Sun Cluster Geographic Edition System Administration Guide

Applying Patches to a Sun Cluster Geographic Edition System

Observe the following guidelines and requirements to patch Sun Cluster Geographic Edition software:

ProcedureHow to Prepare a Sun Cluster Geographic Edition System for Patches

  1. Ensure that the cluster is functioning properly.

    To view the current status of the cluster, run the following command from any node:


    % cluster status
    

    See the cluster(1CL) man page for more information.

    Search the /var/adm/messages log on the same node for unresolved error messages or warning messages.

  2. Become superuser on a node of the cluster.

  3. Remove all application resource groups from protection groups.

    This step ensures that resource groups are not stopped when you later stop the protection groups.


    # geopg remove-resource-group resourcegroup protectiongroup
    

    See the geopg(1M) man page for more information.

  4. Perform the preceding steps on all clusters that have a partnership with this cluster.

  5. Stop all protection groups that are active on the cluster.


    # geopg stop -e local protectiongroup
    

    See the geopg(1M) man page for more information.

  6. Stop the Sun Cluster Geographic Edition infrastructure.


    # geoadm stop
    

    Shutting down the infrastructure ensures that a patch installation on one cluster does not affect the other cluster in the partnership.

    See the geoadm(1M) man page for more information.

  7. On each node, stop the common agent container.


    # /usr/sbin/cacaoadm stop
    

    Note –

    You must use common agent container 2, which is located in the /usr/sbin directory. Use the /usr/sbin/cacaoadm -V command to check which version of common agent container you are using.


Next Steps

Install the required patches for the Sun Cluster Geographic Edition software. Go to How to Install Patches on a Sun Cluster Geographic Edition System.

ProcedureHow to Install Patches on a Sun Cluster Geographic Edition System

Perform this procedure on all nodes of the cluster.

Patch the secondary cluster before you patch the primary cluster, to permit testing.

Before You Begin

Perform the following tasks:

  1. Ensure that all the nodes are online and part of the cluster.

    To view the current status of the cluster, run the following command from any node:


    % cluster status
    

    See the cluster(1CL) man page for more information.

    Search the /var/adm/messages log on the same node for unresolved error messages or warning messages.

  2. Become superuser on each node.

  3. On each node, install any necessary patches to support Sun Cluster Geographic Edition software by using the patchadd command.

    If you are applying Sun Cluster patches, use the Sun Cluster methods on both clusters.

  4. After you have installed all required patches on all nodes of the cluster, on each node start the common agent container.


    # /usr/sbin/cacaoadm start
    

    Note –

    You must use common agent container 2, which is located in the /usr/sbin directory. Use the /usr/sbin/cacaoadm -V command to check which version of common agent container you are using.


  5. On one node, enable Sun Cluster Geographic Edition software.


    # geoadm start
    
  6. Add all application resource groups that you removed while you were preparing the cluster for a patch installation back to the protection group.


    # geopg add-resource-group resourcegroup protectiongroup
    

    See the geopg(1M) man page for more information.

  7. Start all the protection groups that you have added.


    # geopg start -e local [-n] protectiongroup
    

    See the geopg(1M) man page for more information.

Next Steps

After you patch the secondary cluster, perform a sanity test on the Sun Cluster Geographic Edition software, and then repeat this procedure on the primary cluster.