Fedora Installation :: Safely Remove Unsupported Packages After FC13 Upgrade?
Jun 2, 2010
I managed to successfully upgrade FC11 to FC13. While doing the post upgrade steps, I made a HUGE mistake. After identifying the unsupported packages, I used "yum remove package1 package2" command to remove the obsoleted ones without realizing the yum was also removing many FC13 packages! Here are the cmd sequence used:
I simply typed Yes and left unattended for a few minutes after yum prompted there were # of packages to be removed as I blindly believed it would do the job. After returning to the computer, I found many installed icons were disappearing, the wireless suddenly turned off etc. I killed the yum process right away but it was too late.
-Is there anything I could do to undo the yum remove process?
I recently did upgrade from F12 to F13 using preupgrade, and I was following the guide from the Fedora project website [URL]
On the section Removing unsupported packages it shows how to find orphaned packages and sugest you remove them. I tried that, found a bunch of packages that I promptly erased. But then I came across
ModemManager-0.4-5.git20100720.fc12.x86_64
I thought "well I don't use a Modem" so I tried to yum erase it. I stopped when I saw that it was trying to also remove the dependencies, a bunch of NetworkManager packages.
My question is, why is ModemManager flagged as an orphan if it has packages depending on it, isn't that wrong?
And also, the recommendation on the guide to remove these packages can be dangerous to people who don't pay attention (imagine someone accidentally removing NetworkManager, no internet).
I will keep the "orphaned" package out of necessity, but something is wrong in this whole deal.
I had a fairly smooth upgrade process using PreUpgrade on my main workstation, but I still have a huge bunch of fc12, some of which have fc13 analogs installed, some of which don't (a total of 317 packages). Also, I still have the Constantine gdm theme, probably as a symptom of these lingering packages. Can I just yum erase these? Do I need to check which have analogs in fc13? I don't know which step of the upgrade I forgot, but it seems I'm stuck somewhere in between, although
To remove pendrive when I click the 'safely remove' tab instead of getting removed from the desktop it reappears again. This problem is there in fedora 14
Took an FC12 system to FC13 via 'preupgrade'There was a REALLY old version of gimp-print-cups in RPM, and the installer announced "a fatal error occurred". However, everything else seemed to download properly and I allowed the upgrader to reboot the system.
a) grub boots the kernel, then the kernel can't find root and panics. It does this because it can't root file system with root=UUID= kernel option. If I manually tweek the grub line to use root=/dev/sdx, kernel finds its root and boots up to the next problem.
b) GUI environment not launched. Init scripts die on "error loading modules.dep". Sure enough - one is not there. Ran "depmod" to generate it ,and rebooted. System reboots into GUI, where I login to find the next problem.
c) Evolution won't start. Reason turns out the evolution-data-server from FC12 is still present. Did a check on just how many other FC12 RPM's are present: 1100. Yeah - over 1000. About 850 of the FC12 RPM's also had FC13 replacements installed. Did an "rpm -e" on these. Also had to do some remove action on mkinitrd, nash, plymouth-scripts-0.8.0, and parted-1.9.0 as there are replacements for these utilities under different package names. That cleaned up the system some, to present me with the next problem.
d) yum package repository still linked into FC12. Removed fedora-release-12-2.noarch and then I could "yum install" evolution and a few other packages properly.
I've done one other FC12->FC13 upgrade today, and that went flawlessly. So, I think in general the upgrade works fine. It just looks like if preupgrade finds anything amiss, it doesn't take care of a lot of things and generally leaves you with an unbootable mess to clean up.
Anyone else seeing this? I do 'safely remove' to remove USB flash drive. disappears. Five seconds (or less) later, it reappears. The second time I do 'safely remove', it stays gone.
Using Xubuntu,I have upgraded from Karmic to Lucid. Post upgrading when attempting to install a new package with aptitude, it is reported that cups is "BROKEN" and a host of packages are marked for removal.
I am unable to login to kde after upgrade to fc14 from fc13 as user , the splash screen works for seconds and returns back to login screen, but I am able to do as root I also managed to login to KDE (failsafe session) and also I am able to login as normal user to gnome desktop
last week I encountered the following problem: The update manager suggests a distribution upgrade (strange enough since I already have 10.4 and 10.10 is not released yet). The even worse part is that 24 packages shall be removed from my system, including acpi, cryptsetup, network-manager, nvidia-current, plymouth-x11, samba and wine. About 5 weeks ago I experimented (without success) with disabling plymouth because I wanted textmode during booting. I found some warnings that one should not completly remove plymouth because lots of dependencies including cryptsetup (which is vital to my system). So I did not remove plymouth (a look in Synaptic confirms this), but I may have changed some settings concerning plymouth.I am not sure what this upgrade is about.
I have just upgraded my FC12 installation to FC13. The initial boot failed (just blank screen). Second time I received 'selinux targeted policy relabel is required'. I let that run, but still the boot did not suceed.
I then went in to verbose boot and saw that the boot would hang at libvirtd. So I disabled libvirtd for anything above run level 1. Next boot failed at atd, disabled atd. Next boot failed at 'monthly Smolt checkin'.
At that point and booted in to single user mode and dissabled SELinux, thinking that was the cause of all the problems. That did not help.
The update process has also removed my previous kernels, so I can't test a different kernel
I am having Fedora 11 installed on my laptop. I installed Ubuntu 9.04 a few days back but I don't seem to like it. I had installed grub loader of Ubuntu 9.04.
Can I safely remove 1 distro without screwing up the other? I have Linuxmint as secondary and ubuntu as the last.I want to replace ubuntu.If I just delete the partitions/format and install my other os which is OpenSuse 10.03 will this work.will opensuse see linuxmint and make grub understand?
I have an external HDD which I use under Fedora. After finishing with my work, I unmount all the mounted partitions of the external HDD & then proceed to switch it off. The HDD partitions are unmounted but they are still visible(but not mounted) under computer. The HDD still seems to spin. But when I click on "Safely Remove Drive" they disappear from Computer & also the HDD stops spinning albeit the switch on HDD being powered on. So what should I use "Unmount" or "Safely Remove". What is the difference between the two?
Until a recent software update, I encountered no problems when 'Safely Removing' my external hard drive.
After the following update:
Aug 23 14:36:03 Installed: kernel-devel-2.6.35.14-95.fc14.x86_64 Aug 23 14:36:13 Installed: kernel-2.6.35.14-95.fc14.x86_64
The system freezes when I try to safely remove the drive. What I see is the blue screen with the Fedora logo, the caps lock key lights up and the system is totally frozen. Following is the information on my external drive gleamed from the messages log file when the device was mounted:
Aug 26 07:53:03 localhost kernel: [ 496.855476] usb 1-1.2: new high speed USB device using ehci_hcd and address 7 Aug 26 07:53:04 localhost kernel: [ 496.942025] usb 1-1.2: New USB device found, idVendor=1058, idProduct=0704 Aug 26 07:53:04 localhost kernel: [ 496.942031] usb 1-1.2: New USB device strings: Mfr=1, Product=2, SerialNumber=3
Last nights upgrade of java-1.6.0-openjdk-1.6.0.0-43.1.8.2.fc13.i686 and java-1.6.0-openjdk-plugin-1.6.0.0-43.1.8.2.fc13.i686 prevent the loading of Ameritrades Command Center.Downgrading to version 1.6.0.0-37 fixes the problem. I'll post the bugzilla number as soon as I get the chance to file it.
This has been bothering me for years now...when I go to remove a thumb drive from my computer, I have two options when I right click the device eject and safely remove. What on earth is the difference supposed to be?
got my upgraded fedora 14 partition to boot, but I find out that I had to reinstall my nvidia driver, no big deal. seemed to go ok, but now when I go applications->system tools -> terminal I get the "starting terminal" thing on the bar, and then nothing. Can't figure out whats up.
How to add packages using X-Window's add/remove packages option in RHEL-5.3 as it shows only the currently installed package and and does not show any thing when we click the button "available packages" ?
I upgraded from FC11 to FC13 last night. The desktop has not been stable since. It will run for anything from 5 to 30 minutes before it locks up. Interesting, if I am already ssh'd in, I can perform commands, but a new SSH will hang, and if I try to SU to root in an already logged in ssh, it hangs.
However, I did finally manage to get the messages that were logged to /var/log/messages at the time of the desktop freeze.
Here are the messages. The first 3 about high speed usb may or may not be relevant. The freeze appeared to happen around the 4th message in the log below.
Code: Sep 2 13:16:57 starbug kernel: usb 2-4.2: reset high speed USB device using ehci_hcd and address 6 Sep 2 13:26:38 starbug kernel: usb 2-4.2: reset high speed USB device using ehci_hcd and address 6 Sep 2 13:31:20 starbug kernel: usb 2-4.2: reset high speed USB device using ehci_hcd and address 6 Sep 2 13:38:03 starbug kernel: INFO: task jbd2/dm-0-8:558 blocked for more than 120 seconds.
im using fedora 14 and i have a slow internet connection. i want 2 install some packages from the fedora 14 dvd instead of downloading from internet using add/remove packages. i tried to edit /etc/yum.repos.d/fedora.repo and /etc/yum.repos.d/fedora-updates.repo but it dint work.
I was having a job in /etc/event.d. It was working fine upto FC12. With upgrade to Fedora 13, it stopped working . Searching the internet, It looks like /etc/event.d is obsolete and I need to use /etc/init. Also the filename has to have a '.conf' extension. So I have renamed my job as /etc/init/svscan.conf.However, it fails to start automatically on reboot . It works if I manually issue the command % sudo /sbin/initctl emit qmailstartBut fails to start automatically on reboot. Is there anything wrong in my script below ? Let me know if I have missed out anything?
% cat /etc/init/svscan.conf # svscan - runlevel compatibility # WARNING: This file was auto-generated. Do not edit!
I have a problem with LVM. My installation of Fedora 13 is on 2 SATA hard disks, which are set as /dev/sda and /dev/sdb. On sda there are two partitions (boot and a physical part of the volume group) and on sdb is only one partition (other physical part of the volume group). Both physical parts of the volume group are used for one volume group and the VG is splittet into multiple logical groups. The system works and all seems to be ok. But I saw a strange thing on last reboot:
The device /dev/sda was not recognized during the boot sequence, instead the system has build a device /dev/md126p1 with the same geometry and partitions and it has the same disk id as sda.
The volume group is also ok, the system boots into the desktop environment, but the system-config-lvm says that /dev/sda is not initialized and the volume group consists of sdb and md126p1. fdisk -l knows three devices (sda, sdb and md126p1). I think, that is the reason for my problems with preupgrade, which stops after the reboot with a LVM problem (root not found). Can I correct that without data loss?
I'm running Fedora 15 GNOME3 on a Compaq Presario C700 laptop.
I have an external Transcend HDD. I have three partitions on it.
When I insert it, it's automatically detected and the partitions mounted and works perfectly well but when I "Safely Remove" the disk, Fedora freezes and hangs. The mouse is immobilized and none of the keyboard actions work. I can't even enter the Virtual Terminal. The music playing in the background loops a short 1 second buffer or so.
However, if I eject all the partitions by clicking on the eject button in Nautilus and then "Safely Remove Drive", it does not crash.
I have 350GB external Western Digital USB hard Drive.When I try to remove it from the system by executing Safely Remove Drive menu the fedora 15 system gets stuck.The processor starts giving a hum sound and it goes on even if it is left for half an hour in the stuck state.The Mouse is not working and everything is halted.
Is there a good/simple way to do this?OK - Under certain other OSs, the 'safely remove' option on usb devices will...1/ On usb pen drive with a power/activity led, switch it off (trivial nicety, but reassuring).2/ On usb HDDs, spin down the disk - seems more important.I don't know if it really is but 'pulling the plug' on a spinning HD makes me nervous!I have tried a number of linux distros recently and only ubuntu 10.04 with gnome seems to do the same. Most others (including openSUSE 11.2, kde 4.3.5 - which I have) only seem to unmount
I am currently running the xen (64 bit) kernel, but want to move to the non xen kernel(64 bit) while retaining my carefully crafted system. I tried this once before by unticking the "virtualisation" and it removed the xen kernel, leaving me with nothing to boot from.
After some recent upgrade of my Debian Testing i386, on ThinkPad T400s, I am receiving panic message upon Safely Remove Drive.When I insert external HDD, it is automatically detected and the partitions mounted and works perfectly well. But when I "Safely Remove" the disk, Debian freezes and hangs. Nothing works (mouse or keyboard), and even X crashes and I get frozen terminal.I never experienced anything similar
