Oracle® Database

4.1.1 Operating System Requirements

Red Hat Enterprise Linux 5 and Oracle Linux 5 on Linux x86 and Linux x86-64 require Update 2.

The documents that shipped with the media are superseded by the updates on OTN, which have updated operating system requirements.

http://www.oracle.com/technetwork/indexes/documentation/index.html

4.1.2 Swap Space Error Message

If you encounter an Oracle Universal Installer error indicating inadequate swap space size, but your swap space meets the requirements listed in the Oracle Database Installation Guide, then you can ignore this error.

This issue is tracked with Oracle bug 7700929.

4.1.3 ODBC Support for SUSE Linux Enterprise Server 11 on Linux x86

You must install the following 32-bit ODBC RPMs on Linux x86 to use ODBC on SUSE 11:

unixODBC-32bit-2.2.12 (32-bit) or later
unixODBC-devel-32bit-2.2.12 (32-bit) or later

4.1.4 ODBC Support for SUSE Linux Enterprise Server 11 on Linux x86-64

You must install the following 32-bit ODBC RPMs on Linux x86-64 to use ODBC on SUSE 11:

unixODBC-2.2.12 or later
unixODBC-devel-2.2.12 or later
unixODBC-32bit-2.2.12 (32-bit) or later 

4.1.5 Back Up grub.conf Before Installing Oracle Grid Infrastructure for a Cluster on Oracle VM

During Oracle Grid Infrastructure installation for a cluster, when rootcrs.pl installs Oracle ACFS in an unprivileged domain running a Xen or Oracle VM kernel, the base Linux driver installation routines may overwrite grub.conf with an incorrect version. Contact Oracle Support if you see the following error message:

# xm create vm.cfg
Using config file "./vm.cfg".
Error: Boot loader didn't return any data!

Workaround:

To prevent this error, follow these steps to copy /boot/grub/grub.conf to a backup, and then restore it after the Oracle ACFS installation.

1. Check for Xen kernel in a domU:

   # uname -r
   

2. Copy grub.conf to backup:

   # cp -a /boot/grub/grub.conf /boot/grub/grub.conf.bak
   

3. Install Oracle ACFS using the root script:

   # rootcrs.pl
   

4. Restore original grub.conf:

   cp -a /boot/grub/grub.conf.bak /boot/grub/grub.conf
   

This issue is tracked with Oracle bug 9268593.

4.2.1 Connection Load Balancing Issue

A single Oracle Database Client can request only a specific SCAN listener to connect to a database using SCAN and easy connect method. As a result, load balancing by the DNS round-robin is not available.

Workaround:

Connect to a database using the following configuration, specifying LOAD_BALANCE=on, in the tnsnames.ora file:

ORCL =
  (DESCRIPTION =
    (LOAD_BALANCE=on)
    (ADDRESS = (PROTOCOL = TCP)(HOST = stscan1)(PORT = 1521))
    (CONNECT_DATA =
     (SERVER = DEDICATED)
      (SERVICE_NAME = srv.world)
     )
  )
    

This issue is tracked with Oracle bug 8256753.

4.2.2 Oracle Real Application Clusters Upgrade Error When Upgrading from Release 10.1.0.5 to 11.2

When upgrading from Oracle Database release 10.1.0.5 on Linux x86-64, the following errors may occur in the Database Upgrade Assistant:

ERROR at line 1:

• ORA-00604: error occurred at recursive SQL level 3

• ORA-00060: deadlock detected while waiting for resource

Workaround: Complete the following steps before upgrading to avoid the errors:

1. Before starting the release 11.2 upgrade, purge the dba_recyclebin through SQL*Plus using the following command:

   SQL> purge dba_recyclebin;
   

2. Disable the recycle bin using the following command:

   ALTER SYSTEM set "_recyclebin"=FALSE;
   

3. Perform the upgrade.

4. Enable recycle bin after the release 11.2 upgrade using the following command:

   ALTER SESSION SET recyclebin=ON;
   

This issue is tracked with Oracle bug 8475185.

4.2.3 Error When Installing Oracle Grid Infrastructure with the Grid Naming Service (GNS) and the OCR and Voting Disks Stored on Oracle ASM

When installing Oracle Grid Infrastructure for Oracle Database 11g Release 2 with GNS and the OCR and voting disks on Oracle ASM on Linux x86-64, root.sh may fail with error:

PROT-1: Failed to initialize ocrconfig

The failing command is ocrconfig -upgrade user group, and TNS-12560 can be found in sqlnet log file.

Workaround:

Check the /etc/group entry for the group being used by OCR, temporarily reduce the number of users listed for this group, and restart the installation.

This issue is tracked with Oracle bug 8979500.

4.2.4 Cluster Verification Utility Issue During Oracle Clusterware Upgrade

While upgrading Oracle Clusterware of a previous release to Oracle Grid Infrastructure 11g Release 2 (11.2.0.1.0), Cluster Verification Utility (CVU) may fail if the permission and ownership on the raw devices configured with Oracle Cluster Registry and Voting Disks are incorrect.

Workaround:

Check CVU's output in the installation log file and fix the permissions and ownership on the raw devices. Then, click Retry to run CVU again.

This issue is tracked with Oracle bug 8737732.

4.2.5 RPM Errors with Oracle VM

When Oracle Database is installed on a Linux environment configured with Oracle VM, Oracle Universal Installer may show package check failures, and the RPM command returns the following error:

# rpm
Floating point exception

Workaround:

Upgrade the RPM package to 4.4.2.3 or later.

This issue is tracked with Oracle bug 8626425.

4.2.6 C++ XDK Demos and Pro*C++ Static Demos Linking Error

The C++ XDK demos and Pro*C++ Static demos fail to link with the -lstdc++ library giving the following error on Linux x86-64 with SUSE Linux Enterprise Server 10, SUSE Linux Enterprise Server 11, and Red Hat Enterprise Linux 5:

/usr/lib64/gcc/x86_64-suse-linux/4.1.0/../../../..
/x86_64-suse-linux/bin/ld: cannot find -lstdc++_nonshared

Workaround:

Contact your operating system vendor.

This issue is tracked with Oracle bug 8396301.

4.2.7 Oracle RAC Database Installation Error on SUSE Linux Enterprise Server 11

The following error occurs while performing an Oracle RAC database installation on SUSE Linux Enterprise Server 11:

Error securing Database Control, Database Control has not been brought up on nodes
remote_node(s). Please execute the following command(s) on nodes: remote_node(s).

1) Set the environment variable ORACLE_UNQNAME to Database unique name 
2) /u01/app/aime/product/11.2.0/racdb/bin/emctl config emkey -repos-sysman_pwd Password_for_SYSMAN_user -host node
-sid Database_unique_name
3) /u01/app/aime/product/11.2.0/racdb/bin/emctl secure dbconsole -sysman_pwd Password_for_SYSMAN_user -host node
-sid Database_unique_name 
4) /u01/app/aime/product/11.2.0/racdb/bin/emctl start dbconsole

To secure Em Key, run /u01/app/aime/product/11.2.0/racdb/bin/emctl configemkey -remove_from_repos -sysman_pwd
Password_for_SYSMAN_user

Workaround:

Follow the instructions shown in the error message.

This issue is tracked with Oracle bug 8680497.

4.2.8 Error When Installing Oracle RAC Database on Oracle Linux 5.0 Update 3

When installing Oracle RAC database on Oracle Linux 5 Update 3, the Perform Prerequisite Checks screen reports Node Connectivity failure.

Workaround:

Check Ignore All to ignore this error and proceed with the installation.

This issue is tracked with Oracle bug 8753154.

4.2.9 Oracle Grid Infrastructure for a Standalone Server Installation Error on SUSE Linux Enterprise Server 10

The root.sh script displays the following error when installing Oracle Grid Infrastructure for a standalone server on SUSE Linux Enterprise Server 10:

/bin/env: Command not found

Workaround:

Ignore this error and proceed with the installation.

This issue is tracked with Oracle bug 8732857.

4.2.10 Oracle Database Reupgrade Issue on Linux x86-64

When you reupgrade Oracle Database release 10.2 to release 11.2, after an initial upgrade to release 11.2 and a subsequent downgrade to release 10.2, the upgrade process hangs at 80% with Database Configuration Assistant. For manual upgrades, the hang may be seen in the spooled log while the c1101000.sql upgrade script is running.

This issue is tracked with Oracle bug 8742500.

4.3.1 Error While Loading Shared Library When SELinux is Enforcing on Oracle Linux 5.x and Red Hat Enterprise Linux 5.x

SQL*Plus and Oracle Call Interface (OCI) program calls fail with SELinux in the Enforcing mode on Oracle Linux 5.x and Red Hat Enterprise Linux 5.x.

Workaround: Shift SELinux to Permissive mode on the system.

This issue is tracked with Oracle bug 9817031.

4.3.2 Insufficient Memory Target Errors

On Linux systems, if the operating system /dev/shm mount size is too small for the Oracle system global area (SGA) and program global area (PGA), then you encounter the following error:

ORA-00845: MEMORY_TARGET not supported on this system. 

The cause of this error is an insufficient /dev/shm allocation. The total memory size of the SGA and PGA, which sets the initialization parameter MEMORY_TARGET or MEMORY_MAX_TARGET, cannot be greater than the shared memory file system (/dev/shm) on your operating system.

Background

Automatic Memory Management (AMM) has been updated in Oracle ASM 11g Release 2. It manages both the SGA and PGA together. It is managed by the Memory Manager Process (MMAN). In this release, note the following changes to AMM:

• It uses MEMORY_TARGET instead of SGA_TARGET

• It uses MEMORY_MAX_TARGET instead of SGA_MAX_SIZE (defaults to MEMORY_TARGET)

• It uses memory allocated by /dev/shm

If the value of max_target is set to a value greater than the allocation for the /dev/shm size, then you may encounter the error ORA-00845: MEMORY_TARGET not supported on this system.

$ df -k 
 
Filesystem 1K-blocks Used Available Use% Mounted on 
shmfs 6291456 832356 5459100 14% /dev/shm

Solution

Increase the /dev/shm mountpoint size.

For example:

# mount -t tmpfs shmfs -o size=7g /dev/shm

To make this change persistent across system restarts, add an entry in /etc/fstab similar to the following:

shmfs /dev/shm tmpfs size=7g 0

4.3.3 Help Files Translation Issue for Spanish and French

Oracle Universal Installer does not copy the translated help files to Oracle home directory when you select the Spanish or French language.

Workaround:

On the Select Product Languages page, select both the Latin American Spanish and Spanish to get the help files translated in Spanish; and select both the Canadian French and French for the translated help files in French.

This issue is tracked with Oracle bug 9147106.

4.3.4 Incorrect Reverse Ping Command from Oracle Management Service

Oracle Management Service provides incorrect reverse ping command with Oracle Enterprise Manager Database Control. The emoms.properties file shows:

emdrep.ping.pingCommand=/usr/sbin/ping6 <hostname>

Workaround:

Change the property in the emoms.properties file to:

emdrep.ping.pingCommand=/bin/ping -c 3 -w 30 <hostname>

This issue is tracked with Oracle bug 9211800.

5.1.1 Oracle Unbreakable Enterprise Kernel for Linux

Oracle offers two Linux x86-64 kernels for Oracle Linux 5 Update 5, or Red Hat Enterprise Linux 5 Update 5:

• The Unbreakable Enterprise Kernel optimized for running Oracle software and hardware

• The Red Hat compatible kernel for strict Red Hat compatibility

5.1.2 Multiple Private Interconnects and Oracle Linux

Due to a Reverse Path Filtering bug fix, Oracle RAC and Oracle Exadata systems running kernel 2.6.31 and later require specific settings for the rp_filter parameter. When using multiple NICs for a private interconnect, the interconnect packets may be blocked or discarded if the rp_filter parameter is not set.

Set the rp_filter value for the private interconnects to either 0 (no filtering) or 2 (loose filtering). Do not set the private interconnect NIC to 1 (strict filtering) as it may cause connection issues on the private interconnect. For example, where eth1 and eth2 are the private interconnect NICs, and eth0 is the public network NIC, set the rp_filter of the private address to 2 (loose filtering), the public address to 1 (strict filtering), using the following entries in /etc/sysctl.conf :

net.ipv4.conf.eth2.rp_filter = 2
net.ipv4.conf.eth1.rp_filter = 2
net.ipv4.conf.eth0.rp_filter = 1

Oracle Linux 5.6 (Enterprise Linux 5 Update 6) includes a fix using initscripts-8.45.33-1.0.4.el5.i386.rpm, which sets the kernel parameter net.ipv4.conf.default.rp_filter to 2 (relaxed mode). After you apply Unbreakable Linux Kernel on top of Oracle Linux 5.6, you may not need to make manual changes, because the rp_filter value of all the NICs is set to 2. If you require strict reverse path filtering on the public network, then set the public NIC rp_filter to 1.

https://support.oracle.com

5.2.1 JDBC Connection Timeout Errors

On some servers, applications can encounter JDBC connection timeouts, due to errors with JVM attempting to access /tmp and reading the /dev/random device. This error may have multiple causes relating to hardware or operating system issues.

Workaround:

Contact an Oracle Support Engineer for further information.

This issue is tracked with Oracle bug 6485149.

5.2.2 Oracle Universal Installer Hangs During Oracle Grid Infrastructure Installation on SUSE 10

Due to an issue with the SUSE Enterprise Linux 10 operating system, Oracle Universal Installer may experience intermittent hangs of five to six minutes during Oracle Grid Infrastructure for a Cluster installation.

Workaround:

None. Wait for the hang to resolve, and proceed with the installation.

This issue is tracked with bug 9949520.

5.2.3 Oracle9i Real Application Clusters Upgrade Issue

When using Database Upgrade Assistant to upgrade an Oracle9i Real Application Clusters database to Oracle Database 11g Release 2, the upgrade fails if the Oracle Grid Infrastructure home is non-shared and was configured using the Oracle Grid Infrastructure Configuration Wizard.

Workaround:

1. As a user with root privileges, use the following commands to copy the libcmdll.so and libwddapi.so files to the remote nodes:

   scp grid_user@node1:/etc/ORCLcluster/oracm/lib/libcmdll.so
    /etc/ORCLcluster/oracm/lib
   scp grid_user@node1:etc/ORCLcluster/oracm/lib/libwddapi.so
    /etc/ORCLcluster/oracm/lib
   

   Enter the grid_user password when prompted.

2. Use Database Upgrade Assistant to upgrade Oracle RAC.

This issue is tracked with Oracle bug 10029705.

5.2.4 Missing Events After an Oracle Clusterware Upgrade

If you upgrade Oracle Clusterware or Oracle Grid Infrastructure for a cluster to 11g Release 2 (11.2.0.2), using a rolling upgrade, then Oracle Clusterware may not publish fast application notification (FAN) events and Oracle Enterprise Manager cluster management events on the upgraded clusters.

Workaround:

Perform a rolling restart of the Oracle Clusterware software by running the crsctl stop crs and crsctl start crs commands on each cluster member node in sequence.

This issue is tracked with Oracle bug 10015238.

5.2.5 Installing Predownloaded Software Updates

By default, software updates are download to the /tmp directory. This directory is deleted at the end of an installation. If you do not download and install the software updates during the same installation session, then the software updates are not available.

Workaround:

Specify a new download location for the software updates on the Download Software Updates screen.

This issue is tracked with Oracle bug 9975999.

5.2.6 Square Boxes Appear When Installing on Korean Locale on Red Hat Enterprise Linux 5

When installing on a Korean locale (ko_KR.utf8 or ko_KR.euckr), the characters on the Oracle Universal Installer screens are shown as square boxes.

Workaround:

Create the following symbolic link:

# cd /usr/share/fonts
# ln -s korean ko

This issue is tracked with Oracle bug 9746773.

5.2.7 Required Oracle Grid Infrastructure Patch Set Updates

If you prepare to upgrade Oracle Grid Infrastructure release 11.2.0.1 on Linux x86 and Linux x86_64 by applying patch for bug 9413827, then the rootupgrade.sh script for Oracle Grid Infrastructure release 11.2.0.2 displays the following error:

The fixes for bug 9413827 and 9706490 are not present in the 11.2.0.1 crs home.
Apply the patches for these bugs in the 11.2.0.1 crs home and then run 
rootupgrade.sh

Workaround

Apply patch 9655006 to the 11.2.0.1 Grid home before upgrading to 11.2.0.2.

This issue is tracked with Oracle bug 10036834.

5.2.8 Oracle Database Vault Administrator Login Issues

While trying to log in to Oracle Database Vault Administrator, you may get the following error:

java.security.NoSuchProviderException: no such provider: SUN 

Workaround:

Use any one of the following workarounds to configure Database Vault policy:

• Connect to the 11.2.0.2 Linux Database Vault enabled database using the 11.2.0.1 Oracle Database Vault Administrator from any platform.

• Use Database Vault administrative packages.

This issue is tracked with Oracle bug 10188865.

5.3.1 Oracle Notification Service Issue Associated With Physical Memory

The Oracle Notification Service server may occasionally fail on Linux x86-64 systems with 32 GB of physical memory. The Oracle Notification Service server is immediately restarted, and no action is required of an administrator.

This issue is tracked with Oracle bug 9203989.

5.3.2 SQL Developer Fails to Start on IBM: Linux on System Z

SQL Developer might fail to start and the following error message appears:

Type the full pathname of a J2SE installation (or Ctrl-C to quit), the path
will be stored in ~/.sqldeveloper/jdk

Workaround:

Download the latest version of SQL Developer from the Oracle Technology Network (OTN) Web site and start SQL Developer pointing to the appropriate JRE:

http://www.oracle.com/technetwork/developer-tools/sql-developer/overview/index.html

This issue is tracked with Oracle bug 10339897.

5.3.3 Required Oracle Warehouse Builder Patches on IBM: Linux on System Z

If you plan to install Oracle Warehouse Builder, then after you install Oracle Database, but before you create the Oracle Warehouse Builder repository, you must complete the following procedure:

1. Download patch 12326108 from the My Oracle Support Web site.

2. Extract the contents to a directory, for example, /tmp/12326108.

3. Verify that the ORACLE_HOME environment variable is set to the 11.2.0.2 DB home.

4. Run the Opatch utility with the right parameters to proceed with the Oracle Warehouse Builder one-off installation:

   cd /tmp/12326108
   opatch apply
   

5. Navigate to $ORACLE_HOME/owb/patches/bug12326108.

6. Run the script, fix_bug12326108.sh.

7. Edit the following files:

   OMBPlus.sh

   Add the following line at line 15, under the s390 entry:

   s390x) OPTS="-Xdisableexplicitgc";;
   

   Add the -d64 entry to the following line:

   $JAVAPATH/bin/java $JAVA64FLAG -Xms64M -Xmx768M...
   

   reposinst.sh

   Add the -d64 entry to the following line:

   $JAVAPATH/bin/java $JAVA64FLAG -Xms128m -Xmx1024m...
   

   run_service.sh.orig

   Add the -d64 entry to the following line:

   $JAVAPATH/bin/java $JAVA64FLAG -Xmx768M...
   

   setowbenv.sh

   Add $ORACLE_HOME/lib to the clsspath line:

   CLASSPATH_LAUNCHER="-classpath ../admin:../admin/launcher.jar:$ORACLE_HOME/lib:$CLASSPATH:...
   

8. After you have edited the files, you can install Oracle Warehouse Builder.

This issue is tracked with Oracle bugs 11772716 and 11934843.

6.2.1 Square Boxes Appear When Installing on Korean Locale on Red Hat Enterprise Linux 5

When installing on a Korean locale (ko_KR.utf8 or ko_KR.euckr), the characters on the Oracle Universal Installer screens are shown as square boxes.

Workaround:

Create the following symbolic link:

# cd /usr/share/fonts
# ln -s korean ko

This issue is tracked with Oracle bug 9746773.

6.2.2 JDBC Connection Timeout Errors

On some servers, applications can encounter JDBC connection timeouts, due to errors with JVM attempting to access /tmp and reading the /dev/random device. This error may have multiple causes relating to hardware or operating system issues.

Workaround:

Contact an Oracle Support Engineer for further information.

This issue is tracked with Oracle bug 6485149.

6.2.3 Offline Services After Completing a Shared Home Upgrade

After upgrading a shared home Oracle Clusterware from an earlier release to Oracle Clusterware release 11.2.0.3, you may encounter offline service errors.

Workaround:

Use the command crsctl start res from the Grid_home/bin directory to start offline service resources. For example, if the service resource that is offline is ora.orcldb.srv2.orcldb4.srv, then enter the following command:

# cd /u01/app/grid/11203/bin
# ./crsctl start res ora.orcldb.srv2.orcldb4.srv

This issue is tracked with Oracle bug 12844130.

6.3.1 Time Zone Issue with Oracle Linux 6

When Oracle Grid Infrastructure is installed on Oracle Linux 6, the logs may show incorrect time zone information if the system Timezone string contains a space. This can be checked through the /etc/sysconfig/clock file.

Workaround:

You must use the following workaround on the local node (the node on which Oracle Universal Installer is running) of a cluster before starting Oracle Grid Infrastructure installation:

Modify the /etc/sysconfig/clock file to replace any space with an underscore (_) in the time zone name. For example, Los_Angeles instead of Los Angeles.

Oracle recommends that you apply this workaround on all the nodes of a cluster so that if Oracle Universal Installer is run at a later time from any other node, the time zone is correctly calculated.

6.3.2 Java-Based Tools Issue on Oracle Linux 6 and Red Hat Enterprise Linux 6

Java-based tools such as Oracle Universal Installer, Oracle Database Configuration Assistant, and Oracle Enterprise Manager Configuration Assistant cannot display non-English multi-byte character language messages on Oracle Linux 6 (32-bit and 64-bit) and Red Hat Enterprise Linux 6. For example, languages like Japanese, Chinese, Korean, and so on, are not displayed by the Java-based tools on Oracle Linux 6 (32-bit and 64-bit) and Red Hat Enterprise Linux 6.

Workaround:

Download patch 14712133 from the My Oracle Support web site and follow the instructions in the README file of the patch.

This issue is tracked with Oracle bug 12991286.

6.3.3 Issues While Running Asynchronous I/O on Red Hat Enterprise Linux 5 Update 6

If you are using Red Hat Enterprise Linux 5 Update 6 or later with kernel version 2.6.18-238 or later, then the I/O's to Oracle Cluster File System 2 may hang due to a Red Hat kernel bug. This issue does not occur with Oracle Linux kernels.

Workaround:

Contact Red Hat for a fix.

This issue is tracked with Oracle bug 12725273.

6.3.4 Required Oracle Warehouse Builder Patches on IBM: Linux on System Z

If you plan to install Oracle Warehouse Builder, then after you install Oracle Database, but before you create the Oracle Warehouse Builder repository, you must complete the following procedure:

1. Download patch 12973496 from the My Oracle Support Web site.

2. Extract the contents to a directory, for example, /tmp/12973496.

3. Verify that the ORACLE_HOME environment variable is set to the 11.2.0.2 DB home.

4. Run the Opatch utility with the right parameters to proceed with the Oracle Warehouse Builder one-off installation:

   cd /tmp/12973496
   opatch apply
   

5. Navigate to $ORACLE_HOME/owb/patches/bug12973496.

6. Run the script, fix_bug12973496.sh.

7. After you have edited the files, you can install Oracle Warehouse Builder.

This issue is tracked with Oracle bugs 11772716.