Red Hat Training

A Red Hat training course is available for Red Hat Satellite

13.2. Performing Critical Updates to the Server

Red Hat releases critical updates to Red Hat Satellite in the form of an Erratum. The method used to provide errata updates to Red Hat Satellite is determined by whether it is in online or offline mode.

Important

Apache RPMs do not restart the httpd service upon installation. Conducting a full update of the Red Hat Satellite Server (such as with the command yum update) might cause Apache to fail. To avoid this, make sure to restart the httpd service after upgrading it.

13.2.1. Performing Critical Updates to an Online Server

Updates to an online Satellite Server are done using the yum tool. All Satellite services, except for the database, are shut down during the update process.

Procedure 13.1. Perform Critical Updates to an Online Server

  1. Stop the Satellite services, except for the Satellite database.
    # rhn-satellite stop --exclude postgresql
    
  2. Take a backup of the Satellite's database in a working state. Run the following command and replace the [FILENAME] option with the full path to the backup file that you want to create. This location needs to be writable by the PostgreSQL user:
    # db-control online-backup FILENAME
    
  3. Apply the updates:
    # yum update
    
    Apply all Satellite updates. Updating the schema without updating the rest of the Satellite components can cause issues with the Satellite database.
  4. Update the database schema using spacewalk-schema-upgrade command.
    # spacewalk-schema-upgrade
    
    This process will update your database schema to latest version. The spacewalk-schema-upgrade command will inform you with the results of the upgrade and the locations of schema upgrade log files. To double-check if the schema update completed successfully, run the following commands:
    # rpm -q satellite-schema
    # rhn-schema-version
    
    If the output versions match, continue with the process. Otherwise restore the database with db-control restore /path/to/backup.
  5. Restart Red Hat Satellite:
    # rhn-satellite start --exclude postgresql
    
  6. Clear the search index:
    # service rhn-search cleanindex
    
    It is recommended to clean the search index. The above command triggers the creation of a new one, which in most cases completes within thirty minutes to an hour. You might experience issues with the search features of Satellite 5 if you do not clean the index.