Help us improve your experience.

Let us know what you think.

Do you have time for a two-minute survey?

 
 

Upgrading Contrail Networking Release 19xx with RHOSP13 to Contrail Networking Release 2011 with RHOSP16.1

The goal of this topic is to provide a combined procedure to upgrade Red Hat OpenStack Platform (RHOSP) from RHOSP 13 to RHOSP 16.1 by leveraging Red Hat Fast Forward Upgrade (FFU) procedure while simultaneously upgrading Contrail Networking from Release 19xx to Release 2011. The procedure leverages the Zero Impact Upgrade (ZIU) procedure from Contrail to minimize the downtime.

The downtime will be reduced by not requiring extra server reboots in addition to the ones that the RHOSP FFU procedure already requires for Kernel/RHEL upgrades.

Refer to Red Hat OpenStack Framework for Upgrades (13 to16.1) documentation for details on RHOSP 13 to RHOSP 16.1 Fast Forward Upgrade (FFU) procedure of OpenStack Platform environment from one long life version to the next long life version.

Access ContrailImageTag located at README Access to Contrail Registry 20XX.

  1. Follow chapter 2—Planning and preparation for an in-place upgrade through chapter 8.3— Copying the Leapp data to the overcloud nodes of Red Hat OpenStack Framework for Upgrades (13 to16.1) procedure.
  2. Use predictable NIC names for overcloud nodes
    1. Log in to the undercloud as the stack user.
    2. Run the playbook-nics.yaml playbook on all overcloud nodes.

      $ ansible-playbook -i ~/inventory.yaml playbook-nics.yaml

      The playbook sets the new NIC prefix to em. To set a different NIC prefix, set the prefix variable when running the playbook:

      $ ansible-playbook -i ~/inventory.yaml -e prefix="mynic" playbook-nics.yaml

    3. Run the playbook-nics-vhost0.yaml playbook on all overcloud compute nodes.

      $ ansible-playbook -i ~/inventory.yaml -l overcloud_Compute $my_dir/playbook-nics-vhost0.yaml

      For details, see https://github.com/tungstenfabric/tf-deployment-test/blob/master/rhosp/ffu_ziu_13_16/tf_specific/playbook-nics-vhost0.yaml.

    4. Reboot overcloud nodes using the standard reboot procedures. For more information, see Red Hat rebooting nodes documentation.
  3. Follow chapter 8.3—Copying the Leapp data to the overcloud nodes through chapter 19.2—Upgrading Controller nodes with external Ceph deployments of Red Hat OpenStack Framework for Upgrades (13 to16.1) procedure.
  4. Upgrade compute nodes to OpenStack Platform 16.1.
    Note:

    If you are not using the default stack name (overcloud), set your stack name with the --stack STACK NAME option replacing STACK NAME with the name of your stack.

    1. Source the stackrc file.

      $ source ~/stackrc

    2. Migrate your instances. For details, see RedHat Migrating virtual machine instances between Compute nodes documentation.
    3. Run the upgrade command with the system_upgrade_prepare and system_upgrade_run tags.

      $ openstack overcloud upgrade run --stack STACK NAME --tags system_upgrade_prepare --limit overcloud-compute-0

      $ openstack overcloud upgrade run --stack STACK NAME --tags system_upgrade_run --limit overcloud-compute-0

      To upgrade multiple Compute nodes in parallel, set the --limit option to a comma-separated list of nodes that you want to upgrade.

    4. Run the upgrade command with no tags to perform Red Hat OpenStack Platform upgrade.

      $ openstack overcloud upgrade run --stack STACK NAME --limit overcloud-compute-0

  5. Follow chapter 19.4—Synchronizing the overcloud stack through chapter 21.4—Synchronizing the overcloud stack of Red Hat OpenStack Framework for Upgrades (13 to16.1) procedure to complete the upgrade.
    Note:

    To speed up the overcloud upgrade process, you can upgrade 1/3 of the control plane at a time, starting with the bootstrap nodes. For details, see Speeding Up An Overcloud Upgrade.

Related Documentation