Tag Info

New answers tagged

2

This is how I fixed my problem. This is quite the messy situation. You may fix this by cleaning out rpm database. To minimize risk, make a backup of files in /var/lib/rpm/ using cp command: mkdir /root/backups.rpm.mm_dd_yyyy/ cp -avr /var/lib/rpm/ /root/backups.rpm.mm_dd_yyyy/ To fix this problem, try: # rm -f /var/lib/rpm/__db* # db_verify ...


2

My CentOS knowledge is a bit out of date, but you may need to also specify the timezone in /etc/sysconfig/clock. As a last-ditch effort, of course, you could: chatter +i /etc/localtime It is surprising that tzdata munges /etc/localtime though. Have you made it a symlink to /usr/share/zoneinfo/...? Are you perhaps updating glibc at the same time as ...


2

glibc upgrades often require restarting running daemons (because the name service switch [NSS] ABI changes, or just to get the daemon to actually used the upgraded version). That's probably what you're seeing.


0

For part of your question and for completeness, if you only wanted to reduce/limit the bandwidth consumption for yum, there is the throttle option that can be either enabled globally or for a specific plugin or repo There are other settings that you can leverage to use more of the yum cache and do not refresh repository metadata as often by default. See ...


1

A source rpm can create one or more binary rpms. Unless you are building your own binary rpms, you don't need to worry about the source rpm. So repoquery is showing you that bind-9.9.4-14.el7_0.1.src.rpm is the source rpm for bind-license. That same source rpm creates bind, bind-libs, bind-license, bind-utils (and many more).


0

repoquery will query a repository, that means a remote server which stores a bunch of RPM, and tell you some informations about it (ie: files included in this RPM). repoquery doesn't care if the package you are requesting is installed on your system unless you add the --installed flag. I guess the command you typed will output nothing if you add this flag: ...


0

I think it is caused by PackageKit. You have to check for PackageKit and disable it (I assume it is CentOS with systemctl, otherwise you can use service and chkconfig): systemctl stop packagekitd systemctl disable packagekitd Another approach is to open /etc/yum/pluginconf.d/refresh-packagekit.conf with a text editor, and change enabled=1 to enabled=0. ...


0

I asked and answered a similar question on the Fedora Q&A site. It turns out that Yum stores its history in an SQLite database in /var/lib/yum/history. I was able to concoct this SQL join that shows all the manually installed packages along with their installation date and times. In my case, I was only interested in the packages I'd manually ...


1

As I am not sure it is possible in the 'Requires' part of the script to specify a repo as a dependency, it may be better to include the repo file in the package you are building, or add as a dependency a RPM that provides it. A existing solution for including the repo file is the one used in the Google Chrome package, which is to add the repo file and GPG ...


0

In our case, it was a firewall issue ... we were allowing our host outbound http (port 80) access, but not https (443). The regular repos are accessed via the former, but EPEL repos are via the latter. So w/o EPEL in the loop a 'yum update' (or whatever) was working fine, but w/ EPEL, it was bombing, ==> One of the configured repositories failed ...


0

This is how I fixed the issue. This will be help you For me this issues occur because of the IPV6 internet connection in the Fedora. So first we have to disable IPV6 totally. here what you have to do Open Terminal Type su and enter to log in as the super user Enter the root password Type cd /etc/modprobe.d/ to change directory to /etc/modprobe.d/ Type vi ...


1

Download the latest version of the libcurl rpm from rpmfind.net or your preferred mirror and run: rpm -e --nodeps libcurl rpm -ivh libcurl PS: I think that your problem is same as Bugzilla 960765


1

Like shibley said, this would defeat the purpose of the package manager. It would be wiser to download the source and compile it with the version of the dependency you need. Hopefully, for you, the version you want will work without any source mods.


1

Yum doesn't have a way to do this. It would counter the purpose of the package manager. Moreover, it will usually render a package dysfunctional. Package maintainer's put effort into ensuring dependencies are properly defined and not overly restrictive. That said, you can install the rpm with the rpm utility using the --nodeps flag. As a result, it ...


0

There are cases where you use rpm first, to perform some preliminary stuff, before yum. A good example is MySQL. (1) rpm -iv mysql-community-release-el6-5.noarch.rpm At this point the mysql/yum repository is ready and you can start relying on yum only to get the main mysql and the rest of optional packages: (2) yum install mysql-community-server See ...


0

Solved! From askapache: "Do Not set "ProxyRequests On". Setting ProxyRequests On turns yourserver into an Open Proxy. There are 'bots scanning the Web for openproxies. When they find you, they'll start using you to route aroundblocks and filters to access questionable or illegal material. Atworst, they might be able to route email spam through your proxy. ...


0

It sounds like your system is missing a Python cryptographic library file that is required by yum. You should get better diagnostic information by launching the python command interpreter and entering import yum. You can then edit your question to include this information.


1

You could try rebuilding the rpmdb rpmdb --rebuilddb This fixed a similar problem for me.


0

Because they are arch dependent. Either rebuild the .src.rpm on the arch. you care about (the one in the source repos. is built on a random supported arch), or download and unpuck the .src.rpm and yum-buildep on the kernel.spec.


0

You should do ldd /lib64/libldap-2.4.so.2, check output for 'undefined references', find package which provides that undefined lib, find your repo, manually download the rpm which provides that lib, install it with rpm -i <missing>.rpm. Most probably your problem is due to bug in libldap or your manual intervention into packaging system. I happen to ...



Top 50 recent answers are included