I just did an upgrade and it included the kernel. As time goes by, I will be stuck with a lot of kernels to choose from, and lower disk space. How do I remove an older kernel and everything it's related to it without breaking stuff?
My company offers Linux preloads in RHEL, Fedora and Ubuntu. I have Fedora 13 installed on my laptop. I want to install it on a desktop but the 2.6.33 kernel hangs the desktop when the Fedora 13 or Ubuntu Lucid DVD boots. I have the Ubuntu Jaunty preload on the desktop but would prefer Fedora. The preload includes a few apps for my job that aren't included with Ubuntu. Is it possible to select an older kernel to boot to and/or use that during the install? I think it is the SCSI controller causing this. I have the same model machine at home configured the same way except the onboard SCSI is disabled, I am using SATA drives in it.
I had the 180.xxx nvidia drivers on my FC10 setup.I installed the 185.xxx and am running into issues. It gives me the fatal server error: no screens found message. Before that it says that the nvidia kernel module is 180.xxx and the driver component has 185.xxx. I can't seem to figure out how to uninstall the previous module to make this work. I removed the yum kmod nvidia. I'm running this using sh nvidia.185xxx.xxxx.run . The weird thing is that it sometimes would work but then I would restart and it would break again. Thanks in advance,
How can I install older fedora kernel? Because I accidentally removed the working kernel among three kernels. I've tried yum install kernel-2.6.30.10-105.2.4.fc11.i586 but fails.
Is there any legal way to install an older version of linux-kernel (say 2.6.35) on 11.04? By legal I mean - no source-compiling and third-party repo adding.
I am building a series of custom kernels...for one of them i am using a kernel that is older than my currently installed ones.
I am using rpm to install this kernel and it will not install, period. i don't get why i can't install this....what does it matter that i am installing an older kernel? and why is fedora/rpm designed to not let me do this?
Code:
There should be an easy way to get around this, but i haven't figured it out yet. i've googled around, no luck. i bounce between kernel versions all the time without problems on my system, so why is rpm setup for strict versions on a kernel? and is there no way around this, and i just have to delete all three of those kernels, i don't really want to. 2 of them are awesome and i shouldn't have to build them again, that would be complete crap! and i wouldn't have this problem on my debian setup, so why fedora? (i generally think fedora is better)
I just had a hard drive failure on my mythtv box that I had been running F14 on. I reinstalled F14 and just got about everything up and running like normal again except that I upgraded the kernel by mistake when I was upgrading everything else to current levels. The problem is that I run a hauppauge pvr150 on this mythtv box, and the code for capturing on this card is broken on the latest kernel version. I have the kernel headers and development files installed for the newest kernel that I can't really use, but I didn't install the kernel headers for the older kernel before the upgrade. Now, I am looking for a way to forcefully install the older kernel development files so that I can compile my the proprietary nvidia drivers against it for my adapter. I downloaded the rpm files from the fedora 14 repository directly. Is there an option for rpm that I can use to force the installation?
I need to prevent that the latest kernel update removes the only kernel that still works in my computer. how can i do this? Currently I have 3 linux kernels versions:
But only the oldest works. With the new update there is a new kernel version (2.6.32-24) that possibly will not boot (like previous 2.6.32-* kernels) and also I expect that, as in previous updates, the oldest kernel will become inaccessible, thus rendering my ubuntu unbootable.
PS: unsolved threads on the underlaying problem: Can't boot default kernel after upgrading from 9.xx to 10.04 lucid Upgrade to 10.04 freezes on the Ubuntu screen
Seem my rotation part is not removing files older than 90 days. Anybody know what is wrong?
Code: #!/bin/sh #navigate to the desired backup location cd /public/backup/linux #dump the MySQL entirely, output file is dated mysqldump -u root -pmt1jxz68f2 --all-databases > "`date +%Y%m%d`.sql" #backup the web folder
How do you go about removing old kernel from other distros from the grub menu. For example I have lucid and ubuntu studio installed. Lucid controls the grub2. So how would I remove the old kernel out of the ubuntu studio?
Is removing previous older kernels "vmlinuz", when updates do get a newer version an easy task, or do the more recent Linux like Ubuntu 9.10 do that automatically, and only keep a few older previous kernels ?
I installed the kernel source RPM to add a patch to a driver kernel module that didn't support my NIC. Now that the patch is part of the official source tree I no longer need the kernel source and I want to delete it. Can I just do rm -rf rpmbuild or do I need to run the reverse of the rpm commands described in Section 1 of [URL].
How do I remove old kernels from the Fedora grub menu? I have three currently and I want to keep only one old version to fall back if I have any problems with the current one. When I was using Ubuntu, I had Ubuntu Tweak which had an option to clean up old kernels from the grub menu .. is there something similar for Fedora?
Here's how to always have only the latest kernel version in your grub: Code: sudo gedit /etc/grub.d/10_linux
Find these lines: Code: list=`for i in /boot/vmlinu[xz]-* /vmlinu[xz]-* ; do if grub_file_is_not_garbage "$i" ; then echo -n "$i " ; fi done`
And change it with this: Code: for ver in /boot/vmlinu[xz]-* ; do list="$ver" done
Notice two differences. 1. There's no ` before "for" and after "done" keywords. 2. There's no list= before "for"
If you want to change the name of entry, scroll down to the end of the file and find last two occurrences of: Code: linux_entry ${OS} stands for Ubuntu and ${version} stands for your kernel version. I changed this into: Code: linux_entry "${OS} 9.10 Karmic Koala"
You can use your own imagination. If you want to have specific name for your Windows entry, in /boot/grub/grub.cfg copy everything between: Code: ### BEGIN /etc/grub.d/30_os-prober ### and Code: ### END /etc/grub.d/30_os-prober ### paste it in /etc/grub.d/40_custom
Change the text between double quotes right after menu entry. (For example, my 40_custom looks like this now: Code: #!/bin/sh exec tail -n +3 $0 # This file provides an easy way to add custom menu entries. Simply type the menu entries you want to add after this comment. Be careful not to change the 'exec tail' line above. menuentry "Windows 7" { insmod ntfs set root=(hd0,1) search --no-floppy --fs-uuid --set bcdc24dddc249424 chainloader +1 } ) and make 30_os-prober not executable: Code: sudo chmod -x 30_os-prober
If you want to remove recovery mode entry, just uncomment this line in /etc/default/grub: Code: #GRUB_DISABLE_LINUX_RECOVERY="true" PS Don't forget to run Code: sudo update-grub After you finish editing everything! And check in /boot/grub/grub.cfg if everything is ok!
I have winxp and ubuntu 10.04 dualboot. They were working ok. Today I removed old *21 kernel image and headers so grub updated the confs. That's all I did that could cause the win no longer boot. It starts booting, the screen goes black and the PC reboots. I tried safe mode, it started to load some dlls as it usually shows in safe mode but then still reboot.
Code: $ sudo fdisk -l Disk /dev/sda: 160.0 GB, 160041885696 bytes 255 heads, 63 sectors/track, 19457 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes
I have looked for existing threads on this issue, but found no matches, so I am starting a new one.I first installed Ubuntu 9.04 on my laptop from a bought CD last year. When I later installed the kernel updtates with the Update Manager, I noticed the the list of versions growing when I booted up. Then something horrible happened. After installing maybe the third or fourth kernel update, I couldn't boot into Ubuntu, it failed every time. After that I left Ubuntu broken on my laptop for a while.I have now reinstalled from the CD again However, I am now very wary when it comes to installing even the important security updates, never mind the recommended ones! I prefer to keep it simple, because I don't want to do any more reinstalls for a while. So far I have just one kernel headers update on the boot list, and I am trying to figure out how to remove the previous one. I do not see the point in having a growing list of them again.
I have Ubuntu 10.10 installed on my laptop.Its a dual boot with XP, have been using ubuntu/xp for a while now..Till now i always used to remove older kernel images after ubuntu update, then update grub and do "sudo apt-get autoclean" never had any problem at all.i skipped the second step, i removed the old kernel image and did "Autoclean" on apt-get but forgot to update grub.
Now after i have reboot, when grub loads i can see only 2 "Memtest" lines in grub list and 1 "windows xp" line.the "Linux kernel generic" line is missing. I can boot in xp (have done just that to post this) but since there is no line in grub list to boot into Ubuntu, cant boot in Ubuntu.Is there any option, something i can do at "Grub" to boot into Ubuntu?
As it written here [URL]. Fedora 13 does not include the kernel-source package provided by older versions, since only the kernel-devel package is required now to build external modules. Linux kernel, does not belong to fedora. Fedora is derivative work isn't it? Or you rewrite all kernel in fedora 13, and fedora is not linux any more? According to GPL, derivative works must be licensed as GPL. And so you must provide ALL sources for free. Including kernel sources. So, where can I get kernel sources for fedora 13? Or I'm wrong, and you shouldn't provide them if you wont?
I am having a problem getting F10 to boot after install.I downloaded the DVD, burned it and installed from it. No errors during the installation but when I go to reboot the system comes up saying:"VolGroup00 not foundUnable to access resume device (/dev/volgroup00/logvol01)Mount: error mounting /dev/root/ on /sysroot as ext3:h file or directory"This is being installed into an AMD Opteron Blade with an Adaptec SC2000 controller card, 4 SCSI disks configured in a RAID 5 array.I gather it's looking at the wrong device for a boot track/loader, but not sure how to fix thisAny ideas? I figured if it found it during install it should find it during boot but that doesn't seem to be the case
I tried installing Fedora 11 on my i386 using the 6 CD set that I downloaded. Since my system has only 128M of RAM, the text install mode was used. The 1st CD installation seemed to work OK. After 181 out of 181 packages were installed, the system rebooted and did not ask for me to insert the other CDs. now when my system comes up, it is always in text mode.
Q. Is there a command I can use to have the remaining CD's installed so I can install the graphics system?
I am trying to install Fedora 13 on a new system from CD. DISK 1 runs but terminates without using disks 2-5. There is a system I can access as 'root' but I was not prompted to create a unique user or which packages should be installed.How do I complete the process?
I recently installed fedora 9. Everything works fine after install (except sound but thats a different story). When i run updates everything seems to have worked fine but when i reboot afterwards i get to the login screen which works but as soon as i log in it freezes? the only way to restart is to press the power switch...
upgraded from Fedora 10 to 11, had both KDE and Gnome desktops installed under 10 but now upgraded to 11 I get a boot error when truing to boot to KDE and most KDE apps that used to run under Gnome no longer do so
I'm having trouble installing Fedora 11 using the URL method. When I put in a F11 netinstall CD, or even Disc 1, I press tab on the first option and add "askmethod" to the end of the text. Each time I point the installer to my local FTP and press enter after typing in the FTP url, the installer just hangs there. I've even tried deleting ALL the files from the F11 folder on my FTP server and downloaded ALL the files at: ftp://ftp.linux.ncsu.edu/pub/fedora/...edora/i386/os/
Still...the same thing. When I pointed the installer to: ftp://ftp.linux.ncsu.edu/pub/fedora/...edora/i386/os/ Everything worked fine, albeit slow as it was DL from the NCSU FTP server. Am I doing something wrong with my FTP server? It is anon, so there is no authentication issues.
In Ubuntu Server 10.04, the kernel naming convention went sort of like "vmlinuz-2.6.32.21-server" for servers and "vmlinuz-2.6.32.21-generic" for desktops... Kernel numbers changing for the kernel version, but last signifying whether it was a desktop or a server version of the kernel... Right? At least as much as I've noticed with my 64bit installs.
I recently got a 32bit test server up on Ubuntu 10.10 and I think I'm either confused, maybe things might be different in 10.10 versions or my prior assumptions were wrong? On this test machine, I installed Ubuntu Server 10.10 32bit. After the install, I upgraded packages, then I installed kubuntu-desktop, xubuntu-desktop and ubuntu-desktop (in that order). I know the implied resource and security issues-- but this was a test machine to work something out. My observation on this machine is that I cannot find a kernel image on it with "...-server". All the versions of kernel images are "...-generic" All my previous installs of Ubuntu Server have been 64bit installs... and I just assumed all server kernels had "...-server" as the naming convention.
- Is this different for 10.10? - Is this a difference in the 32bit server/64bit server kernel? - Did my installs of the GUI's somehow change to the desktop kernel? If so, all my testing in now invalidated- as I'm supposed to be testing on the "server kernel."
A few days ago yast did update my computer to a new kernel-desktop 2.6.31.12-0.1.1.
My mainboard is an ASUS P4P800-VM with Intel ICH5 chipset. This new kernel is not able, to poweroff my computer. Older kernels were able to poweroff the computer.
How can I roll back to the older kernel? Yast does only offer the recent kernel.
In /boot there are only files of the recent kernel.
ASUS P4P800 Intel Pentium 4 HT, 3000 MHz, Frontside 800 MHz/Cache 1 MB 3 GB Ram AGP nVidia 7600GS/512 MB, nVidia driver 190.53, installed from yast. PCI WLan TP-Link 951N (Atheros chipset, WLan N, driver ath9 was automatically installed)
what i did was, remove evolution mail from synaptic, what i wanted to do was just remove the indicator applet from the task bar. i read a bunch of bad stuff about removing evolution from synaptic vs just removing the applet.
im worried. did i break anything or put my security at risk. after, i used a command (older) (sudo apt-get install ubuntu-desktop)to install ubuntu desktop. because i thought that it would fix evolution. then i went to synaptic and installed a package called evolution. i rechecked evolution in applications menu. however, i notice that i have both a checkable evolution and two evolution icons. nothing 'seems' broken. im not sure if it ever was. and evolution calender pops up as normal, as does the the installed plain evolution. they both seems to be an exact copy of the other.
all i really wanted to do was remove the indicator applet. did i make a serious mistake. since ive had ubuntu, ive reformatted a lot because i was worried i made a mistake of some kind. however now im into the more "make a mistake and fix it stage' as im pretty happy with my current desktop and have worked hard to customize it. the command, sudo apt-get remove indicator-messages removed the mail icon. i still am worried that i broke something, or put my security at risk. also, now i have two mail icons. evolution mail and calendar, and another just called evolution.