Setting up SUSE Linux Enterprise Server 11 SP2 and Red Hat Enterprise Linux 6.2
This chapter provides an overview of the steps that are to prepare a Linux guest for the installation of a Single Instance Oracle 11gR2 database. It also includes information about both distributions.
The first step is to become familiar with the following Oracle support notes (which are available at http://support.oracle.com):
•Note 1306465.1: Getting Started - 11gR2 Grid Infrastructure, ASM and DB (IBM: Linux on System z)
•Note 1290644.1: Requirements for Installing Oracle 11gR2 on SLES11 on IBM: Linux on System z (s390x) Also review note: OHASD fails to start on SuSE 11 SP2 on IBM: Linux on System z [ID 1476511.1]
•Note 1470834.1: Requirements for Installing Oracle 11gR2 on RHEL 6 on IBM: Linux on System z (s390x)
•Note 1086769.1: Ensure you have prerequisite rpms to install Oracle Database and AS10g(midtier) on IBM: Linux on System z (s390x)
There are specific sections for SUSE Linux Enterprise Server 11 and Red Hat Enterprise Linux 6 and a common section that applies to both.
This chapter includes the following topics:
4.1 Installing Oracle 11gR2 on SUSE Linux Enterprise Server guest
The minimum requirement to install Oracle 11gR2 is SUSE Linux Enterprise Server 10 SP3, or later. However, SUSE Linux Enterprise Server 11 SP2 (3.0.13-0.27) + is currently available and is preferable for 11gR2 because it incorporates support of various features of System z hardware. To verify your release, use the following command:
# cat /proc/version
Linux version 3.0.13-0.27-default (geeko@buildhost) (gcc version 4.3.4 [gcc-4_3-branch revision 152973] (SUSE Linux) ) #1 SMP Wed Feb 15 13:33:49 UTC 2012 (d73692b)
 
 
Tip: For SUSE Linux Enterprise Server 11 SP2, be sure to review the My Oracle Support note OHASD fails to start on SuSE 11 SP2 on IBM: Linux on System z [ID 1476511.1].
4.1.1 Linux required RPMs for SUSE Linux Enterprise Server 11
This section describes the following RPMs that are required to install Oracle on a SUSE Linux Enterprise Server 11 guests:
•Linux Base Installation
When you are installing on a system, select the majority of the rpms that are required at installation time to help prevent issues with rpm dependencies. For example, selecting the “C” Libraries and functions contains the majority of the “C” library rpm packages.
•RPM checker
You should use rpm checker to verify your system configuration before Oracle software is installed.
Download the appropriate RPM checker from the bottom of the My Oracle Support (MOS) Note 1306465.1. The rpm checker checks that the required rpms for Oracle Grid and database installs. This prevents problems with the installation of Oracle. You must log on to the My Oracle Support website and select the SUSE Linux Enterprise Server 11 RPM checkers S11 Grid Infrastructure/Database RPM checker 11.2.0.2 (1.38 KB), which is found at this website:
Extract the download file and then install the extracted rpm to verify your Linux rpm requirements. The rpm checker does not actually install anything. Instead, the checker uses the dependencies of rpm to check your system, as shown in the following example:
# rpm -i ora-val-rpm-S11-DB-11.2.0.2-1.s390x.rpm
****************************************************************************
* Validation complete - Your sles11 OS has required rpms for Oracle 11.2 *
****************************************************************************
 
Recommendation: Run the RPM checker command as the root user.
•Required RPMs
The following rpm packages are required for each version of Linux. The RPM release numbers can be higher than the minimum versions listed here:
Review the Note: 1383381.1 - 11.2.0.3 PREREQ CHECK WARNING FOR MISSING compat-libstdc++-33.3.2.3-47.3 ON IBM: LINUX ON SYSTEM Z ON SLES 11
 
Important: Certain packages require the 31 bit (s390) and the 64 bit version (s390x) of the rpm to be installed.
The following packages should be installed as part of a base installation:
binutils-2.20.0-0.7.9.s390x.rpm
glibc-2.11.1-0.17.4.s390x.rpm
glibc-32bit-2.11.1-0.17.4.s390x.rpm
ksh-93t-9.9.8.s390x.rpm
libaio-0.3.109-0.1.46.s390x.rpm
libaio-32bit-0.3.109-0.1.46.s390x.rpm
libstdc++33-3.3.3-11.9.s390x.rpm
libstdc++33-32bit-3.3.3-11.9.s390x.rpm
libstdc++43-4.3.4_20091019-0.7.35.s390x.rpm
libstdc++43-32bit-4.3.4_20091019-0.7.35.s390x.rpm
libgcc43-4.3.4_20091019-0.7.35.s390x.rpm
make-3.81-128.20.s390x.rpm
The remaining rpm requirements can be installed by selecting all the “C” Libraries and extensions or by manually installing each of the following rpms:
libaio-devel-0.3.109-0.1.46.s390x.rpm
libaio-devel-32bit-0.3.109-0.1.46.s390x.rpm
sysstat-8.1.5-7.9.56.s390x.rpm
glibc-devel-2.11.1-0.17.4.s390x.rpm (requires linux-kernel-headers-2.6.32-1.4.13.noarch.rpm)
gcc-4.3-62.198.s390x.rpm (requires gcc43-4.3.4_20091019-0.7.35.s390x.rpm)
glibc-devel-32bit-2.11.1-0.17.4.s390x.rpm
gcc-32bit-4.3-62.198.s390x.rpm (requires gcc43-32bit-4.3.4_20091019-0.7.35.s390x.rpm and libgomp43-32bit-4.3.4_20091019-0.7.35.s390x.rpm)
libstdc++43-devel-4.3.4_20091019-0.7.35.s390x.rpm
gcc-c++-4.3-62.198.s390x.rpm (requires gcc43-c++-4.3.4_20091019-0.7.35.s390x.rpm)
libstdc++43-devel-32bit-4.3.4_20091019-0.7.35.s390x.rpm
libstdc++-devel-4.3-62.198.s390x.rpm
libcap1-1.10-6.10.s390x.rpm
 
The following rpm command is used to verify the full extensions of the rpms. Some of the requirements need the s390 (31 bit), and some need the s390x (64 bit) version of the rpm:
# rpm -qa --queryformat="%{n}-%{v}-%{r}.%{arch}.rpm" | grep <package>
4.1.2 Network Time Protocol TIME option
If you are performing an Oracle Grid/Automated Storage Manager (ASM) installation, the grid install performs a system check to verify that the Cluster Time Synchronization Service is set to prevent the system time from being adjusted backward.
If you are installing Oracle Grid for Single Instance ASM, or Oracle RAC you should modify the NTP configuration to include the "slueing: option with the -x parameter.
Edit the /etc/sysconfig/ntp file and add the -x flag, as shown in Example 4-1.
Example 4-1 Modifying the NTP configuration
#NTPD_OPTIONS="-g -u ntp:ntp"
NTPD_OPTIONS="-x -g -u ntp:ntp"
Restart the network time protocol daemon after you complete this task by running the following command as the root user:
# /sbin/service ntp restart
Shutting down network time protocol daemon (NTPD) done
Starting network time protocol daemon (NTPD) done
# ps -ef | grep ntp | grep -v grep
ntp 56945 1 0 11:06 00:00:00 /usr/sbin/ntpd -p /var/run/ntp/ntpd.pid -x -g -u ntp:ntp -i /var/lib/ntp -c /etc/ntp.conf
Next, we must configure the system by using the following command so that the NTP daemon is started on reboot:
# chkconfig --level 35 ntp on
You might encounter the problem that is shown in Example 4-2 when Oracle runs its system pre-checks.
Example 4-2 Clock synchronization error
PRVE-0029 : Hardware clock synchronization check could not run on node xxxxx"
To resolve this problem, add the following lines to the /etc/init.d/halt.local file:
CLOCKFLAGS="$CLOCKFLAGS --systohc"
#/sbin/hwclock --systohc
4.2 Installing Oracle 11.2.0.3 on a Red Hat Enterprise Linux 6 guest
The section describes the process that is used to install Oracle 11g R2 on a Red Hat Enterprise Linux 6 (RHEL 6) guest only.
For more information about how to install Red Hat Enterprise Linux 6 for an Oracle Database, see Appendix A, “Setting up Red Hat Enterprise Linux 6.3 for Oracle” on page 311.
For Oracle Database 11gR2 (11.2.0.3), the minimum version is RHEL 6.2, kernel -2.6.32-220 or higher. This was certified in Q1 2013.
To check the version of RHEL you have installed, use the following command:
# cat /proc/version
Linux version 2.6.32-220.el6.s390x ([email protected]) (gcc version 4.4.5 20110214 (Red Hat 4.4.5-6) (GCC) ) #1 SMP Wed Nov 9 08:20:08 EST 2011.
You also should review the following notes:
•Note 1377392.1: How to Manually Configure Disk Storage devices for use with Oracle ASM 11.2 on IBM: Linux on System z Red Hat 6
•Note 1459030.1: 11.2.0.3 Grid Installer Hangs at 75% When Using DASD Softlink Device
•Note 1514012.1: runcluvfy stage -pre crsinst generates reference data is not available for verifying prerequisites for RHEL 6.
4.2.1 Verify SELinux is permissive or disabled
Oracle 11gR2 on an RHEL 6 system must have SELinux disabled or set to permissive or sqlplus does not work properly. To verify your SELinux setting, check the /etc/selinux/config file and ensure the SELinux setting is not set to Enforcing; otherwise, specify SELINUX=permissive and reboot:
# cat /etc/selinux/config
# This file controls the state of SELinux on the system.
# SELINUX= can take one of these three values:
# enforcing - SELinux security policy is enforced.
# permissive - SELinux prints warnings instead of enforcing.
# disabled - SELinux is fully disabled.
SELINUX=permissive
# SELINUXTYPE= type of policy in use. Possible values are:
# targeted - Only targeted network daemons are protected.
# strict - Full SELinux protection.
SELINUXTYPE=targeted
To change dynamically, you can run the getenforce command as root to change the SELinux security mode, as shown in the following example:
# getenforce (returns "Enforcing")
# setenforce 0
# getenforce (returns "Permissive")
 
 
Important: Disabling SELinux or setting it to “permissive” mode can have security considerations. It is possible to keep SELinux enabled and add manual exclusion rules. For more information, see Oracle note [457458.1], How to disable or set SELinux to permissive mode.
4.2.2 Linux required RPMs for Red Hat Enterprise Linux installations
For more information about how to set up a Linux guest with all the required rpms for an Oracle Database installation, see Appendix A, “Setting up Red Hat Enterprise Linux 6.3 for Oracle” on page 311.
The rpm checker for RHEL 6 is ora-val-rpm-EL6-DB-11.2.0.3-1.s390x.rpm.
4.2.3 Setting NTP TIME for Red Hat Enterprise Linux (optional only for Oracle Grid installations)
Oracle Grid/ASM performs a system check to verify that the Cluster Time Synchronization Service is set in such a way as to prevent the system time from being adjusted backward.
If you are installing Oracle Grid for Single Instance ASM or Oracle RAC, you should modify the NTP configuration to include the slewing option with the –x parameter.
To do this on RHEL, edit the /etc/sysconfig/ntpd file and add the -x flag, as shown in the following example:
# cat /etc/sysconfig/ntpd
# Drop root to id 'ntp:ntp' by default.
OPTIONS="-x -u ntp:ntp -p /var/run/ntpd.pid"
# Set to 'yes' to sync hw clock after successful ntpdate
SYNC_HWCLOCK=no
# Additional options for ntpdate
NTPDATE_OPTIONS=""
Restart the network time protocol daemon after you complete this task as the root user with the following command:
# /sbin/service ntpd restart
Next, configure the system so that the NTP daemon is started on reboot by using the following command:
# chkconfig --level 35 ntpd on
4.3 Customization that is common to SUSE Linux Enterprise Server and Red Hat Enterprise Linux
This section describes the following customization that must be done to the Linux guest that is common to SUSE Linux Enterprise Server 11 and Red Hat Enterprise Linux 6:
•Kernel parameters
•Oracle User Groups and accounts
•File Descriptor limits
•User directories
•Other RPMs
4.3.1 Required parameters for Oracle
This section describes the required parameters for Oracle.
Kernel parameters
As the root user, ensure that the required Kernel parameters are set in /etc/sysctl.conf file, as shown on Example 4-3. The recommended Kernel requirements are listed in My Oracle Support notes, such as Requirements for Installing Oracle 11gR2 on SLES 11 on IBM: Linux on System z (s390x), ID 1290644.1 and Requirements for Installing Oracle 11.2.0.3 RDBMS on RHEL 6 on IBM: Linux on System z (s390x), ID 1470834.1.
Example 4-3 Sample /etc/sysctl.conf
# Oracle Kernel Specific parameters
#
#fs.file-max = 512 x oracle processes (for example 6815744 for 13312 processes)
fs.file-max = 6815744
# fs.aio-max-nr = 3145728 (use for really large concurrent I/O databases
fs.aio-max-nr = 1048576
#kernel.shmall = set to (sum of all sga's on system) / 4096 or a default of 2097152
kernel.shmall = 2097152
#kernel.shmmax=MAX (1/2 the virtual RAM , largest SGA_MAX_SIZE/SGA_TARGET on system)
kernel.shmmax = 4218210304
kernel.shmmni = 4096
kernel.sem = 250 32000 100 128
net.ipv4.ip_local_port_range = 9000 65500
net.core.rmem_default = 262144
net.core.rmem_max = 4194304
net.core.wmem_default = 262144
net.core.wmem_max = 1048576
kernel.spin_retry = 2000
#vm.nr_hugepages = 4000 (Use for large SGA's > 10 GB)
Complete the following steps to change these values:
1. Make a copy of the /etc/sysctl.conf files if you are making any changes. Use the -p option to preserve the date, as shown in the following example:
# cp -p /etc/sysctl.conf /etc/sysctl.conf.old
2. Edit the /etc/sysctl.conf file.
3. Run the sysctl -p command for the kernel parameter changes to take effect.
Network configuration
You should comment out any IPV6 (see number 2 in Example 4-4) entries from your /etc/hosts file if you are not using IPv6 IP addresses. Also, the first line of the /etc/hosts should contain local hosts, as shown in number 1 in Example 4-4.
Example 4-4 The hosts file in the lab environment.
# cat /etc/hosts
127.0.0.1 localhost.localdomain localhost 1
# special IPv6 addresses
#localhost ipv6-localhost ipv6-loopback 2
 
9.82.34.164 ora1.wsclab.washington.ibm.com ora1
# Additional Required Only for Oracle RAC install
9.82.34.165 ora2.wsclab.washington.ibm.com ora2
10.0.0.164 ora1-priv.wsclab.washington.ibm.com ora1-priv
10.0.0.165 ora2-priv.wsclab.washington.ibm.com ora2-priv
9.82.34.167 ora1-vip.wsclab.washington.ibm.com ora1-vip
9.82.34.168 ora2-vip.wsclab.washington.ibm.com ora2-vip
#
# If Not using Oracle SCAN IP's for Oracle then setup 2 DNS entries as below
#
#9.82.34.166 ora-cluster.wsclab.washington.ibm.com ora-cluster crs
#9.82.34.169 ora-cluster-scan.wsclab.washington.ibm.com ora-cluster-scan
9.82.34.167 ora1-vip.wsclab.washington.ibm.com ora1-vip
9.82.34.168 ora2-vip.wsclab.washington.ibm.com ora2-vip
Oracle also requires that the host name be the fully qualified domain name, with a corresponding entry in the /etc/hosts file, as shown in the following example:
# hostname
ora1.wsclab.washington.ibm.com
4.3.2 Oracle RAC installations only
For Oracle RAC installations, more IP addresses are required (see Example 4-5) for the other nodes in the RAC cluster (ora2), the Private Interconnect (-priv), and the variable IPs (-vip), which are created when the Oracle Grid starts.
Example 4-5 Other IP addresses identified in the lab environment
10.0.0.164 ora1-priv.wsclab.washington.ibm.com ora1-priv
10.0.0.165 ora2-priv.wsclab.washington.ibm.com ora2-priv
 
Tip: For other steps and requirements, see Chapter 3, “Network connectivity options for Oracle on Linux on IBM System z” on page 29 and Appendix B in the Installing Oracle 11gR2 RAC on Linux on System z, REDP4788.
One other network interface on each server must be created, such as hsi0 - virtual HiperSocket. This network interface is between the Linux Guests for Oracle’s Interconnect and should be on a private non-routable interface (192.x.x.x or 10.x.x.x). Only the nodes in the RAC cluster should contact the private interface.
You also require two other IP addresses for the Oracle Virtual IPs (VIPs) that must be on the same subnet as the public eth0 interface, as shown in Example 4-6.
Example 4-6 Oracle VIPs
9.82.34.167 ora1-vip.wsclab.washington.ibm.com ora1-vip
9.82.34.168 ora2-vip.wsclab.washington.ibm.com ora2-vip
Finally, you need three SCAN IP addresses to be defined as Class A DNS entries, as shown in Example 4-7 (there are three IP addresses for each DNS entry) for the new 11gR2 Oracle RAC systems. These should also be on the same subnet as the public interface.
Example 4-7 DNS SCAN entries
rac-scan IN A 9.82.34.166
rac-scan IN A 9.82.34.167
rac-scan IN A 9.82.34.168
 
# Note 3 IPs to one DNS (host file entry, but we require DNS entries for this to work)
#
9.82.34.166rac-scan.<domain name>rac-scan
9.82.34.167rac-scan.<domain name>rac-scan
9.82.34.168rac-scan.<domain name>rac-scan
If you cannot set up DNS SCAN entries at this time, you can define two /etc/host entries on each of the nodes, but you receive a warning that can be ignored during the installation, as shown in Example 4-8.
Example 4-8 Non-SCAN Oracle RAC configuration
9.82.34.166 ora-cluster.wsclab.washington.ibm.com ora-cluster crs
9.82.34.169 ora-cluster-scan.wsclab.washington.ibm.com ora-cluster-scan
 
Important: When the two network interfaces for Oracle RAC are configured (public and private interfaces), you must have ARP enabled (that is, NOARP must not be configured). The root.sh script fails on the first node if NOARP is configured.
Example 4-9 shows the command ifconfig -a run as though root user.
Example 4-9 ifconfig output
eth0 Link encap:Ethernet HWaddr 02:00:00:00:00:05
inet addr: 9.82.34.164 Bcast: 9.82.63.255 Mask:255.255.255.224
inet6 addr: fe80::200:0:100:5/64 Scope:Link
UP BROADCAST RUNNING MULTICAST MTU:1492 Metric:1
RX packets:6503 errors:0 dropped:0 overruns:0 frame:0
TX packets:199 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:697138 (680.7 Kb) TX bytes:24804 (24.2 Kb)
 
hsi0 Link encap:Ethernet HWaddr 02:00:00:00:00:06
inet addr:10.0.0.164 Bcast:10.0.0.255 Mask:255.255.255.0
inet6 addr: fe80::ff:fe00:6/64 Scope:Link
UP BROADCAST RUNNING NOARP MULTICAST MTU:8192 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:5 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 b) TX bytes:390 (390.0 b)
4.3.3 Create and verify required UNIX groups and Oracle user accounts
When Oracle 11gR2 is installed, Oracle recommends that two groups be created: one for the group named dba, and another group called oinstall. (It is possible to install with one group; for example, dba.)
If only database executable files are installed, often one user account called oracle is created.
If Oracle Grid for Oracle ASM or a Real Application Cluster (RAC) system is installed, another user account called grid should be created to manage the grid infrastructure components.
As part of a grid infrastructure installation, Oracle changes certain directories and files to have root access privileges. Having separate user IDs (one for grid, and one for oracle) makes it easier to configure the environment variables that are required to maintain each environment.
To verify that the Linux groups and users were created, you can view the group and password files by using the following commands:
# cat /etc/passwd | grep oracle
# cat /etc/group
If your users and groups were not created, run the commands that are shown in Example 4-10 to create users. Having consistent group IDs (that is, 501) and user IDs (for example, 502) across nodes is required, particularly if you share storage or files between systems.
Example 4-10 Commands to create users
/usr/sbin/groupadd -g 501 oinstall
/usr/sbin/groupadd -g 502 dba
/usr/sbin/useradd -u 501 -g oinstall -G dba -s /bin/ksh -m grid
/usr/sbin/useradd -u 502 -g oinstall -G dba -s /bin/ksh -m oracle
Use the following commands to set the passwords for the grid and oracle users:
# passwd grid
# passwd oracle
 
Tip: If you encounter INS-06101 errors, verify that the host name is correct in the /etc/hosts file and host name is fully qualified.
4.3.4 Setting file descriptors limits for the oracle and grid users
As the root user, edit or verify the /etc/security/limits.conf file. If you created a separate user for the Oracle Grid user, the file descriptor limit or ulimit entries for the grid user should be created, as shown in Example 4-11.
Example 4-11 Unlimit entries
#vi /etc/security/limits.conf
grid soft nofile 1024
grid hard nofile 65536
grid soft nproc 2047
grid hard nproc 16384
#
oracle soft nofile 1024
oracle hard nofile 65536
oracle soft nproc 2047
oracle hard nproc 16384
#
# Use memlock for Huge Pages support (commented out)
#* soft memlock 3145728
#* hard memlock 3145728
Ensure that the /etc/pam.d/login file has an entry for pam_limits.so. Also, you should make a backup if changes are made to /etc/pam.d/login and to test any changes with a superuser or login before logging off, as a typographical error can make future logins problematic, as shown in Example 4-12.
Example 4-12 Making a backup and verification
# cp /etc/pam.d/login /etc/pam.d/login.old
# cat /etc/pam.d/login
#%PAM-1.0
auth required pam_nologin.so
…
session optional pam_mail.so standard
session required /lib/security/pam_limits.so
session required pam_limits.so
session optional pam_mail.so standard
To increase the limits at oracle logon, as the Oracle User, verify the oracle user's profile (for example, /home/oracle/.profile for ksh users) and ensure the following lines were added:
#vi .profile
ulimit -n 65536
ulimit -u 16384
Another method is to modify the main system profile by adding the following lines to the file called /etc/profile, as shown on Example 4-13. Change this if the oracle user is using a separate user shell program, such as csh or bash.
Example 4-13 Modifying the main system profile
If [ $USER = "oracle" ]; then
ulimit -u 16384
ulimit -n 65536
fi
# Optional Grid User
If [ $USER = "grid" ]; then
ulimit -u 16384
ulimit -n 65536
fi
You should then log in as the oracle user and run the command ulimit -a to verify everything is configured correctly, as shown in Example 4-14.
Example 4-14 Verifying the configuration
# su - oracle
$ ulimit -a
address space limit (kbytes) (-M) unlimited
core file size (blocks) (-c) 0
cpu time (seconds) (-t) unlimited
data size (kbytes) (-d) unlimited
file size (blocks) (-f) unlimited
locks (-L) unlimited
locked address space (kbytes) (-l) unlimited
nice (-e) 0
nofile (-n) 65536
nproc (-u) 16384
pipe buffer size (bytes) (-p) 4096
resident set size (kbytes) (-m) unlimited
rtprio (-r) 0
socket buffer size (bytes) (-b) 4096
stack size (kbytes) (-s) 10240
threads (-T) not supported
process size (kbytes) (-v) unlimited
4.3.5 Pre-create user directories for product installs
If you are installing Oracle Grid, the GRID_BASE directory must be different from the grid’s product directory (where the executable files are installed). In Example 4-15, /oracle is the logical volume for installing the Oracle executable files.
Example 4-15 Creating directories for Oracle Grid
# Make Directories for Oracle Grid
mkdir -p /u01/grid/base
mkdir -p /u01/grid/11.2
chown -R grid:oinstall /u01/grid
chmod -R 775 /u01/grid
# Make Directories for Oracle User (database)
mkdir -p /u01/oracle/11.2
chown -R oracle:oinstall /u01/oracle
chmod -R 775 /u01/oracle
Do not set any of the standard environment variables. If standard environment variables are already set, unset the standard ORACLE_HOME and LD_LIBRARY_PATH with a comment, as shown in lines 1 and 2 in Example 4-16 and Example 4-17.
Example 4-16 Example of /home/grid/.profile
$ cat .profile
export ORACLE_BASE=/u01/grid/base
export GRID_BASE=/u01/grid
#export ORACLE_HOME=$GRID_BASE/11.2
#
# comment out the following lines for use later, do not have set for runInstaller
#
#export PATH=$ORACLE_HOME/bin:$ORACLE_HOME/OPatch:$PATH:. 1
#export LD_LIBRARY_PATH=$ORACLE_HOME/lib:$LD_LIBRARY_PATH 2
umask 022
#defaults for shell startup for ulimits of oracle user
ulimit -u 16384
ulimit -n 65536
Example 4-17 Example of /home/oracle/.profile
$cat .profile
 
export ORACLE_BASE=/u01/oracle
#export ORACLE_HOME=$ORACLE_BASE/11.2
#
# comment out the following lines for use later, do not have set for runInstaller
#
#export PATH=$ORACLE_HOME/bin:$ORACLE_HOME/OPatch:$PATH:. 1
#export LD_LIBRARY_PATH=$ORACLE_HOME/lib:$LD_LIBRARY_PATH 2
umask 022
#defaults for shell startup for ulimits of oracle user
ulimit -u 16384
ulimit -n 65536
 
4.3.6 Other rpm for grid installs for SUSE Linux Enterprise Server and Red Hat Enterprise Linux
If you are performing an Oracle RAC install, you must install the cvudisk-1.0.9-1 rpm package from the Oracle 11gR2 distribution media, as shown in Figure 4-1.
Figure 4-1 Installing other RPMs
You can create a fix up script or install the RPM from the software distribution on each of the nodes in the RAC cluster.
To run the fixup scripts, complete the following steps:
1. Open a terminal window.
2. Log in as the root user.
3. Run the scripts.
4. Return to the window that is shown in Figure 4-2 and click OK.
Figure 4-2 Running fixup scripts
Example 4-18 shows running the fixup script.
Example 4-18 Running the fixup script
root@lnx007 CVU_11.2.0.2.0_grid]# ./runfixup.sh
/usr/bin/id
Response file being used is :./fixup.response
Enable file being used is :./fixup.enable
Log file location: ./orarun.log
Installing Package /tmp/CVU_11.2.0.2.0_grid//cvuqdisk-1.0.9-1.rpm
Preparing... ########################################### [100%]
1:cvuqdisk ########################################### [100%]
The RPM can also be found in the <Grid CD>/Disk1/rpm/cvuqdisk-1.0.9-1.rpm directory and installed in advance.
 
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset