FedoraThis forum is for the discussion of the Fedora Project.
Notices
Welcome to LinuxQuestions.org, a friendly and active Linux Community.
You are currently viewing LQ as a guest. By joining our community you will have the ability to post topics, receive our newsletter, use the advanced search, subscribe to threads and access many other special features. Registration is quick, simple and absolutely free. Join our community today!
Note that registered members see fewer ads, and ContentLink is completely disabled once you log in.
If you have any problems with the registration process or your account login, please contact us. If you need to reset your password, click here.
Having a problem logging in? Please visit this page to clear all LQ-related cookies.
Get a virtual cloud desktop with the Linux distro that you want in less than five minutes with Shells! With over 10 pre-installed distros to choose from, the worry-free installation life is here! Whether you are a digital nomad or just looking for flexibility, Shells can put your Linux machine on the device that you want to use.
Exclusive for LQ members, get up to 45% off per month. Click here for more info.
I installed FC6 on a Dell PE2950 server, dual core, dual CPU 8G ram etc. After installing it ran fine for at least 2 weeks and I needed to move it to a different workbench, so I shut it down and moved it. Now I get a No IRQ Handler for vector error, and I have to pull the plug. I believe it is happening at the network level as it doesnt do it if my nics are unplugged.
Anyway, I believe this is kernel related as a few searches have turned up similiar issues running 2.6.19 kernel in FC6. So how can I downgrade the kernel in this without network connectivity, and without rebuilding the system?
When the system boots and you get the count-down screen, hit enter. You'll be presented with the list of installed kernels. Select the one you want and hit enter.
How did you upgrade the kernel? If you used yum or just rpm to install a new kernel .rpm package, chances are you do have the old kernel there too -- check your /boot for what files reside there. Basically because the kernel can be installed from a .rpm binary package, you could just find your preferred version of the kernel from the web, download it (and it's dependencies -- correct versions of them, note) and install them using rpm. Read
Code:
man rpm
I guess that explains which command line options you use to downgrade and possibly force downgrading. Because rpm doesn't like you having two different versions of the same package installed simultaneously, you may have to do some tricks to downgrade (not difficult), but it doesn't prevent you from having multiple kernels around. Generally when upgrading a kernel you should never remove the old kernel before you're 100% sure (or even then) that the new one works and you don't need the old one anymore. That's why, when you upgrade your kernel in a binary distribution like Fedora, it doesn't (read: should not) usually remove the old kernel files.
If you happen to have the old kernel files in your /boot you can simply use your bootloader (GRUB nowadays I guess) to boot one of the older ones. And if you like that permanent, configure your bootloader to do so. There are sites on the web with more precise information about this (kernel.org too), and what it means when you upgrade a kernel or have multiple kernels which you want to use, so go ahead -- search and read. My advice (though this should have been heard before, and not after, this like usually) is that if you're doing a kernel upgrade -- especially on a server, and especially especially on an "important" "work" "server", first test the new kernel on an identical test environment for a long enough time (how long is that? ) before doing the upgrade on the work machine, and when you do, always have the old kernel there for backup.
This was easier than I thought it would be. I never saw a choose kernel at boot screen when it started up, but I found the grub.conf file and commented out the newer kernel, it had both listed. Gave it a reboot and it came up fine. Real easy fix!!!!!
If you are not seeing the count-down, your monitor may be taking a while to sync (and the timeout expires before you see it). You can increase the timeout in the grub.conf with, for example:
Yes, yum update will install a new kernel if it is available. It will leave the prior kernel available to be selected as I indicated above.
My machine was running 2.6.19-1.2911.fc6. Before the DST change, I allowed my computer to be updated by installing about 80 packages. After the install, I had 2.6.19-1.2911.6.5.fc6. Once updated, my nvidia drivers failed to load; it says I don't have nvidia drivers for 2.6.19-1.2911.6.5.fc6. I've tried all kinds of stuff to get the nvidia drivers to work with 2.6.19-1.2911.6.5.fc6, including a yum update for which there was nothing to install, but have had no luck. Booting into 2.6.19-1.2911.fc6 does not have the problem so I've been manually selecting the older kernel at boot time for about a week and a half.
So, System Patches are also suppose to update the kernel too... Right?
The nvidia kernel driver is third party software. As a result, the rpm for the kernel has no knowledge of it - it's not marked as a dependency. You need to manually update the nvdia driver each time you update the kernel.
The official nvidia installer works with the latest kernel.
LinuxQuestions.org is looking for people interested in writing
Editorials, Articles, Reviews, and more. If you'd like to contribute
content, let us know.