Edit

Share via

Troubleshoot common issues in the yum and dnf package management tools for Linux

  • Article

This article discusses common issues that you might encounter when you use yum and dnf package management tools to install or update applications on Azure virtual machines (VMs), and provides solutions to these issues.

Caution

This article references CentOS, a Linux distribution that is nearing End Of Life (EOL) status. Please consider your use and planning accordingly. For more information, see the CentOS End Of Life guidance.

Overview

The yum command-line package management tool is used in Linux distributions that employ the RPM Package Manager. The tool name is an acronym for "Yellowdog Updater Modified" and was originally developed for Yellow Dog Linux. The yum tool has gained widespread usage, particularly in RPM-based distributions such as Red Hat Enterprise Linux (RHEL), CentOS, Oracle Linux, Mariner, and Fedora.

The dnf command-line package management tool, or "Dandified Yum," is a modernized and enhanced package manager utility for RPM-based Linux distributions.

Scenario 1: Repository access issue

You encounter errors that apply to certificates or network connectivity.

Solution 1

For an RHEL system, see Troubleshoot Red Hat RHUI certificate issues in Azure or Troubleshoot Red Hat RHUI connectivity issues in Azure.

For other Linux distributions, make sure that communication to the corresponding repository is allowed through port 443. Additionally, make sure that the operating system uses a valid certificate if a valid certificate is required.

Scenario 2: Dependency issue

You might experience any of several errors when you encounter a dependency conflict during an update. The following subsections display the output symptoms of typical dependency errors.

Symptom 2a

Error: 
 Problem: package leapp-0.16.0-2.el8.noarch requires python3-leapp = 0.16.0-2.el8, but none of the providers can be installed
  - package python2-leapp-0.16.0-1.el7_9.noarch conflicts with python3-leapp provided by python3-leapp-0.16.0-2.el8.noarch
  - package python3-leapp-0.16.0-2.el8.noarch conflicts with python2-leapp provided by python2-leapp-0.16.0-1.el7_9.noarch
  - cannot install the best update candidate for package leapp-0.16.0-1.el7_9.noarch
  - problem with installed package python2-leapp-0.16.0-1.el7_9.noarch
(try to add '--allowerasing' to command line to replace conflicting packages or '--skip-broken' to skip uninstallable packages or '--nobest' to use not only best candidate packages)

Symptom 2b

Error: 
 Problem 1: package systemd-udev-239-78.el8.x86_64 requires systemd(x86-64) = 239-78.el8, but none of the providers can be installed
  - cannot install the best update candidate for package systemd-udev-239-74.el8_8.5.x86_64
  - package systemd-239-78.el8.x86_64 is filtered out by exclude filtering
 Problem 2: package systemd-container-239-78.el8.x86_64 requires systemd(x86-64) = 239-78.el8, but none of the providers can be installed
  - cannot install the best update candidate for package systemd-container-239-74.el8_8.5.x86_64
  - package systemd-239-78.el8.x86_64 is filtered out by exclude filtering
 Problem 3: package systemd-pam-239-78.el8.x86_64 requires systemd = 239-78.el8, but none of the providers can be installed
  - cannot install the best update candidate for package systemd-pam-239-74.el8_8.5.x86_64
  - package systemd-239-78.el8.i686 is filtered out by exclude filtering
  - package systemd-239-78.el8.x86_64 is filtered out by exclude filtering
 Problem 4: systemd-libs-239-74.el8_8.5.i686 has inferior architecture
  - package systemd-239-74.el8_8.5.x86_64 requires systemd-libs = 239-74.el8_8.5, but none of the providers can be installed
  - cannot install both systemd-libs-239-78.el8.x86_64 and systemd-libs-239-74.el8_8.5.x86_64
  - cannot install the best update candidate for package systemd-libs-239-74.el8_8.5.x86_64
  - problem with installed package systemd-239-74.el8_8.5.x86_64
(try to add '--allowerasing' to command line to replace conflicting packages or '--skip-broken' to skip uninstallable packages or '--nobest' to use not only best candidate packages)

Solution 2

When you encounter a dependency issue, start by identifying the specific combination of package and version that's missing or incompatible. The yum error message typically indicates which package-version combination is required but isn't available.

Solution for symptom 2a: Delete the older package

For symptom 2a, yum rejects a transaction and outputs an error message. An incompatibility between the python2-leapp package from an older release and the latest release causes the error.

To fix this error, you must remove the older package that causes the conflict (in this case, python2-leapp-0.16.0-1.el7_9.noarch) by running the following yum remove command:

sudo yum remove python2-leapp-0.16.0-1.el7_9.noarch

Then, you can continue the update or installation of the required package.

Solution for symptom 2b: Remove the package exclusion from the configuration file

For symptom 2b, yum rejects a transaction and outputs an error message. An exclusion filter causes the error.

To fix this error so that you can resume updating or installing the required package, make sure that the systemd package isn't excluded in either the /etc/yum.conf or /etc/dnf.conf configuration file. To check whether the systemd package exclusion is occurring, run a command such as the following cat/grep combination:

sudo cat /etc/yum.conf | grep -i exclude

If the systemd package is being excluded, the following output appears:

exclude=systemd

To end this package exclusion, use a text editor (such as vi, vim, or nano) to remove or comment out the entry for the "exclude" line in the /etc/yum.conf or /etc/dnf.conf file.

Scenario 3: Wrong version of Python

The yum, dnf, and rpm tools are written in Python and depend on the Python interpreter to operate correctly. Specifically, they rely on certain features and libraries that the Python interpreter provides to run their code and interact with the system. Changing the default version of Python that's installed on the system can cause issues within yum or dnf because of a dependency on specific Python versions and configurations.

For example, you might experience one of the following errors if you don't have the correct version of Python installed:

  • "Invalid syntax"

    yum list all
  File "/usr/bin/yum", line 30
    except KeyboardInterrupt, e:
                            ^
SyntaxError: invalid syntax

  • "No module named yum"

    yum repolist
There was a problem importing one of the Python modules
required to run yum. The error leading to this problem was:

   No module named yum

  • "Bad interpreter: No such file or directory"

    bash: /usr/bin/yum: /usr/bin/pythonX.X: bad interpreter: No such file or directory

Solution 3

By consulting the official documentation for your Linux distribution, you can confidently verify the default Python version and compare it against the version that's installed on your system. This process helps to ensure consistency and compatibility between software dependencies and system configurations.

After you identify the default version of Python for your Linux distribution, you can determine the currently installed version by running one of the following commands:

sudo ls -al `which python`

sudo ls -lrth /usr/bin/python*

sudo rpm -V python

If you verify that an unsupported version of Python is in use, you have several options to fix this issue.

Remove the symlink to the unsupported version of Python and create a symlink to the supported version of Python by running the following commands:

sudo cd /usr/bin
sudo unlink python
sudo ln -s python2.7 python  # Modify if necessary.

Solution 3b: Reinstall Python by using RPM

Reinstall the Python package by running the following rpm command:

sudo rpm -ivh python-<release>.<arch>.rpm --replacepkgs --replacefiles

Solution 3c: Reinstall Python from a rescue VM

Reinstall the Python package from a rescue VM. For instructions, see Missing important system core libraries and packages.

Scenario 4: Duplicate packages

If an interruption occurs while yum is updating, yum might retain the old package versions alongside the new packages that it's installing. Duplicate packages can cause confusion in system operations during subsequent use of the tool and eventually cause "protected multilib" errors or misleading dependency issues.

The following scenarios are typical:

  • Running yum update over Secure Shell (SSH) and experiencing connectivity disruptions during the update process

  • Sending a SIGINT and a kill -2 signal to the yum process, or selecting Ctrl+C to terminate the yum process while it's actively running

  • Experiencing a system restart midway through the update process

Solution 4

There are two possible solutions: Try to complete the transaction, or manually remove the duplicate packages. The steps to complete these approaches differ depending on the distribution and version of your Linux installation.

Solution 4a: Try to complete the transaction

  1. Try to complete the transaction by running the following yum-complete-transaction command:

    sudo yum-complete-transaction

    If the operation is successful without leaving any more duplicates, you can skip step 2.

  2. If duplicates are present, resolve this transaction by specifying the --cleanup-only parameter in the yum-complete-transaction command:

    sudo yum-complete-transaction --cleanup-only

Solution 4b: Manually remove duplicates

  1. Remove the duplicate packages manually by running the following commands. The yum check command might require a considerable amount of time, but obtaining its output is crucial for proceeding with these steps:

    sudo tar -cjf /tmp/rpm_dbbkp.tar.bz2 /var/lib/{rpm,yum}
sudo yum check &> /tmp/yumcheck
grep "duplicate" /tmp/yumcheck | awk '{ print $NF }' | egrep -v "\:" > /tmp/duplicaterpms
grep "duplicate" /tmp/yumcheck | awk '{ print $NF }' | egrep ":" | awk -F':' '{ print $NF }' >> /tmp/duplicaterpms
for i in $(cat /tmp/duplicaterpms); do sudo rpm -e --justdb --nodeps $i; done
sudo yum update

  2. After the duplicates are removed, if a new kernel was included in the failed transaction, reinstall the kernel to make sure that the correct generation of Grand Unified Bootloader (GRUB) entries, the initial RAM disk (initrd), and the initial RAM file system (initramfs) are installed.

    sudo yum remove kernel-<newversion>-<release>.<arch>
sudo yum install kernel-<newversion>-<release>.<arch>

Scenario 5: Yum doesn't work and shows a '404 Not Found' error

A "404 Not Found" error message in the yum output indicates that yum couldn't find the package or resource on the server. This issue typically occurs when yum tries to download or access a package from a repository, but the package file or metadata that's associated with it is missing or unavailable on the server. This scenario has various causes. Some of these are the following:

  • The package or resource was removed from the repository.

  • The repository URL or configuration is incorrect.

  • Network issues prevent yum from accessing the repository server.

  • There was temporary server downtime or maintenance.

Examples of this error are shown in sample output for various Linux distributions.

Red Hat Enterprise Linux 9 for x86_64 - Supplem  46  B/s |  14  B     00:00    
Errors during downloading metadata for repository 'rhel-9-for-x86_64-supplementary-rhui-rpms':
  - Status code: 404 for https://rhui4-1.microsoft.com/pulp/repos/content/dist/rhel9/rhui/9.1/x86_64/supplementary/os/repodata/repomd.xml (IP: 52.142.4.99)
Error: Failed to download metadata for repo 'rhel-9-for-x86_64-

Solution 5

The following possible solutions apply to various Linux distributions.

Make sure that you follow the correct process when you switch between Extended Update Support (EUS) lifecycle repositories and non-EUS repositories, or from non-EUS to EUS. For more information, see Image update behavior.

If the error typically occurs because an incorrect value is used for /etc/yum/vars/releasever and /etc/dnf/vars/releasever, or because the value should be absent (for non-EUS repositories), follow these steps:

  1. Check whether your system uses EUS or non-EUS repositories. To do this, run the following rpm/grep command combination:

    sudo rpm -qa | grep -i rhui

    The file name format that the command produces depends on the repository type, as shown in the following table.

    Repository type File name format
    EUS rhui-azure-rhelX-eus-<new-version>-<release>.noarch
    non-EUS rhui-azure-rhelX-<new-version>-<release>.noarch

  2. Determine the current value for the release version by running the following cat command.

    RHEL version Command
    RHEL 7 sudo cat /etc/yum/vars/releasever
    RHEL 8 and 9 sudo cat /etc/dnf/vars/releasever

  3. Modify and use the correct value for the release version:

    • If your system is using EUS repositories, make sure that the appropriate version lock is applied. In the following example, the version lock is added for RHEL 9.2 because RHUI EUS repositories are used:

      sudo sh -c 'echo 9.2 > /etc/dnf/vars/releasever'

      Note

      Support for RHEL7 EUS ended on August 30, 2021. We recommend that you no longer use EUS repositories in RHEL7.

      RHEL 8.x versions for EUS are available. These versions include RHEL 8.1, 8.2, 8.4, 8.6, and 8.8.

      RHEL 9.x versions for EUS are available. Currently, these versions include RHEL 9.0 and 9.2.

      For more information about version availability, see Red Hat Enterprise Linux Life Cycle.

    • If your system is using non-EUS repositories, make sure that you remove the version lock for the /etc/yum/vars/releasever or /etc/dnf/vars/releasever file by running the following command.

      RHEL version Command
      RHEL 7 sudo echo "" > /etc/yum/vars/releasever
      RHEL 8 and 9 sudo echo "" > /etc/dnf/vars/releasever

Scenario 6: RPM database issue

The following errors appear if you run the rpm, up2date, yum, dnf, or tdnf command:

error: rpmdb: BDB0113 Thread/process 24669/140693557245760 failed: BDB1507 Thread died in Berkeley DB library
error: db5 error(-30973) from dbenv->failchk: BDB0087 DB_RUNRECOVERY: Fatal error, run database recovery
error: cannot open Packages index using db5 - (-30973)
error: cannot open Packages database in /var/lib/rpm
CRITICAL:yum.main:

These errors might occur for any of the following reasons:

  • Broken RPM databases
  • Another application using the database
  • Insufficient space in the /var/lib/rpm directory

Solution 6

Rebuild the RPM database by following these steps:

Important

Before you do these steps, make sure that enough free disk space is available.

  1. Create a backup of the /var/lib/rpm directory by running the tar command:

    sudo tar zcvf /var/lib/rpm-backup.tar.gz /var/lib/rpm

  2. To avoid stale locks, remove the __db* files and verify the integrity of the Packages file:

    sudo cd /var/lib/rpm
sudo rm -f __db.*
sudo /usr/lib/rpm/rpmdb_verify Packages

  3. Rebuild the RPM database by running the rpm command:

    sudo rpm -vv --rebuilddb

Scenario 7: Yum command fails and returns '[Errno 14] HTTPS Error 403 --Forbidden'

The following errors appear if you run the yum command on a Red Hat 7.x VM that's connected to RHUI.

https://rhui4-1.microsoft.com/pulp/repos/content/dist/rhel/rhui/server/7/7Server/x86_64/rh-common/os/repodata/repomd.xml: [Errno 14] HTTPS Error 403 - Forbidden

Solution 7

  1. Check whether a third-party curl package is installed on your VM:
sudo rpm -qa | grep -i curl

rpm -q --queryformat '%{VENDOR}\n' curl libcurl

curl-7.73.0-2.0.cf.rhel7.x86_64 
libcurl-7.73.0-2.0.cf.rhel7.x86_64
libcurl-devel-7.73.0-2.0.cf.rhel7.x86_64 

city-fan.org repo http://www.city-fan.org/ftp/contrib/

If any third-party package is installed, go to step 2.

Important

The curl packages from third-party sources are provided with their own binaries and certificates that are not recognized by Red Hat. This incompatibility causes yum to encounter issues.

  1. Use either of the following methods to dowload the latest version of the curl, libcurl, and libcurl-devel packages that are provided for RHEL 7.9:

    • Download the packages manually by logging on to Red Hat Download, and then upload the rpms files to the affected VM.

    • Log on to a working RHEL 7.9 VM (PAYGO), download the required packages by using the yumdownloader command, and then copy the rpms files to the affected VM:

      sudo yumdownloader curl.x86_64 libcurl.x86_64 libcurl-devel.x86_64

  2. Downgrade the curl packages by using the rpms files from step 2:

    sudo cd /path/location/of/rpms/downloaded
sudo yum downgrade curl-X.XX.0-XX.el7_9.X.x86_64.rpm libcurl-X.XX.X-XX.el7_9.X.x86_64.rpm libcurl-devel-X.XX.X-XX.el7_9.X.x86_64.rpm --disablerepo=*

Third-party information disclaimer

The third-party products that this article discusses are manufactured by companies that are independent of Microsoft. Microsoft makes no warranty, implied or otherwise, about the performance or reliability of these products.

Third-party contact disclaimer

Microsoft provides third-party contact information to help you find additional information about this topic. This contact information may change without notice. Microsoft does not guarantee the accuracy of third-party contact information.

Contact us for help

If you have questions or need help, create a support request, or ask Azure community support. You can also submit product feedback to Azure feedback community.