Results 1 to 7 of 7
Not that I need or want to do this, but I got to wondering...
Is there a practical limit to how new a kernel you can use with how old ...
- 04-18-2006 #1
Really new kernels on really old distro versions?
Not that I need or want to do this, but I got to wondering...
Is there a practical limit to how new a kernel you can use with how old a distro? I realize a lot of drivers and things get deprecated over time, but for instance, could you run Mandrake 8.2 with a 2.6.16 kernel? Has anyone had the experience of running a really new kernel on a really old distro version?
- 04-18-2006 #2
I don't see why this wouldn't work, though things like modules and maybe a few other things might break.
I've managed to boot Slackware 10.0 (not old I know) with both the 2.4.* kernel and the 2.6.* kernel. If you change the kernel in something like an old mandrake I don't see much difference to changing the kernels in Slack.
dylunio
- 04-18-2006 #3
I have Redhat 6 lying around somewhere. Just for the fun of it, (and all the new things I'm sure I will learn!) I might see if I can get a 2.6 kernel running with it. Should be interesting.
- 04-18-2006 #4
Have fun Dapper Dan
- 04-18-2006 #5
As long as your distro respect the "Current Minimal Requirements" as stated in <linux-source>/Documentation/Changes, it should be working.
"To express yourself in freedom, you must die to everything of yesterday. From the 'old', you derive security; from the 'new', you gain the flow."
-Bruce Lee
- 04-18-2006 #6
I wouldn't expect any problems, as long as you compile the modules at the same time. As far as I'm aware, nothing else cares what version the kernel is. That's one thing I really like about Linux--everything can be updated independently, and on an ongoing basis, so there's never a need for a re-install or a whole-system upgrade.
Stand up and be counted as a Linux user!
- 04-25-2006 #7Just Joined!
- Join Date
- Apr 2006
- Location
- Petaluma, CA, USA
- Posts
- 3
I have just had a 2.6.12 kernel running redhat 6.1; however, it was accidental, and due to unsolved weird behavior between lilo and the kernel (it should boot debian on root=hdb9, but boots redhat on root=hda9)! I'm sure one can do a lot better intentionally, with a properly compiled kernel.
Originally Posted by Dapper Dan
In this case, of course, there's lots of messages about incorrect kernel, map, obsolete system calls, etc, and X fails to run, the system gets to run level 3, and after various stumbles it becomes possible to log in on the tty console. I've got a pretty complete log grep'd out of the kernel messages now for a most unusual boot; here's a a few selections (a little while later, the message "Welcome to RedHat" show up, but that's not logged):
Apr 25 01:27:15 linus kernel: [4294667.296000] Linux version 2.6.12-10-386 (buil
dd@terranova) (gcc version 3.4.5 20050809 (prerelease) (Ubuntu 3.4.4-6ubuntu8.1)
) #1 Sat Mar 11 16:13:17 UTC 2006
And, the kernel got the right root passed by lilo (349 = hdb9) :
Apr 25 01:27:15 linus kernel: [4294667.296000] Kernel command line: BOOT_IMAGE=ub2612 ro root=349.
a few selected error lines:
Apr 25 01:27:15 linus kernel: Inspecting /boot/System.map
Apr 25 01:27:15 linus kernel: Cannot find map file.
Apr 25 01:27:16 linus kernel: [ 60.737338] device-mapper: 4.4.0-ioctl (2005-01-12) initialised: dm-devel@redhat.com
Apr 25 01:27:41 linus cups: Warning: /boot/System.map has an incorrect kernel version.
Apr 25 01:27:41 linus cups: Warning: /usr/src/linux/System.map has an incorrect kernel version.
Cheers!
Johnpipe


Reply With Quote
