- Objectives
- Prerequisites
- Introduction
- Management Server Functions
- Section 2.7: Solaris OE Installation Preparation on Each Cluster Node
- Section 2.8: Install the Solaris 8 Operating Environment on Each Cluster Node
- Section 2.9: Install Recommended Solaris OE Patches on Each Cluster Node
- Section 2.10: The Solaris OE Post Installation and Configuration
- Section 2.11: Configure Additional Cluster Management Services
- Appendix A: System Configuration Files
Section 2.9: Install Recommended Solaris OE Patches on Each Cluster Node
For local (manual) installations, you must ensure that the latest Solaris OE patches are installed on each cluster node, per the following recommendations.
NOTE
The Solaris OE (and patch) installation is performed only during local (manual) installations. Always verify installation is correct for each cluster node, as described below.
Key Practice: On each cluster node, install the latest Solaris OE recommended patches from the SunSolveSM program website, as described in this section. Maintaining the latest recommended Solaris OE patches assures that your system has its highest reliability.
NOTE
The method of obtaining patches through the http://sunsolve.sun.com website may deviate from the following procedure. Check with your local Sun service provider on the best method for getting the required patch clusters for your current operating environment.
If you have access to the Internet, go to http://sunsolve.sun.com, and click the Patches option on the left-hand column.
Key Practice: The SunSolve program is a contract service from Sun Enterprise Services. It is a good idea to subscribe to this service, especially if you are running a production server.
Key Practice: Create a /PATCHES directory on a dedicated server (clustadm repository) to store all of the required patches. This enables centralized patch management.
Step 2.9.1Verifying the Existence of the Patch Directory
For local (manual) installations, change the directory to where the appropriate Solaris OE patch-cluster files are stored. Following these key practices, this next example assumes that the patch repository has been created on clustadm:
clustadm# cd /JumpStart/PATCHES/SOLARIS_8/8_Recommended clustadm# ls {{verify existence of patches}}
Step 2.9.2Sharing the Patch Directory
For local (manual) installations, verify that the PATCHES directory is shared to each cluster node by entering the following command on the Sun Cluster 3.0 software administration workstation (clustadm). If the /JumpStart directory is not already shared, repeat steps from Section 2.3 (refer to the share(1M) man page).
clustadm# share - /JumpStart ro,anon=0 ""
Key Practice: Always refer to the individual patch README files to review any install prerequisites before installing patches. Using this practice could possibly prevent conflicts with other patches, software, boot-prom variables, or other unknowns.
Step 2.9.3Installing the Patches
For local (manual) installations, change the directory to the Solaris OE patches directory, and install all of the required patches on each cluster node. For example, enter the following commands on each cluster node:
# mount -F nfs clustadm:/JumpStart/PATCHES/SOLARIS_8/8_Recommended /mnt # cd /mnt # ./install_cluster Are you ready to continue with install? [y/n]: y Installing <PATCH NUMBER>... Installing <PATCH NUMBER>... ... Installing <PATCH NUMBER>...
NOTE
Some patches included in the patch cluster may fail installation with a Return code 2 error message because they were already included in the Solaris OE base software installation.
Key Practice: Review the installation log file for each patch applied (for example, /var/sadm/patch/..) to verify that it was successfully installed. Determine and resolve the cause of any patch installation error messages that occurred. Refer to the Sun Cluster 3.0 U1 Error Messages Manual for error message translations (see Appendix B for information on obtaining this manual).
Step 2.9.4Verifying the Patch Installation
For local (manual) installations, verify that the required patches are installed. For example, to list all of the patches that were applied, enter the following command into both nodes.
# patchadd -p | more
For specific installation messages, refer to the /var/sadm/patch/.. directory. Note the /var/sadm/README file when navigating the directory structure.
NOTE
Both the /usr/sbin/patchadd -p and /usr/bin/showrev -p commands displays a list of patches that have been added to the system.
Use the /usr/bin/patchadd -p | grep patch_number to check individual patches.
# patchadd -p | grep patch_number
Step 2.9.5Rebooting the Cluster Nodes
For local (manual) installations, reboot both cluster nodes by performing the next two steps in succession.
First, enter the following command into each cluster node:
# shutdown -g0 -y
Second, reboot each cluster node by entering the following command at the OpenBoot PROM prompt:
ok reset
Summary of Key Practices
Always install the latest (required) Solaris OE patches. Create a /PATCHES directory on a dedicated server to store all of the patches. This enables centralized patch management. Refer to the individual patch README files to review any install pre-requisites before installing the patches. Review the installation (and patch) log files to determine, and resolve, the cause of any software installation error messages. |