Fedora X86/64bit :: 14 Completely Unusable; Probably Because Kernel Update
May 24, 2011
My main systems runs Fedora 14 x64 with the latest updates (I can't check anymore, but I believe it has the .13.91 x64 kernel). Since this afternoon it became unsuable. First, Opera Browser crashed and the crash report app of the browser showed up. Clicking one of the options there resulted to X crash, if I understood what really went on. No input, mouse or kb, I couldn't go to tty etc. I had to do a hard reset. Next boot proceeded normally up to the point that GNOME desktop is fully loaded. ABRT began reporting numerous kernel oops.
Rebooted again (noticed that init.d fails, so hard reset again) and tried past two kernel versions, same results.
Yum crashes and so does everything. I tried to revert all recent updates with yum history, but yum crashed.
Last update was May 22, and it updated kmod-nvidia.
Quote:
uname -a
Linux innovator 2.6.35.13-91.fc14.x86_64 #1 SMP Tue May 3 13:23:06 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux
initial crash
Package: kernel
[code]...
before the reboot Opera 11.11 had crashed and it's own bug report tool came up. Clicking the "do not report this error now" option probably crashed X, because the screen went black and no input was possible (no mouse or keybord, so I couldn't go to tty either. Had to do a hard reset) Bug Reports:Kernel oops report was uploaded
One of the numerous crashes
Package: kernel
Latest Crash:Τρι 24 Μάι 2011 09:19:56 μμ
Can anyone else confirm that Inkscape (from the FC15 distribution repo) is running as slow as a wet wig?
I have a fast machine (8 core 3.4ghz, 4GB RAM, 64bit FC15) everything else graphics wise is smooth and sexy but with inkscape try and zoom in with NO image loaded and it tic, tic, tic, jerky movements etc
Load an image and it is completely unusable, and has actually crashed my machine a couple of times when i was trying to work with a 5MB SVG file
I recently bought a second monitor and I had this set up perfectly fine with two panels on my primary display and one on my secondary. I've now gone away for the weekend (leaving the monitor) and I went to turn on my laptop and it didn't work.
When I login to a gnome session (or indeed a failsafe gnome session) I just got two horizontal white bars (where the panels would be). I switch to a terminal, login and run top and see that gnome-panel is on 100%. Running 'killall gnome-panel' does nothing (tried a few times).
I've had to install xfce4 just to type this message. Is there any way I can 'reset' gnome-panel or any other fix? Or even a workaround would be nice. I'm on 9.10 by the way. I am going to upgrade at some point but its not really an option yet.
I am unable to boot my Fedora 14 after last nights kernel update. When I start the boot up and hit F2 the system fails after outputting the line : Starting vservers of type 'default'. At some point the console displays an error message :
serial8250 : Too much work for irq17
How do I restore my previous version? I know some people have suggested in the past - just modify the grub.conf but how? When I boot off the cd I only get access to the liveuser disk. I need to have access to my own /boot/grub/grub.conf file.
I'm new to Fedora/Linux trying it out because I'm becoming hacked off with Windows taking so much time to do anything. The problem I have is that every now and again the computer slows down so drastically that the mouse movements are jerky, response to clicks/key inputs take about 2-3 mins to get a response and the only solution is to press and hold the power button. I'm running a Packard Bell Dot M/U. It has mobile celeron 1.3GHz 64 bit, 3GB ram. Intel wi-fi, atheros LAN (i think), HDMI output. I believe that it is v similar to an Acer Aspire 1410 (but not 100% same).
I just tried to update to kernel 2.6.27.15-170.2.24.fc10 (x86_64) at the suggestion of the gnome update widget, and the process hung. A bunch of other things were updating at the same time, many of which were related to audacity or wine.
I'm not sure just what was being updated, because the option to review which updates are available does not work - when I click on it, nothing happens, but when I take the other choice (install all updates), I take pot luck. It would be nice to be able to choose which programs to update, and to be able to upgrade the kernel independently, but I don't seem to be able to do that at the moment.
Keep trying update the packages (36 in total) and every time I keep getting this error ERROR with rpm_check_debug vs depsolve: kernel-x86_64 is needed by kmod-nvidia-2.6.26.6-49.fc8-173.14.12-5.fc8.2.x86_64 kernel-x86_64 is needed by kmod-nvidia-2.6.26.8-57.fc8-173.14.12-5.fc8.3.x86_64 Please report this error at [URL]
Initial FC12 install worked great, kernel 2.6.31.5-127.fc12.x86_64 allowed radio kill switch functionality Fn-F11 for wlan and Fn-F12 for bluetooth.The last two kernels have broken that functionality.
Since 11.04 was released, I've used it in Gnome and now also KDE, and this problem is in both. Aptitude is completely unusable. The main issue is that when downloading packages, it disconnects every minute if not every half a minute, making package installing a very hard process and making the configuration of a new OS take days (just to install 1GB of packages). Ontop of the disconnections, aptitude doesn't even want to go full speed when it IS connected.
The other issues involve repositories. Whether it's the official Ubuntu package servers, ANY of its many mirrors, or even my own custom PPAs, connecting to ANY repository when I reload my packages is a hit-or-miss, and it's completely impossible to get them all in one refresh (in fact, I don't think I've been able to access the main server since upgrading. Ever. I've been on the US server). This makes installing all of my packages take even LONGER. Also, to add insult to injury, the list files in /var/lib/apt/lists constantly corrupt themselves and make themselves unusable, ESPECIALLY the PPA lists (anything that uses launchpad has to be disabled at all times unless I have to update its program, or else I have to delete those list files every day).
Googling any of these issues separately gets me nowhere, so I'm making a topic, and it better not be the millionth topic I've made here that never got solved. The problem can't be my router or modem because my brother runs aptitude just fine on his 11.04 laptop. This issue is within my computer.
I meant apt-get. You mean they aren't the same program with two ways to run them? Yes, apt-get is the problemed one. I never actually type "aptitude" into the Terminal. As a consequence, Synaptic and KPackageKit and any other apt-get-using program is messed up as well.
I have had a good look round the forums but cannot come up with anything quite like this and the problem prevents me from using my NAS box and stuff is starting to stack up on my main PC now...
I have a Dell desktop running Ubuntu 9.04, a Shuttle running Mythbuntu 9.10 and a Netgear ReadyNAS Duo, all of which worked fine until this problem started. I have done nothing to any config files nor have I modified any settings from the outset, as far as I can remember I have just used everything as it came. This problem probably started a month or so back when update manager suggested a Samba update on both PC's which I accepted. I don't fire up my NAS box more than once or twice a month to back all my photos, music, video etc. so it was a while before I did (following the update) I could see the NAS and the Shuttle from the Dell but couldn't browse them I just got the error message 'Unable to mount location. Failed to retrieve share list from server' which I gather is all too common. I have searched around a bit on these forums and have tried a few things in the command line to gather information and I found something that worried me rather a lot. From Dell:
Code: geoff@geoff-dell:~$ findsmb *=DMB +=LMB IP ADDR NETBIOS NAME WORKGROUP/OS/VERSION
[Code]...
I haven't done any updates on the NAS box firmware so it can't be that and now I don't even see the shares in filemanager. What on earth is going on??? I know there are many other threads similar to this but but I am hoping that the slight difference might make this more obvious to solve. I forgot to mention that now I don't even see the shares in filedmanager - another thing that happened all by itself!
I downloaded 64-bit version of Fedora 11 and burned the iso on my dvd. When I rebooted the machine it booted up from the dvd but it showed an error message that kernel "image not found". And it stops there itself. Can someone show me some way except downloading the iso again?
I upgraded my F13 kernel from 2.6.33 to 2.6.34. But, the newer kernel isn't booting. It starts the booting process and then goes idle. how shall I troubleshoot?
I've managed to get some sound through the system with kernel 2.6.40.1. I use multiple sound cards, I have problems with keeping the same order. I added in etc/modprobe.conf The list of sound devices, putting my D66 card as device 3, Rebooted and cat /proc/asound/cards showed all fore devices in the correct order. I got audio through one of the apps , and started to try and get another ,wsjt, to function.
I have been running F14 as a 32-bit system (original m/b was 32-bit). Now the system is on a 64-bit dual-core processor board and I'd like to change from the 32-bit kernel to 64-bit. I downloaded a 64-bit kernel (kernel-2.6.35.6-45.fc14.86_64.rpm) and attempted to install it. It won't install and Package Installer returns a message of the sort that says this file "has incompatible architecture.
Valid architectures are ['athlon', 'i686', 'i586', 'i486'....]". My processor is an Athlon 64 x2. Is it possible to do what I want to do (I only want the kernel at the moment)? Is there some Fedora file I need to modify to allow the system to install the 64-bit kernel for this m/b? Do I need another file (or files) installed before I install the kernel?
I have a new work laptop (Dell M6400) that I currently have F10 i386 on. I would like to take advantage to the 64bit processor, but don't know if I need to completely re-install my F10, or if I just have to add the proper 64 bit kernel and applications.
Ive loaded 64studio (hda6) grub in MBR, and fed10 grub on its partition (hda8), but from Studio it seems i cant Mount fed10 files to look at the Kernel path from MBR. (studio did not auto pick it up, like it did for Mandriva on hda7)
Does anyone know what it is ? something like kernel /boot/vmlinuz.....and Initrd(hd0,7)/boot.initrd.img .....then i can Edit menu.Lst in MBR grub.
I am running a vmware on a 64 bit fedora 9. I have already installed kernel header, but the vmware could not find the kernel header, and ask to input the kernel header path. I added /usr/include to the vmware, it complained not a right kernel header.
The -162 kernel upgrade just released failed to boot in my x86-64 system. Additionally, both previous kernels failed in (probably) the same way. The progress icon proceeded all the way to its 'f' image, and then hung. I touched escape key, and found several disturbing messages among a dozen or so lines --
mdadm: No arrays found in config file or automatically. A dynamic linking error occurred: /lib64/multipath/libcheckdirectio.so: no such file or directory awk: cmd. line:1: fatal: cannot open file `/etc/inittab` for reading (Invalid Argument). Init: rcS post-stop process (807) terminated with status 2s. I got into rescue mode with my install disk, then looked for libcheckdirectio.so in /mnt/sysimage/lib64 but it was not there. Also, locate -d /mnt/sysimage/mlocate/mlocate.db/libcheckdirectio.so found nothing.
Ran chroot and executed the fakd diagnostic (which I had not seen before) and rebooted. Same problem - although now the screen is telling me SELinux is requiring a targeted policy relabel. My dual-boot-to-vista system has been erratic (requiring fsck several times, and confounding the Vista partition to require file checks), primarily since a large install of Eclipse and multiple plug-ins AFAICR.
I have an issue with kernel-2.6.31.9-174.fc12.x86_64. I ran the updates on Christmas Eve and it updated from kernel-2.6.31.6-166.fc12.x86_64 to kernel-2.6.31.9-174.fc12.x86_64. When I rebooted, I no longer received the Fedora Bubble boot screen. It comes up with the Fedora boot bar (across the bottom). Also, I receive failures during boot with the following:
I've downloaded 2.6.36-rc8 vanilla kernel, then I copied .config file from my current working kernel 2.6.32.21-168.fc12.x86_64, then I've configured, compiled and installed kernel like this:
Code: make gconfig make -j4 all (or make all) make modules_all make install The last command edits my grub.conf file and writes this: [Code]....
I checked my .config and ACPI, and File Systems are built into kernel and not loaded as modules... And, I have LVM but my /boot partition isn't in it, so I don't HAVE to use initrd, right? How can I boot from a vanilla kernel without initrd ?
In an attempt to ameliorate some other problems (lm_sensors, and loopback), and based on several forum entries, I decided to update the default f15 kernel (2.6.38.35) using kernel-2.6.39-1.fc16.x86_64.rpm from rawhide. It seemed to install without problems and when I rebooted it seemed to hang just before completing the boot process. The on screen boot messages ended with:
Started Display Manager Started LSB: daemon for libvirt ........ Starting LSB: suspend/resume libvirt guests on shutdown/boot [ 12 ...... about a dozen similar lines ending with [25.030991 hci_cmd_timer: hci0 command tx timeout then it apparently hangs
However, at that point I can log into to a console using Cntl-Alt-f2, so most, if not all, the crucial boot processes seem to have completed. There is an issue with the nVidia driver that I will deal with by modifying /boot/grub/grub.conf to re-enable nouveau, but I am stumped by what is causing the hang up. There is a certain amount of traffic regarding a similar problem with Ubuntu 11.04, but no solution other than it seems to be related to a BIOS bug (NFI)
what sub-system uses the "hci_cmd_timer" and why it might interfere with the boot process? Has anyone else run into a similar problem with f15?
Is there a better way to get an updated kernel in f15? I do want to keep all the Red Hat patches and back ports, so reluctant to build a kernel from scratch from upstream sources at kernel.org (and maybe just a bit apprehensive ;-})
I have F10 x86_64 installed on my desktop. When I update the packages, everything works well, except curl-7.19.4-6.fc10.x86_64 which conflicts with the installed curl-7.18.2-6.fc8.i386 (Fedora 8!). The problem seems to be that my repo list does not have the info about the F10 i386 repository. This is the report from "yum repolist"
------------------------------------------------------ Loaded plugins: refresh-packagekit repo id repo name status adobe-linux-i386 Adobe Systems Incorporated enabled: 17 fedora Fedora 10 - x86_64 enabled: 14,303
[code]...
I don't want to simply remove the f8.i386 package, because it'll take away a lot of other i386 packages that I'm sure I'll miss in the future (nspluginwrapper- .3.0-2.fc10.i386).
The other solution seems to be updating the curl-***.i386 package. However, I'm somewhat embarrased to confess that I don't know where to find the repository information...
