DO NOT SHARE THESE DOCS WITH CUSTOMERS!

This is an LA release that will only be provided to a select number of PLM-sanctioned customers (PDFs only). Contact PLM for details.


In this section:

Related articles:

SBC release 8.1 consolidated high-availability (HA) redundancy models into two supported options: N:1 HA and 1:1 HA. Because of changes related to HA consolidation, upgrading a 1:1 HA deployment from any prior release to release 8.1 or later requires deleting your current nodes and re-instantiating them using Heat templates that incorporate HA model changes. Deleting the nodes is service-impacting.

SBC SWe in 1:1 HA mode can be deployed either with or without an Insight EMS system in the deployment. If your deployment includes an EMS (a Direct-Single deployment), your configuration database is backed up on the EMS and can be downloaded again after you re-instantiate the nodes. Otherwise, deleting your current nodes will result in loss of your current configuration. To save and then restore your own configuration backup, refer to Saving the SBC Configuration using EMA and Restoring the SBC Configuration using EMA.

After you upgrade to release 8.1 or higher, the required database changes are in place and you can resume using the replacement-based upgrade procedure for subsequent upgrades to higher releases. Refer to Upgrading SBC SWe 1:1 HA Nodes on OpenStack using Heat Templates (from release 8.1 or later) for that procedure.

Note

For upgrades from release 7.2 to 8.x or higher, a Direct-Single SBC node can go to an unregistered state on the EMS while coming up, which pauses the upgrade. This can occur because of the conversion of the policy database from Oracle to Postgres. To continue, reboot the node to return it to a registered-online state.

Upgrade Process

Perform the following activities to upgrade an SBC SWe 1:1 HA deployment to SBC release 8.1 on OpenStack using Heat.

Upgrade the Insight EMS

If there is an EMS in your deployment, upgrade it first before re-instantiating the SBC nodes. Refer to Steps for Migrating EMS Standalone on OpenStack or Steps for Migrating GR on OpenStack. 

Download the Software Image

Download the required .QCOW2 and .sha256 files from the Customer Portal.

Upload the Software Image to OpenStack

To upload the .QCOW2 file to OpenStack, navigate to Project > Compute > Images. For more information, refer to Creating a Glance Image.

Update the Heat Template 

Refer to Instantiating SBC SWe on OpenStack using Heat Templates for information on the process to develop Heat templates and instantiate SBC nodes.  Refer to the current release notes for information on how to access the current set of sample Heat templates.

Note that as a result of HA model consolidation, in the Heat templates, the user data parameter for specifying the HA model for a deployment is called 'SbcHaMode'. For 1:1 HA deployments this must be set to '1to1'.  If you have an EMS in the deployment and want to download your prior configuration after re-instantiating the SBC nodes, be sure to specify the same EMS IP address and cluster parameters when instantiating the nodes.

Delete the Existing Nodes

Use the Horizon GUI to delete the stacks corresponding to the current nodes. 

  1. Log into the Horizon GUI.
  2. Access the list of current stacks list: Project > Orchestration > Stacks
  3. Locate the entries for the SBC nodes in the stack list.
  4. In the drop-down menu adjacent to the entries, select Delete Stack. 

Instantiate the Nodes

Use standard OpenStack Heat commands or the OpenStack Horizon dashboard to instantiate the nodes using the 8.1 software image and Heat templates. Refer to Instantiating SBC SWe Using a Heat Template for additional information.  

Note

The new SBC nodes you create in OpenStack will have a new Universal Unique Identifier (UUID) value. If a node is configured to use node-locked licensing, after re-instantiating the node you must generate a new license for the node using the new UUID.