Tag Info

New answers tagged

0

The hint during installation is pretty clear, no? Automatic installation of initscripts for your platform/distro is not supported. Please, in directory according to your plartform/distribution create symbolic link 'avgd' to initscript for AVG daemon /opt/avg/av/etc/init.d//avgd.all. Registering 'avgd' service to runlevels... Automatic ...


0

Put localpkg_gpgcheck=1 in /etc/yum.conf and ensure you always yum to install rpms. If you have an rpm on the local files system just do yum install <rpm-on-local-filesys> Make sure you also set these in all the repos in /etc/yum.repos.d gpgcheck=1 enabled=1 gpgkey=file:///etc/pki/rpm-gpg/SOME-APPROVED-GPG-KEY As you mentioned, the best you'll ...


0

Try running $ sudo rm /var/lib/rpm/.rpm.lock


2

You are actually asking multiple questions at once: how to create a repo how to resolve dependencies of a package (without internet connection) what benefits does a repo bring how to convince somebody a repo is actually beneficial Some of them are already covered here, but you should refine your question what you ask for in particular and then ask ...


0

It appears that yum no longer can find octave. I'm fairly sure that I was able to find it via yum search octave last week. Where can I find old yum repositories? The octave package is available through the EPEL repository for CentOS. The easiest way to set this up is: # yum install epel-release But you can also follow the instructions on the linked ...


0

Just to add to Mark's link; I recently had to upgrade to 6.6 (specifically). My method was: yum --releasever=6.6 update


0

To fix this: yum remove epel-release Re-download the epel-release RPM Install it yum install pyliblzma


1

As you mentioned that it should be an automated shell script, at least for the two yum commands you would need to add -y so yum will assume an answer of "yes" for all questions it will ask. See the yum man page, relevant excerpt: -y, --assumeyes Assume yes; assume that the answer to any question which would be asked is yes. ...


0

Satellite is Red Hat's product for lifecycle management, configuration management, etc, and is more than capable of meeting your stated requirements: https://access.redhat.com/products/red-hat-satellite Why are you not using it?


1

# yum remove libxcb-devel # rpm -Uvh --oldpackage libxcb-1.5-1.el6.i686.rpm These are two different packages. The -devel version is for compiling against, the libxcb package without -devel is what applications will be running against. Because you removed the -devel version that there isn't anything installed, so "yum info" only shows you the latest ...


0

AIUI the latest versions of abrt don't actually install anything, they just use yum to download the packages and it unpacks the data itself. Update: https://wiki.centos.org/TipsAndTricks/ABRT#head-6ec8c2ca60fa7a4f2e8167a19299ea6d61217df2 suggests that abrt-action-analyze-ccpp-local unpacks them to /var/cache/abrt-di/. Unless you have the very latest yum ...


1

First you need to define what "upgrade" means. Eg. upgrade with only a newer kernel will be an install (and likely an erase of an older one). If someone does "upgrade foo" does that count? What about "distro-sync" that ends up just doing upgrades (or if it doesn't)? Does it count if the transaction failed? Then stop greping output, it will only lead to pain ...


2

If you want to know what version is installed, just run: rpm -q libxcb-devel If you want to prevent upgrades to this package, you can add the package to the yum exclude configuration. Add the following to the main section in /etc/yum.conf: exclude=libxcb-devel The library version (e.g., in libxcb.so.1.1.0) very seldom tracks the package version, and ...


0

The simplest thing is a cronned job to wget a tree off a public repo box out in the world (mirror.centos.org anyone?) and have it mirror it all in /var/www/html/reponame . Later on, strongly consider looking into the cobbler project from redhat. It's a great repo manager, and later on will help you build repos into distros, glue that with profiles to ...


0

Is your application certified for RHEL6? If so, this is then a support case; and since you're paying for this, do leverage it first. If your application isn't certified for RHEL6, then it's a learning opportunity. "It this certified for RHEL(version)" should be in the first 5 questions you ask about a given product, as you want to be able to use the ...


0

Your existing login session didn't load with the wheel group membership. Now that it's been granted, you can incorporate it by entering: newgrp wheel ... without having to log out or reboot.


2

You have several pieces to your question, and I can't claim familiarity with all of the commands you run, but here's my take: Either run the whole script as root (directly) or via sudo. That way, you won't need to run sudo in the script itself. If you require sudo for a particular step, then you'll either need to set your ID up with a NOPASSWD flag in ...


2

Group changes on unix are not recognized by existing login sessions; assuming, say, a Linux system with the usermod command: $ groups user $ sudo usermod -G wheel $USER ... $ grep user /etc/group | grep wheel wheel:x:10:user $ groups user To see the group change, any existing sessions (e.g. SSH, X11, etc.) must be exited, and a new session made (e.g. open ...


1

You almost answered your question. Here is a way you can find latest 5 updated packages: cat /var/log/yum.log | grep Updated: | tail -5 Output example: Aug 05 13:28:34 Updated: virt-manager-common-1.1.0-9.git310f6527.fc21.noarch Aug 05 13:28:34 Updated: glusterfs-libs-3.5.5-2.fc21.i686 Aug 05 13:28:35 Updated: virt-manager-1.1.0-9.git310f6527.fc21.noarch ...


0

Have you tried to configure local yum server and local repositories for it? If no then please follow this link, and just start from the YUM configuration. I have mentioned the steps below as well: For GUI interface in RHEL 6.2 follow below steps::: For GUI interface in RHEL 6.2 you have to install all the gnome packages of RHEL 6.2. Using rpm it is not ...


3

Actually it is --enablerepo=repoglob and --disablerepo=repoglob, where repoglob can stand for a wildcard like * or full/part of a repo name. Wildcards should be quoted. This will enable or disable only the respective repos ad-hoc for that one command and will not do any changes to files in /etc/yum.repos.d/. For the initial question, if you run yum without ...


0

enabling or disabling a repo in CentOS is equivalent to zipping or unzipping a repo in /etc/yum.repos.d. for example, a fresh install of CentOS has the CD repo enabled, if you want to disable it, simply gzip the repo and gunzip the other to use them. Don't forget to yum update after you do that.


0

While I think you are tying to solve a human problem programmatically, which is not yet ready to be solved so, you may want to look into the mock builder tool. I'm told it can assess dependencies and install those as required.


0

Assuming that you are using CentOS because of the tag I can say this: Additional packages are often in 3rd party repos. Information on additional CentOS repos is available at http://wiki.centos.org/AdditionalResources/Repositories Pay attention to the reference on yum-priorities.


0

Repos are either defined via .repo files in /etc/yum.repos.d or via plugins, which are usually defined via files in /etc/yum/pluginconf.d If you run yum repolist --noplugins does your repo in question still show up? If you want to know the URLs of the packages from your mistery repo, you can use yumdownloader --urls packagename to see the URLs. ...


0

Downgrade is very straightforward when package have not any dependencies, which affect the downgrade and is more difficult with packages that have dependencies. YUM downgrade does not resolve dependencies automatically, so it must be done manually. If you wan to that then use yum downgrade package. (kindly tell the error message, i will edit my answer then)


0

Are you mismatching your architecture? Uname -i vs filenames on the CD I'm seeing a LOT of i686 there, where I may want to see x86_64; or the other way around. When I see this, and the dependency barf, it's always been an arch mismatch. Good hunting.


1

I don't want to replace nice answer provided by slm. You generally don't use any regular expressions (globs) when searching with yum search since the command search is already looking for sub-strings within the package names and their summaries. How do I know this? There's a message that tells you this when you use yum search. Name and summary matches ...


0

If you use yum search you don't have to use asterisk(*) character. Yum search searches for the whole text you entered. Try using "yum search zzip".


1

I recently had a similar problem with missing files on openSUSE 13.2 and fixed it with: rpm -Va --nodeps --nodigest --noscripts --nosignature --nofiledigest --nosize --nomode --nordev --nouser --nogroup --nomtime\ | awk -e '($1=="missing") && ($2=="d") {system("rpm -qf "$3);}; ($1=="missing") && ($2!="d") && ($2!="c") {system("rpm ...


0

The problem here is that you haven't configured your server to accept the key that signed the package as a valid GPG key for this repo. The easiest solution is of course to run yum with nogpgcheck, though it's obviously also less secure. But if you've downloaded the package and verified it, and are keeping it in a place where it shouldn't be possible to ...


0

If the package isn't signed, the simplest solution is probably yum --nogpgcheck install chef. If you have a local repository, you can set: gpgcheck=no In the corresponding .repo file.


1

On RHEL/CentOS 6 it's yum groupinfo and on newer versions (since Fedora 19 if I remember correctly) it's yum groups command which is also reflected here: http://yum.baseurl.org/wiki/YumGroups and here: https://fedoraproject.org/wiki/Features/YumGroupsAsObjects Some more information on group handling here: http://yum.baseurl.org/wiki/YumGroupsOverview One ...


-1

Have you considered installing a new system on a different HD (or within a VM) with the same packages you have right now on your Fedora 20 and then copying over from there your entire /var and /usr directories?


1

The main reason a service is masked is to prevent accidental starting or enabling of the service. For example firewalld can be masked with systemctl mask firewalld When starting the firewalld service, you may get the error message Failed to issue method call: Unit firewalld.service is masked. To resolve this, unmask the firewall with systemctl unmask ...


0

The best way is: yum shell <<EOF install foo ts solve remove foo run EOF ...using deplist/etc. will mostly work, but isn't guaranteed to get the exact same result.


0

It is a bad idea to install "unofficial" software, specially on "enterprise" distributions or any sort of machine with critical jobs. You are taking over the full responsibility of security (and other) fixes, and keeping it in synch with the rest of the system's software (libraries, configuration, security configuration, ...). Look for alternatives, check ...


1

You did not specify the distribution you are using. I guess it is rhel/centos 5 or 6: if so, you just need to add the proper EPEL repository to your YUM configuration and then yum update yum install tmux No need to download/compile it manually.


0

To add repo: Adding, Enabling, and Disabling a Yum Repository For tmux: wget https://github.com/downloads/libevent/libevent/libevent-2.0.21-stable.tar.gz tar xzvf libevent-2.0.21-stable.tar.gz cd libevent-2.0.21-stable ./configure && make sudo make install then: git clone git://git.code.sf.net/p/tmux/tmux-code tmux cd tmux sh autogen.sh ...


7

You can use the yum deplist command to generate a list of package dependencies: $ yum deplist bind dependency: /bin/bash provider: bash.x86_64 4.3.39-5.fc21 dependency: /bin/sh provider: bash.x86_64 4.3.39-5.fc21 dependency: bind-libs(x86-64) = 32:9.9.6-10.P1.fc21 provider: bind-libs.x86_64 32:9.9.6-10.P1.fc21 dependency: coreutils ...


0

Thanks to the other posters for giving me the heads up that it was looking for a local repository; it turns out the EPEL repo configuration was (for some reason) set up to override the centos-base repo configuration with a local file path. I removed this and now everything works fine.


1

Clean up the metadata files (all of the repomd.xml files): # yum clean metadata and try again. It also looks like at one time you may have made a local copy of the packages in /var/ftp. Check your /etc/yum.repos.d/CentOS-Base.repo file to confirm it's looking to the Internet for packages. The file should have something like this for at least [base] and ...


2

Your CentOS was configured with a local repo. You'll have to edit your repositories in /etc/yum.repos.d. Here is a sample for /etc/yum.repos.d/CentOS-bas.repo


0

I was able to solve using sudo yum install php-mysqlnd. I must have mistyped it when trying to install the first time. Doh! After installing php-mysqlnd run sudo service httpd restart. Apparently a coworker messed with our mysql drivers. Asked if he changed anything that had to do with php or mysql and he told me he didn't. Looked at .bash_history and ...


0

What you want to do is disable core rpm functionality. All the files contained in an rpm are stored together with a checksum, their permissions and some additional/optional flags in a package. There is no way to disable the change of permissions, as basically you want to be able to restore the default permissions if any file in the package was touched, or ...



Top 50 recent answers are included