Skip to content


Switch branches/tags

Name already in use

A tag already exists with the provided branch name. Many Git commands accept both tag and branch names, so creating this branch may cause unexpected behavior. Are you sure you want to create this branch?

Latest commit


Git stats


Failed to load latest commit information.
Latest commit message
Commit time


Since the move to pulp 3, this script is no more working properly. Currently, I have no time to continue working on that project and try to find a solution.

By looking at the following link,, under Advisory upload, it seems there is a way to upload advisories in json format. If someone can write a patch to add support for that, I'll be more than happy to integrate it


This script imports CentOS Errata into Katello and use preformatted information from

This is a python rewrite of a perl script made by brdude with some modifications, like the use of a redis cache.

To run this script on CentOS you need:

  • pulp-admin-client
  • pulp-rpm-admin-extensions
  • redis server
  • Some python modules
    • lxml
    • PyYAML
    • pyaml
    • redis
    • requests

It has been tested on CentOS 7 with the default python version and with python34 from EPEL. I'm using pew to test it inside a python virtual environment


  • I offer no guarantees that this script will work for you. It is offered as is!
  • I'm not an experimented Python programer, so this script may look horrific to anyone familiar with the language.


Katello settings

The option "Mirror on Sync" has to be set to "No" for the CentOS repositories. if set to "Yes", the repositories will be mirror from upstream and all erratas will be lost, as upstream doesn't publish erratas


pulp-admin must authenticate to pulp. This authentication information can be provided to pulp-admin in two ways.

  1. User certificate (~/.pulp/user-cert.pem)
    If you are using this script with katello, the foreman-installer creates a certificate suitable for use with pulp. You can use the cert by doing the following:
mkdir ~/.pulp/
chmod 0700 ~/.pulp/
sudo cat /etc/pki/katello/certs/pulp-client.crt /etc/pki/katello/private/pulp-client.key > ~/.pulp/user-cert.pem
chmod 400 ~/.pulp/user-cert.pem
  1. Admin configuration file (~/.pulp/admin.conf)
    You can provide the auth credentials in the pulp-admin configuration file. Simply create ~/.pulp/admin.conf, you can get the password from /etc/pulp/server.conf (default_password).
mkdir ~/.pulp/
chmod 0700 ~/.pulp/
sudo cp /etc/pulp/admin/admin.conf ~/.pulp/
sed -i "20,30s/^# host:.*/host: $(hostname -f)/g" ~/.pulp/admin.conf
PULP_PASS=$(sudo awk '/^default_password/ {print $2}' /etc/pulp/server.conf)
cat >> ~/.pulp/admin.conf << EOF

username: admin
password: ${PULP_PASS}
chmod 0400 ~/.pulp/admin.conf

It is probably advisable to not store these credentials in a normal user's home directory. You might consider using the root user for pulp-admin tasks. Then non-privileged users can be given rights explicitly through sudo. If you choose this way, the previous commands are still valid but as you will be connected as root, using sudo will be useless.

Redis server

Modify the configuration file to change the bind address if needed, and to enable persistent storage.

Right now, these scripts are not using authentification to connect to the redis server, so protected mode must be disabled (depending of your redis version).

Configuration file

  • Rename the sample-config.yaml to config.yaml
  • Fill the information for
    • your Katello server
    • the directory that will contains the data files
    • your redis server

For the reposotiries part, this is a little tricky.

  • katello-repository-label: this is the 'Label' fied that you can find in the webui, by clicking on a repository name inside a product
  • pulp_id: this is the 'Backend Identifier' fied that you can find in the webui, by clicking on a repository name inside a product
  • release: this is the CentOS release matching the repository content (must be 6 or 7 right now)


  1. Sync repositories
  2. Run the script to download the last datafiles from Steve Meier and Red Hat sites
  3. Run the script to store errata data into Redis
  4. Run the script to start the creation of the errata into Katello
  5. Wait for the repositories synchronizations triggered by the script to finish and the erratas will appear


Please feel free to make pull requests for any issues or errors in the script you may find.


Imports CentOS (from errata into Katello







No releases published


No packages published