CentOS 5 :: Yum Hang On 5.4 After Manually Remove / Add Packages With Force?
Nov 6, 2009
for a x problem I reinstall the complete x packages. I remove some packages with force.Before this yum works perfect. Exact at this time we have problem with our internet connections and yum hangs somewhere when yum load the repositories and or start the update process.Now yum hangs at start from the command. I can start yum some times and no locking error is rise.strace brings:
After updating to Karmic, Synaptic shows almost all of my installed packages in the category "Installed (manual)", including about half of the packages that belong to a clean Ubuntu installation (e.g. apparmor, apt and hundreds of others). As a result, I can't easily get a list of those packages that I did indeed install manually and may want to remove. Is there a way of removing the "Installed (manual)" flag from all packages?
If I could do this, all packages that do not belong to the core Ubuntu system should show up as "Installed (auto removable)" and I could individually mark only those as manually installed that I really still need and let apt/synaptic uninstall everything else. I know that with today's hard disks, disk usage of installed packages is not an issue. But those packages accumulate over time and need to be updated with every security update and every ubuntu dist-upgrade, wasting time and bandwidth.
If I do a fresh 'netinst' of CentOS 5.4 x64 on a server, what is the correct way to verify that no 32-bit packages were installed or mixed in with x64? Also can someone tell me if it is safe to remove those 32-bit RPM packages? I searched the Wiki for 'Post Install Tips' and could not find anything there or on Google.
Is there a way to manually force sync files? My files have not synced over a day and there's an "unsynchronized" emblem appearing over the Ubuntu One folder. The web interface for Ubuntu doesn't display the proper total size of Ubuntu One nor the current files/folders.
Continuing on my quest to build a system with sole purpose of being a LAMP server, I am trying to have the bare minimum number of packages installed. I'm not trying to disable as many deamons as possible, but actually remove any package that is not required. It might sound a bit extreme but each package that is installed that is not needed is using valuable space on the expensive (and small) 36GB 15krpm SAS disk.
Also it is a damage control approach, in that assuming a cracker finds an exploit in Apache or PHP that allows them to run executables on the filesystem then taking away as many tools as possible will help in damage limitation. For example I don't want an SSH client installed on the system as it has no purpose for me but could be used by a cracker to access other systems on the network, if they compromise that system.
I have installed the absolute minimum from the 5.3 install by deselecting everything during install but that still leaves a great deal of unnecessary packages installed. Seeing as how the system is a 64bit installation and will have no 32bit apps running on it I have first uninstalled all the i386 versions of various libraries. I presume their only function is to provide compatibility for 32bit apps.
I have been looking for information about removing other packages but I don't seem to be able to find information on real dependancies. For example I believe I have no need for dmraid as it is for managing software RAID arrays that I have no intention using. but if I attempt to remove it the dependancies include kernel and mkinitrd. What I can't figure out though is if these are hard dependancies in that those packages will fail to work or soft ones that simply mean without dmraid it will not be possible to configure the software raid module that is loaded into the kernel.
What packages would people recommend to remove, anyone seen a good guide to reducing down CentOS/RHEL to the absolute minimum ? python-elementtree-1.2.6 : Is this really required by YUM ? m2crypto-0.16 x86_64 : Is this really used by YUM, I didn't think YUM connected via SSL ?
I have a centos server x86_64 arch installed - i am getting some issues where i want to update rpms but because there is a equivliant i386 package installed i get dependency errors.Is it safe to run this command to remove all i386 packages - will my system still boot after this yum remove *.i?86
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.
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" ?
when i try to install the flash plugin and chromium in the open suse install remove tool it installs well and downloads but if i do anything else it hangs and makes all the OS unusable and i have to hard shutdown .. i tried to leave it to see if at least like that it can install but then the screen went blank(cause of power saving) and i oved the mouse to continue watching the progress and again it hanged... i am running opensuse 11.3 on an usb stick which i did using the dd_rescue method.it seems it has persistence.
I am just too tired and am missing something in the man pages for aptitude and not getting my search terms right but I can't seem to find a way to make aptitude ignore a couple of supposed broken packages (they work just fine by the way). The only way I can find to remidy the situation is to remove those couple of packages or upgrade several things to their Squeeze counter parts. I really don't want to try and do the upgrade as at home I am stuck on a 56k connection. Removing the broken packages I guess won't be too bad if I can find the debs again after I install a couple of things I want to add to my system. I would however just like to make aptitude ignore the state of my system and try to install what I want anyway.
how to remove a dbvis package as it causing yum to break. I have googled and searched the forums. I found this [URL]. I modified the python script that yum uses to find the offending package. This is the symptoms when i run yum...
Code:
[root@localhost ~]# yum -d 10 list Traceback (most recent call last): File "/usr/bin/yum", line 29, in <module>
I'm trying to force open office to the 3.2.1 version that is available in backports. When I force the openoffice.org package, and try to install it ( with synaptic ), it complains. I assume this is because the dependencies aren't right. Do I have to manually track down all the dependent packages and force their versions to comply as well?
The janitor wants to remove certain applications (svn, virtualbox) which I installed the easy way by going straight to the top and installing a GUI... I then realised the GUIs sucked, so I removed them. Now the computer janitor wants to remove virtuallbox, svn, and who knows what else all because of those stupid packages! how to mark certain packages as deliberatley installed so the computer janitor will take them and their dependencies off the list and let me clean things up?
I'd like to list all packages I installed since the installation. The tricky part is that I don't care for dependencies - only clean list of what I ordered to install. I went through man pages and I did not find anything relevant. Also /var/log/apt/history* doesn't say what I requested and what came as a dependency.
For gentoo-aware folks, I am looking for something like "world" file.
I have 64 bit processor and 64 bit compatible CentOS 5.5 distribution, but some times when I run yum info, I get information about both 32 abd 64 bit installed packages:
For example when I reinstall selinux I see that previously I had both archetecture match packages version:
Reinstalling:
My question is: Is there some rpm or yum configurable option to strict installation to some archetecture, that, that when rpm/yum will try to install package for i386 when it prohibited I will get some error?
I have messed up, I downloaded something about gnash through the repositories trying to get a flash player to feed my little house on the prairie on ..... habit(I know, addiction is sad). Anyway, it didn't work but now it won't let me remove it. I have flash player installed but it's doing the same thing as gnash by itself, I have black lines across the screen instead of a flash player. Is their some way I can force it to remove gnash?
I'm new to Fedora and not sure how things are done so please keep that in mind while you're trying to help me fix my stupid mistake. Thanks
going through synaptic and noticed that a ton of packages are marked as being manually installed when they most definitly did NOT installed them, is there a command i can use to reset all the dependencies?
Is there any way that I can get a list of packages (on the command line) that have been installed manually i.e. all those that haven't been installed as dependencies? I think this must be possible as apt seems to know which dependency packages are no longer required i.e. apt-get autoremove
I was working on my flash drive install of Ubuntu, when I squashfsed my /usr. Long story short, I some how ended up with a working /usr, but with a few packages marked as installed, but not having its components installed (emacs). When I try to remove emacs (emacs23-nox), it gives me numerous errors about files (all relating to emacs) not existing (all in /usr). Thus my questions are as follows:
1) Is there a way to force the removal without it caring about missing packages?
-OR-
2) Is there a way to reload which packages are installed by checking which files exist, etc?
I did a "dirty install" of Maverick over my existing Lucid system. That went very well and I am having no problems with Maverick. However, this morning, I decided to clean off the old Lucid kernels. In the past, after installing a new kernel on the same Ubuntu release, I have done this by running "aptitude search 2.6.32-24", for example, then running "sudo aptitude purge" for the kernel and header files it found.
Now that I have changed releases, aptitude no longer finds the Lucid kernels installed on my system, even though they still reside in the file system and show up on the grub2 menu. So, how do I manually find everything necessary to delete for the old Lucid kernels?
We all know that we can remove installed applications straight from terminal or ubuntu software center. However, sometimes we download .deb files from the web that are not necessarily on the repository. My question is, how do I uninstall a manually installed deb application?
Unbeknownst to me the LVM volume (on LUN) that I removed from my server was in use by another system. So instead of vgexporting the volume group on one of the servers and then lv,vg,pv-removing it on the other I dismantled the LVM devices my servers (as belwo).
0. unmount the filesystem 1. Remove logical volume (lvremove <vg>/<lv>) 2. Remove volume group (vgremove <vg>)
I have installed freeswitch app from opensuse 11.2 repositories. App got broken during the process/setup.., I have removed it (yast) ignoring some of it's dependencies, also I have removed folder /opt/freeswitch manually. In Yast it shows that app is still installed.. when trying to remove it, there is an error, force remove also out puts an error. "whereis" command shows no file, I was also trying to force install it one more time but it fails. Is there a was to delete it permanently?
I want to some how get a list of the packages I installed. I was hoping that I could just list all of the packages that were not installed automatically as a dependency. It turns out that there are 320 packages that match that description (I think). Is there a way to do what I want to do? Shouldn't all of these dependencies have been installed as a handful of meta-packages instead?
I want to remove a keyring package I installed from a repository that I no longer want to use. However, I cannot remove it:
# apt-get remove -y --force-yes debian-xray-keyring Reading package lists... Done Building dependency tree Reading state information... Done
The following packages will be removed: debian-xray-keyring 0 upgraded, 0 newly installed, 1 to remove and 130 not upgraded. After this operation, 49.2 kB disk space will be freed. (Reading database ... 181076 files and directories currently installed.)
Removing debian-xray-keyring ... gpg: key "AB8F901D" not found: eof gpg: AB8F901D: delete key failed: eof dpkg: error processing debian-xray-keyring (--remove): subprocess installed pre-removal script returned error exit status 2 configured to not write apport reports Errors were encountered while processing: debian-xray-keyring E: Sub-process /usr/bin/dpkg returned an error code (1)
I have a question, that may sound silly. I have removed VirtualBox from my Ubuntu install. But the .VirtualBox folder is still existing with a virtual drive of nearly 10 GB. Can I manually remove the folder .VirtualBox with rm -rf without any unwanted side effects?
In general CentOS search automatically after startup for available software updates.Then after some (~20-30) minutes an icon appears in the toolbar which the user can click and install the updates.How can I manually speed up/trigger IMMEDIATELY the search for updates (without waiting for the built-in search)?
I'm trying to install CentOS 5.4-x86 on a "new" motherboard, it's an ABIT Fatal1ty AN9 32X. I have a AMD 6000+ Dual Core running stable at 3.01 ghz. 4GB of ram(4x 1GB). nVidia GeForce 7600. 2x SATA II 160GB WD RAID 0 and 2x SATA II 1TB Seagate RAID 0. This setup has been running fine as a FTP server for months and has been flawless for games for about 3 years.Right off the bat during initial install it hangs before loading any GUI.Last lines:
NFORCE-MCP55: IDE controller PCI slot 0000:00:0c.0 NFORCE-MCP55: chipset revision 161 NFORCE-MCP55: not 100% native mode: will probe irqs later
I'm running into what is quite possibly one of the strangest problems I've ever encountered. We recently had a power loss, and some of our vmware instances didn't shut down correctly. Once of those is our Zabbix testing monitor. From that point on, whenever I run the "ll" command, my terminal freezes. This happens over both ssh and the local console. I've forced a disk check via the "shutdown -rF now" command, and it returned no errors. Since ll is simply an alias for ls, I copied ls (and the entire bin directory) from another Centos 5.3 instance, with no change in behavior.
