How to Upgrade the Software (Dual-Partition)

Language:

Perform this procedure to upgrade each node of the cluster to Oracle Solaris Cluster 4.3 software and, if necessary, upgrade the Oracle Solaris software.

Note - If you intend to upgrade failover zones, do not use the dual-partition upgrade method. Instead, use the standard upgrade or rolling upgrade method.

Perform all steps from the global zone only.

Before You Begin

Perform the following tasks:

Ensure that all steps in How to Prepare the Cluster for Upgrade (Dual-Partition) are completed.
Ensure that the node you are upgrading belongs to the partition that is not active in the cluster and that the node is in noncluster mode.

Tip - You can use the pconsole utility to perform this procedure on multiple nodes simultaneously. See the pconsole(1) man page that is installed with the terminal/pconsole package for more information.

Assume the root role on a node that is a member of the partition that is in noncluster mode.
Subscribe to the ha-cluster publisher that contains the software you want to upgrade to.
```
# pkg set-publisher -G '*' -g URL_for_ha-cluster_publisher ha-cluster
```
Ensure that the solaris publisher is valid.
```
# pkg publisher
PUBLISHER                           TYPE     STATUS   P  LOCATION
solaris                             origin   online   F  solaris-repository
```
For information about setting the solaris publisher, see Adding, Modifying, or Removing Package Publishers in Adding and Updating Software in Oracle Solaris 11.3.
Start the scinstall utility.
```
phys-schost# scinstall
```
The scinstall Main Menu is displayed.
Choose the menu item, Upgrade This Cluster Node.
The Upgrade Menu is displayed.
Follow the menu prompts to upgrade the cluster framework.
Upgrade processing is finished when the system displays the message Completed Oracle Solaris Cluster framework upgrade and prompts you to press Enter to continue.
Quit the scinstall utility.
If you have HA for NFS configured on a highly available local file system, ensure that the loopback file system (LOFS) is disabled.
Note - If you have non-global zones configured, LOFS must remain enabled. For guidelines about using LOFS and alternatives to disabling it, see Planning Cluster File Systems in Oracle Solaris Cluster 4.3 Software Installation Guide.

To disable LOFS, ensure that the /etc/system file contains the following entry:
```
exclude:lofs
```
This change becomes effective at the next system reboot.
As needed, manually upgrade any custom data services that are not supplied on the product media.
Verify that each data-service update is installed successfully.
View the upgrade log file that is referenced at the end of the upgrade output messages.
Install any Oracle Solaris Cluster 4.3 framework and data-service software updates.
For instructions on updating your software, see Chapter 11, Updating Your Software in Oracle Solaris Cluster 4.3 System Administration Guide.
Upgrade software applications that are installed on the cluster.
Ensure that application levels are compatible with the current versions of Oracle Solaris Cluster and Oracle Solaris software. See your application documentation for installation instructions.

Note - If any upgrade procedure instructs you to perform a reboot, you must add the –x option to the boot command. This option boots the cluster into noncluster mode.

Repeat all steps in this procedure up to this point on all remaining nodes that you need to upgrade in the partition.

After all nodes in a partition are upgraded, apply the upgrade changes.

From one node in the partition that you are upgrading, start the interactive scinstall utility.
```
phys-schost# scinstall
```
The scinstall Main Menu is displayed.
Type option number for Apply Dual-Partition Upgrade Changes to the Partition.

Follow the prompts to continue each stage of the upgrade processing.

The command performs the following tasks, depending on which partition the command is run from:

First partition - The command halts each node in the second partition, one node at a time. When a node in the second partition is halted, any services on that node are automatically switched over to a node in the first partition, provided that the node list of the related resource group contains a node in the first partition. After all nodes in the second partition are halted, the nodes in the first partition are booted into cluster mode and take over providing cluster services.

Caution - Do not reboot any node of the first partition again until after the upgrade is completed on all nodes. If you again reboot a node of the first partition before the second partition is upgraded and rebooted into the cluster, the upgrade might fail in an unrecoverable state.

Second partition - The command boots the nodes in the second partition into cluster mode, to join the active cluster that was formed by the first partition. After all nodes have rejoined the cluster, the command performs final processing and reports on the status of the upgrade.

Exit the scinstall utility, if it is still running.

If you are finishing upgrade of the first partition, perform the following substeps to prepare the second partition for upgrade.
Otherwise, if you are finishing upgrade of the second partition, proceed to Step 16.
1. Boot each node in the second partition into noncluster mode.
  - SPARC:
```
ok boot -x
```
  - x86:
    1. In the GRUB menu, use the arrow keys to select the appropriate Oracle Solaris entry and type e to edit its commands.
      For more information about GRUB based booting, see Booting a System in Booting and Shutting Down Oracle Solaris 11.3 Systems.
    2. In the boot parameters screen, use the arrow keys to select the kernel entry and type e to edit the entry.
    3. Add -x to the multiboot command to specify that the system boot into noncluster mode.
    4. Press Enter to accept the change and return to the boot parameters screen.
      The screen displays the edited command.
    5. Type b to boot the node into noncluster mode.
      Note - This change to the kernel boot parameter command does not persist over the system boot. The next time you reboot the node, it will boot into cluster mode. To boot into noncluster mode instead, perform these steps to again add the –x option to the kernel boot parameter command.
2. Upgrade the nodes in the second partition.
  Return to Step 1.

If you changed the RG_system property of any resource groups to FALSE, change the settings back to TRUE.
```
phys-schost# clresourcegroup set -p RG_system=TRUE resourcegroup
```

Next Steps

Go to Completing the Upgrade.

Troubleshooting

If you experience an unrecoverable error during dual-partition upgrade, perform recovery procedures in How to Recover from a Failed Dual-Partition Upgrade.