rhel4 update 5 kernel upgrade


   There are several security fix every fortnight for rhel4.  Is it save to simply upgrade/update all the kernel related package under up2date? In the past i did that for debian and often causing issue on boot up.  Will it cause the same problem? Can anyone give me some advice on how to safely apply the patch with least hassle?

** The rhel4 is currently running mirror1 (rhel software raid).
Who is Participating?
Ideally, you should upgrade the kernel whenever you see an update with a security fix. Practically, since this will require a reboot you'll be doing it at night / over the weekend / etc. I would advise you to do it manually to be able to revert to a previous kernel if something goes wrong.

You should upgrade using

up2date -f

which will upgrade all kernel packages currently installed and skipped in a default configuration. Make sure you have enough space (20 mb or more) on your /boot partition, using

df -h

If you are running out of space there, you need to remove older kernels first. Check the list of installed kernels by running

rpm -q kernel
rpm -q kernel-smp

you can remove the oldest kernels (keep the last 2) using

rpm -e kernel-(full version number)

kernel is a regular uniprocessor kernel
kernel-smp is symmetrical multi-processor kernel used in systems with 2 or more CPU's
-devel packages are only needed if you rebuild the kernels yourself or use features like Dell's DKMS that require kernel source
-hugemem-devel is not needed if you are not running hugemem kernel. hugemem kernel is the unsupported kernel version for use on servers with more than 4Gb RAM. If you are not running this kernel, you don't need it.

You can determine the kernel version you are running by running:

uname -a

It has never caused me any problem in RHEL type systems (and it has in Debian type systems). Unless you use external (and manually added) modules, such as specific storage drivers (like newer cciss), or ocfs2 drivers (by Oracle). Other than that - no problem. Also - upgrading kernel does not erase your previous one. It adds the newer one as the default, but you can still boot using the previous one.
stock99Author Commented:
so i should upgrade all the kernel package as often as possible?

2.6.9-55.ELsmp #1 SMP Fri Apr 20 17:03:35 EDT 2007 i686 i686 i386 GNU/Linux

The option to upgrade are:


should i just go to select "kernel", "kernel-smp" and "kernel-smp-devel"  or all of them ?
Network Scalability - Handle Complex Environments

Monitor your entire network from a single platform. Free 30 Day Trial Now!

If you upgrade using yum, your kernel will be upgraded correctly - that is - your running kernel (and you can get the version by running 'uname -a') would be upgraded to the next upgrade. Incorrect (or unfit) versions won't be installed on your system. Also - you will still have the ability to boot using your previous kernel (the 2nd option in the boot menu).
Hi all,
After I've install RHEL4 AS (which is having 2.6.9-5 kernel version), I immediately run up2date so it download all package (about more than 1 thousand files including latest kernel: 2.6.9-55.0.2) and the update completed overnight. The next day when i booted my computer, I have new option in my grub menu so I selected the 2.6.9-55.0.2 ELsmp x86_64 and within 30 seconds, the kernel panic! The error message i received are:

Mounting root filesystem
mount: error 6 mounting ext3
mount: error 2 mounting none
Switching to new root
switchroot: mount failed: 22
umount /initrd/dev failed: 2
kernel panic - not syncing: Attempted to kill init!

I'm not sure what is the correct way to update to latest kernel but I assume using up2date under this will act like "WIndows update" which will update what ever are relevant? Another question is when I run up2date for the 1st time, will I need to update Errata as well? If yes, is it mean I need to download all Errata from Redhat network and manually update it myself? Sorry, I'm kind of new in RH AS.
It should have worked. Did you do anything special in order to make the system "boot" under RH when you first installed it?
What type of hardware is it? What model? Did you install special disk drivers? Use any RAID controller?
Hi Ezaton,
Actually, I've posted the question here. Please refer to the below link:

Can you boot if you select the old kernel in the list at the boot time?
Yes. I still can boot back to the old kernel which is 2.6.9-5.ELsmp
Try identifying at what version the kernels stop working - i.e. install, say, 2.6.9-22.0.1.ELsmp and see if that one works, if it does, try bumping it up to 2.6.9-42.0.10.ELsmp and so on. If any kernel other than the original one works, locate the version that stops working, and see if the changelog for it mentions any hardware you are using.

If none of these work except the original one, chances are the initrd's that are built for all updated kernels are missing some drivers you have in your initial initrd - you'll need to compare your working initrd with the new ones. initrd files are in /boot and you can unpack them in this manner:

gunzip < /boot/initrd-(version...).img |cpio -i --make-directories

See if you find any device drivers in your original initrd that is not in the newer ones.
Question has a verified solution.

Are you are experiencing a similar issue? Get a personalized answer when you ask a related question.

Have a better answer? Share it in a comment.

All Courses

From novice to tech pro — start learning today.