Oracle® Database

5.1 Latest Upgrade Information

For late-breaking updates and best practices about preupgrade, post-upgrade, compatibility, and interoperability discussions, refer to Note 466181.1 on OracleMetaLink (https://metalink.oracle.com/) that links to "The Upgrade Companion" Web site.

5.2 Error in Copying Files While Upgrading Oracle Database 10g Release 2 to Oracle Database 11g

When you start upgrading second node from Oracle Database 10g Release 2 to Oracle Database 11g, an error message is displayed while copying files during the upgrade process.

Workaround: Run the upgrade session from the node on which you initially installed Oracle RAC.

This issue is tracked with Oracle bug 6501871.

5.3 Restarting the Listener for the Remote Node Displays an Error

Oracle database instance and listeners are not started after the installation is complete if you perform any of the following tasks:

• Oracle 11g Database installation or upgrade on a cluster into an existing ORACLE_BASE directory that has other Oracle home(s)

• Custom installation to install additional components in an existing Oracle RAC database home on Oracle Clusterware

You can run the Oracle_home/bin/srvctl command to display the status of Oracle database instance and listeners. When you try to start the listener for the remote node, the following error message is displayed:

TNS-01151: Missing listener name, LISTENER_NODE_NAME, in LISTENER.ORA

Workaround: Complete the following procedure for the remote node(s) on which the listener displayed an error message:

1. Note the list of nodes on which the additional component installation is being performed. Write down the list of nodes selected in the Node Selection Page in Oracle Universal Installer.

2. For each node selected in step 1, use the following command to backup Oracle_home/network/admin to Oracle_home/network/admin.orig before proceeding with the installation:

   $ cp -R Oracle_home/network/admin Oracle_home/network/admin.orig
   

3. For each node selected in step 1, use the following command to restore Oracle_home/network/admin.orig to Oracle_home/network/admin after completing the installation:

   $ cp -R Oracle_home/network/admin.orig Oracle_home/network/admin
   

4. Run the following command on local node for each nodes selected in step 1:

   $ Oracle_home/bin/srvctl start listener -n node_name
   

5. If the Oracle Database instance for the remote node is down because of the listener, then start the Oracle database an instances that are down using the following command:

   $ Oracle_home/bin/srvctl start databse -d database_name
   

These issues are tracked with Oracle bugs 6412150 and 6612571.

5.4 The /etc/oratab File Lists Multiple Automatic Storage Management Instances

If you install Oracle Clusterware on share Oracle home, the /etc/oratab file lists multiple Automatic Storage instances.

Workaround: Oracle Universal Installer detects this issue during the pre-requisition checks. You can either edit the /etc/oratab file or ignore the error during installation.

The issue is tracked with Oracle bug 5942123.

5.5 Rootpre_aix.zip file location

If you have downloaded Oracle Database 11g release 1 prior to January 18, 2008 from Oracle Technology Network or media and were prompted to execute the rootpre.sh script, then you must install the rootpre_aix.zip file available on Oracle Technology Network:

http://download.oracle.com/otn/aix/oracle11g/rootpre_aix.zip

New customers do not have to download the rootpre_aix.zip file. You can directly download the Oracle Database 11g release 1 software available on Oracle Technology Network:

http://www.oracle.com/technology/software/products/database/oracle11g/111060_aixsoft.html

6.1 Intermittent Errors While Using Oracle 11g Release 1 Direct NFS Client

When using Oracle 11g Release 1 Direct NFS client, the ORA-494 errors are displayed intermittently under certain load conditions.

Workaround: Oracle recommends that you apply the patch mentioned in the Oracle bug 6530092 to Oracle Database 11g Release 1 installation. The Oracle bug contains the details of the patch and application instructions.

This issue is tracked with Oracle bug 6530092.

6.2 Performance Graphs are not Shown in the Memory Access Mode

Monitoring in Memory Access Mode is not supported for this release on this platform. This option is disabled by default. If you enable this option, then the performance charts are not displayed and an error connection to collector memory is displayed at times.

Workaround: Use SQL Mode monitoring instead of Memory Access Mode.

This issue is tracked with Oracle bug 6498439.

6.3 Reactivate the OC4JADMIN User for Oracle Ultra Search

The OC4JADMIN user is deactivated during the Oracle Database 11g installation process. After the installation is complete, you must modify the $ORACLE_HOME/oc4j/j2ee/OC4J_SEARCH/config/system-jazn-data.xml file. Replace the following content of the file:

<user deactivated="true">
<name>oc4jadmin</name>
<display-name>OC4J Administrator</display-name>
<description>OC4J Administrator</description>

with:

<user>
<name>oc4jadmin</name>
<display-name>OC4J Administrator</display-name>
<description>OC4J Administrator</description>

Use the following commands to restart the Oracle Ultra Search middle tier in the Oracle Database:

$ $ORACLE_HOME/bin/searchctl stop
$ $ORACLE_HOME/bin/searchctl start

6.4 Oracle Clusterware Node Addition in Vendor Cluster Environment

When adding a node to the Oracle Clusterware in a vendor cluster environment that uses host names with fully qualified domain names, the rootaddnode.sh utility displays the following error message:

PRKO-2006 : Invalid node name: NODE_NAME

Workaround: For each node that is being added to the Oracle Clusterware, run the following command:

srvctl add nodeapps -n node_name_without_fully_qualified_ domain_name -A node-vip/netmask/Interfaces

CRS_NEW_NODE_NAME_LIST=
CRS_NEW_NODEVIPS=
NETMASK=
NETIFs=

This issue is tracked with Oracle bug 6617499.