How to perform a Split Pair upgrade on 2.5 TPAM appliances (4235463)

Follow the instructions below if you would like to upgrade the Primary and Replica appliances separately.  Doing so will reduce downtime and ensure that one appliance is available while the other is being upgraded.

From the /admin interface of the Primary Appliance:

1. Download a backup from the Primary appliance

- Go to the Backup menu and select Manage Online Backups
- Select the newest backup at the top and click the Download button.
- If you would like to perform a fresh backup, click the Backup Now button
- Refer to the Backup Log and wait for this backup to complete.
- Once it is complete, refresh the list, select the latest backup and click Download.
- Save it locally on your system or to a network file share.

2. Change the Primary to "Maintenance" mode

- Go to System Status/Settings and select Cluster Management
- Select the Primary appliance in the Cluster Members list
- Check the box next to "Run Level" and select "Maintenance" from the drop down
- Click the Change Run Level button
- Click Continue with Change in the confirmation window

3. Force a failover on the Replica to stop replication

- Go to System Status/Settings and select Cluster Management
- Select the Replica appliance in the Cluster Members list
- Click the Force Failover button
- Click Continue with Change in the confirmation window
- Verify that the Replica status has changed to "FailedOver" on the "Cluster Status" tab.

Please note: Users will need to go to the address of the Replica to login to its /tpam interface.  While the Replica is failed over, users can submit requests and retrieve passwords.  Passwords will need to be reset manually.  The Automation Engine will not start on a Failed Over Replica and new Systems or Accounts cannot be added.

4. Reboot the Primary appliance

- Click the Shutdown menu item
- Ensure the "Restart?" box is checked
- Click the Shutdown button
- Close your browser. Wait 5-10 minutes for the reboot to complete.

5. Apply the upgrade patch

- Select the Maint menu and click Apply a Patch
- Click the Select File button and browse for the patch, i.e., paa-console-2.5.910.zip.
- Click the Upload button. Do not navigate away from this page while it is uploading.
- Enter the key and option for this patch into the key and option fields.
- In the Target Replicas list, Ctrl + Click on the Replica's name to deselect it (the Replica is selected by default. In this case, we do not want it to be upgraded automatically.)  Note that if you have multiple replicas, you may wish to push the patch to the remaining (but not the failed over) replicas.  Note as well if you have multiple replicas you will still need to adjust the Failover Timer as per the Release Notes Installation Instructions.
- Click the Apply Patch button.

After the patch has started, refer to the Patch Log to monitor the progress.  The log may be a few minutes behind.  If it seems to stall at any point, allow some time for it to catch up.  The upgrade will be complete when you see the line, "Patch successfully applied to the system".

6. Change the Primary back to Operational mode

- Go to System Status/Settings and select Cluster Management
- Select the Primary appliance in the Cluster Members list
- Check the box next to "Run Level" and select "Operational" from the drop down
- Click the Change Run Level button
- Click Continue with Change in the confirmation window

If all has gone well with the Primary upgrade, it can now be put back into service while the Replica is upgraded.  The process to upgrade the Replica will be the same as the Primary.  Repeat the steps above:

7. Apply the patch to the Replica

8.  Restore Replication

Once both appliances have been upgraded to the same version, you can re-establish the replication.

- Login to the /admin interface on the Primary
- Go to System Status/Settings and select Cluster Management
- Select the Replica appliance in the Cluster Members list
** During the Un-Force Failover, the primary may be inaccessible until the process is completed.
- Click the Un-Force Failover button
- Click Continue with Change in the confirmation window
- The Primary will attempt to sync with the Replica.  Monitor the Cluster Log. 
- Once everything is up to date, the Cluster Status should show as follows:
Primary: Replicating / Operational / Healthy
Replica: Standby / Operational / Healthy

If you experience any problems during the upgrade, please collect the Patch Log and the Cluster Log and submit to Support for review.