and this is my first mail to this group. We have a PC running Centos version 2.6.9-42.0.2.EL. I needed to load a router software called Nistnet and it would not run properly. It is looking for RTC (Real Time clock) support as a module.
I 've gone through the LKM Loadable Kernel module support under Documentation pages of Linux.org and did the following:
located the kernel release: uname -r 2.6.9-42.0.2.EL
next navigated to the /usr/src/kernels/2.6.9-42.0.2.EL-i686 (this is where the kernel is located on this PC) then issued a make menuconfig command
It took me to through the menu and under device drivers> character device drivers i chose the Real time clock support as a module.
Then exited out of the menu and did make modules_install. Then rebooted.
However, the PC while rebooting showed some messages like fast_rtc: lost around 26 interrupts at 8192Hz.
this keeps on appearing on the screen for a while before reboot and after the system reboots, I see that at boot time, the ethernet interface does not come up. It gives the follwoing error: Device eth0 has a different MAC address than expected, ignoring. Also i get the following error messages :
Cannot open /lib/modules/2.6.9-42.0.2.EL/kernel/arch/i386/kernel/ microcode.ko no such file or directory and cannot open lib/modules/2.6.9-42.0.2.EL/kernel/drivers/net/sunrpc/sunrpc.ko no such file or directory.
if I do an updatedb command and then slocate microcode.ko, I still see it under this directory. However, if I navigate to the kernel directory under 2.6.9-42.0.2.EL , there is nothing under it. If i issue ls -al there are a totoal of 16 entries but I cant see them.
Can any one please tell me how to revert back to the original configuration or to undo these changes? I have copied the config-2.6.9-42.0.-EL release from Centos site to the /boot thinking it would revert to the original default config. Then I rebooted the PC but I still get the error-Eth interface has a different MAC id than expected or unable to bring eth0 up.
So, at the boot time, I am currently choosing the kernel version 2.6.9-32-EL. With this I am able to browse the Internet. But I want to make the version 2.6.9-42-EL up. Please some one help me!!
Thanks, Padmaja
----- Original Message ----- From: "nate" centos@linuxpowered.net To: centos@centos.org Sent: Thursday, February 07, 2008 7:00 PM Subject: Re: [CentOS] Reg. Kernel switching
Padmaja wrote:
Can any one please tell me how to revert back to the original configuration or to undo these changes?
Re-install the original kernel RPM. You may have to --force it to overwrite the existing since it's already installed.
nate
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Can you please provide the instructions/ commands to re install the original kernel. Also please let me know if anything went wrong with this re-installation of 2.6.9-42 kernel, will it affect the other kernel from working or I can still continue to use the 2.6.9.34 kernel?
Thanks, Padmaja ----- Original Message ----- From: "nate" centos@linuxpowered.net To: centos@centos.org Sent: Thursday, February 07, 2008 7:00 PM Subject: Re: [CentOS] Reg. Kernel switching
Padmaja wrote:
Can any one please tell me how to revert back to the original configuration or to undo these changes?
Re-install the original kernel RPM. You may have to --force it to overwrite the existing since it's already installed.
nate
CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Padmaja wrote:
Can you please provide the instructions/ commands to re install the original kernel. Also please let me know if anything went wrong with this re-installation of 2.6.9-42 kernel, will it affect the other kernel from working or I can still continue to use the 2.6.9.34 kernel?
find the original RPM for the kernel you want and run
rpm -Uvh <rpm name>
if it fails saying it's already installed run:
rpm -Uvh --force <rpm name>
if it fails saying something newer is already installed run:
rpm -Uvh --oldpackage <rpm name>
The installation of 2.6.9-42 will likely remove the 2.6.9-34 kernel, I *think* you can get around that by using rpm -iv instead of -Uvh, that should install the new kernel along side the old. But you should keep a backup of the other kernel just in case(or track down the original RPM for that as well).
nate
On Fri, Feb 08, 2008 at 06:59:36AM -0800, nate alleged:
Padmaja wrote:
Can you please provide the instructions/ commands to re install the original kernel. Also please let me know if anything went wrong with this re-installation of 2.6.9-42 kernel, will it affect the other kernel from working or I can still continue to use the 2.6.9.34 kernel?
find the original RPM for the kernel you want and run
rpm -Uvh <rpm name>
if it fails saying it's already installed run:
rpm -Uvh --force <rpm name>
if it fails saying something newer is already installed run:
rpm -Uvh --oldpackage <rpm name>
The installation of 2.6.9-42 will likely remove the 2.6.9-34 kernel, I *think* you can get around that by using rpm -iv instead of -Uvh, that should install the new kernel along side the old. But you should keep a backup of the other kernel just in case(or track down the original RPM for that as well).
No. Don't *upgrade* kernels, and use yum.
Get the full package name # rpm -qa | grep kernel
Remove the one you want # rpm -e <pkgname>
Update your system # yum update
Garrick Staples wrote:
No. Don't *upgrade* kernels, and use yum.
Get the full package name # rpm -qa | grep kernel
Remove the one you want # rpm -e <pkgname>
Update your system # yum update
yeah sorry :) I don't use yum on my main systems, everything is through cfengine or manual. Haven't found a good way to use yum in my environments.
nate