8i | 9i | 10g | 11g | 12c | Misc | PL/SQL | SQL | RAC | Linux

Home » Articles » 10g » Here

Oracle Application Server 10g (9.0.4) Cold Failover Cluster Infrastructure On Red Hat Enterprise Linux 3 (RHEL3)

In this article I'll describe the installation of Oracle Application Server 10g (9.0.4) Cold Failover Cluster Infrastructure on Red Hat Enterprise Linux 3 (RHEL3). The article assumes you've performed the standard advanced server installation including the development tools, and have set up shared disks and a cluster service associated with the shared disks.

At the time of writing this method is only documented for 9iAS on Sun Cluster with an amendments for Linux, see 9ias_cfc.pdf.

Download Software

Download the following software.

Unpack Files

First unzip the files.

gunzip ias904_linux_disk1.cpio.gz
gunzip ias904_linux_disk2.cpio.gz
gunzip ias904_linux_disk3.cpio.gz
gunzip ias904_linux_disk4.cpio.gz

Next unpack the contents of the files.

cpio -idmv < ias904_linux_disk1.cpio
cpio -idmv < ias904_linux_disk2.cpio
cpio -idmv < ias904_linux_disk3.cpio
cpio -idmv < ias904_linux_disk4.cpio

You should now have four directories (Disk1, Disk2, Disk3 & Disk4) containing installation files.

Hosts File

The "/etc/hosts" file on each node must contain qualified names for all servers, a cluster alias and the heartbeat IP addresses.

# Do not remove the following line, or various programs
# that require network functionality will fail.	localhost.localdomain	localhost	infra01.mydomain.com	infra01	infra02.mydomain.com	infra02	clust01.mydomain.com	clust01
# Heartbeat IPs
#	infra01-int.mydomain.com	infra01-int	infra02-int.mydomain.com	infra02-int

Set Kernel Parameters

Add the following lines to the "/etc/sysctl.conf" file on each node.

kernel.shmall = 2097152
kernel.shmmax = 2147483648
kernel.shmmni = 142
# semaphores: semmsl, semmns, semopm, semmni
kernel.sem = 256 32000 100 142
fs.file-max = 131072
net.ipv4.ip_local_port_range = 20000 65000
kernel.msgmni = 2878 
kernel.msgmax = 8192 
kernel.msgmnb = 65535

Run the following command to change the current kernel parameters.

/sbin/sysctl -p

Add the following lines to the "/etc/security/limits.conf" file on each node.

*               soft    nproc   16384
*               hard    nproc   16384
*               soft    nofile  65536
*               hard    nofile  65536

Add the following line to the "/etc/pam.d/login" file on each node, if it does not already exist.

session    required     pam_limits.so

The "/etc/services" file should be modified to remove any references to ldap and ldaps. These references will prevent the installer using the default ldap and ldaps ports (389, 636 respectively).

Ensure the the appropriate DNS entries are present for the nodes and the cluster alias. In addition make sure the IP address of the cluster alias (the logical IP address) is associated with the appropriate cluster service. To check this do the following.

If no locical IP address is associated with the service do the following.


Install the following packages on each node.

# From RedHat AS3 Disk 2
cd /mnt/cdrom/RedHat/RPMS
rpm -Uvh setarch-1.3-1.i386.rpm
rpm -Uvh sysstat-4.0.7-4.i386.rpm

# From RedHat AS3 Disk 3
cd /mnt/cdrom/RedHat/RPMS
rpm -Uvh openmotif21-2.1.30-8.i386.rpm
rpm -Uvh ORBit-0.5.17-10.4.i386.rpm
rpm -Uvh libpng10-1.0.13-8.i386.rpm
rpm -Uvh gnome-libs-
rpm -Uvh compat-glibc-7.x- 

Put gcc296 and g++296 first in $PATH variable by creating the following symbolic links on each node.

mv /usr/bin/gcc /usr/bin/gcc323
mv /usr/bin/g++ /usr/bin/g++323
ln -s /usr/bin/gcc296 /usr/bin/gcc
ln -s /usr/bin/g++296 /usr/bin/g++

Install the 3006854 patch on each node.

unzip p3006854_9204_LINUX.zip
cd 3006854
sh rhel3_pre_install.sh

Install the CFC Infrastructure support files.

tar -xvf CFCInfra_linux.tar
cd setup
chown root libloghost.so.1
chgrp root libloghost.so.1
chmod 4755 libloghost.so.1
cp libloghost.so.1 /usr/lib

The following will test if the cluster support is working.

$ hostname
$ LD_PRELOAD=libloghost.so.1; export LD_PRELOAD
$ LHOSTNAME=clust01.mydomain.com; export LHOSTNAME
$ hostname

Do not proceed until the cluster alias is being returned by the hostname command.

Create the new groups and users on each node, making sure the IDs of the groups and users match on each node.

groupadd oinstall -g 500
groupadd dba -g 501
groupadd oper -g 502

useradd -g oinstall -G dba -s /bin/ksh -u 400 oracle
passwd oracle

Create the directories in which the Oracle software will be installed. This step will only be done on the node that is currently running the cluster service as it will be the only one that can currently see the shared disks.

mkdir -p /u01/app/oracle/product/904_infra
chown -R oracle.oinstall /u01

Login as root and issue the following command.

xhost +<machine-name>

Login as the oracle user and add the following lines at the end of the ".profile" file on each node.

# Oracle Settings
TMP=/tmp; export TMP

ORACLE_BASE=/u01/app/oracle; export ORACLE_BASE
ORACLE_HOME=$ORACLE_BASE/product/904_infra; export ORACLE_HOME
PATH=/usr/sbin:$ORACLE_HOME/bin:$PATH; export PATH
PATH=$PATH:$ORACLE_HOME/dcm/bin:$ORACLE_HOME/opmn/bin; export PATH
PATH=$PATH:$ORACLE_HOME/Apache/Apache/bin; export PATH

if [ $USER = "oracle" ]; then
  if [ $SHELL = "/bin/ksh" ]; then
    ulimit -p 16384
    ulimit -n 16384
    ulimit -u 16384 -n 16384

PS1="`hostname`> "
set -o emacs
set filec

Infrastructure Installation

Log into the oracle user on the node running the cluster service. If you are using X emulation then set the DISPLAY environmental variable.

DISPLAY=<machine-name>:0.0; export DISPLAY

Start the Oracle Universal Installer (OUI) by issuing the following command in the Disk1 directory.


During the installation select the appropriate ORACLE_HOME for infrastructure (904_infra) and select the infrastructure installation and proceed as normal.

Post Infrastructure Installation

On completion of the infrastructure installation connect to the Enterprise Manager Website (http://clust01.mydomain.com:1810) using the username "ias_admin" and the password you assigned during the installation. If EM is not available start it with the "emctl start iasconsole" command. Check that all the components are started.

The "$ORACLE_HOME/opmn/conf/opmn.xml" should have the following entries added under the <process-type> tag for all OC4J components.

  <variable id="DISPLAY" value="clust01.mydomain.com:0.0"/>
  <variable id="LD_LIBRARY_PATH" value="/u01/app/oracle/product/904_infra/lib"/>
  <variable id="LD_PRELOAD" value="libloghost.so.1"/>
  <variable id="LHOSTNAME" value="clust01.mydomain.com"/>

Starting and Stopping Services

The cluster support files contain scripts to start and stop the infrastructure. The following are copies of the files that have been modified to support AS10g.

Only the start and stop scripts should need adjustment with the relevant paths, cluster aliases and passwords. The infrastructure can then be started and stopped by running the start and stop scripts as the root user.

To switch the infrastructure between nodes do the following on the node currently running the infrastructure.

Then on the node you are switching to do the following.

For more information see:

Hope this helps. Regards Tim...

Back to the Top.