Drop Down MenusCSS Drop Down MenuPure CSS Dropdown Menu

Simple Oracle 11g Rac setup

In this article I will show you how to perform a basic installation of Oracle RAC 11R2 on Linux 6 using Virtual Box. The cluster we will build will consist of two nodes using ASM as shared storage.

Here are the details of each virtual node:
Node1Node2
Name: oralab1 
Hostname: oralab1.localdomain 
Ram: 3Gb 
Nics: Eth0: 192.168.1.151 #Public IP 
Eth1: 192.168.2.152 #Private IP 
OS: Oracle Linux 6.3 x86_64 euk 
Disks: 1 x 50GB root FS
Name: oralab2 
Hostname:oralab2.localdomain 
Ram: 3Gb 
Nics: Eth0: 192.168.1.156 #Public IP 
Eth1: 192.168.2.157 #Private IP 
OS: Oracle Linux 6.3 x86_64 euk 
Disks: 1 x 50 GB root FS
Shared resources:
  • Shared Disks: 5 disks of 3 GB
  • Scan IPs: 192.168.1.160, 192.168.1.161, 192.168.1.162
  • Scan Name: oralab-cluster
Important notes:
  • The following installation only serves as a demonstration.
  • The host machine must have at least 8GB of RAM and at least 150GB of disk space for root disks and ASM shared disks.
  • For the installation of this version it is necessary that the 3 addresses of the SCAN are resolved by the oralab-cluster name. For this it is necessary to add them to the DNS alias in round robin mode. The DNS configuration will not be in this tutorial.
  • They must synchronize the virtual machines with an NTP server. Oracle recommends that the cluster nodes be synchronized via NTP with a special configuration.
  • The software used throughout the installation will be the latest versions of the same.
  • We will use the oracle-rdbms-server-11gr2-preinstall package for the basic installation settings.
The installation will be carried out in the following order:
  1. Download the necessary software.
  2. Creation of the first virtual machine (Node 1).
  3. Installing the operating system.
  4. Oracle Clusterware Preinstallation Tasks.
  5. Creation of the second virtual machine (Node 2)
  6. Installing Oracle Clusterware.
  7. Installing the Database Software.
  8. Creation of the RAC database.
Let's start !!
PART I - Download the necessary software.
Download the installation media of each of the required software.
Note: For this tutorial I will be using versions 11.2.0.3 of Oracle clusterware and database, this tutorial is similar for version 11.2.0.1 (2).
PART II - Creation of the first virtual machine.
Once the required average is downloaded, we proceed to create our first virtual machine, which will be our node1 as follows.
In the Virtual Box Manager select new virtual machine to start the creation.
Select the amount of memory.
Create the disk for our VM. You can create it by following the screens.
Once our first VM is created, we proceed to edit the properties and add a second network interface in bridged mode.
Note: Interface: Adapter 1 must also be in briged mode.
PART III - Operating System Installation
We started our virtual machine with the average of Oracle Linux and started the installation.
Once I loaded anaconda installer, we clicked Next to start our installation.
We select the language of the system and the keyboard.
We select our type of storage
Click on delete the entire date.
We put our hostname.
We select our TimeZone.
We create the root password
We select the Replace existing Linux system option and continue
In this screen it gives us to choose the type of installation of wanted according to the role of server. I particularly like to select the components I want, so we will select Curstomize now at the bottom of the screen.
In the next screen we select the packages of desire, for this installation we will select the following ones:
Server / system administration tools / oracle-RDBMS-server-11gR2-preinstall 
server / system administration tools / oracleasm-support 
Desktops / desktop 
Desktop / Desktop DEBUGIN and performace tools 
Desktops / Desktop platform 
Desktops / General Purpose Desktop 
Desktops / Grafical Administration tools 
Desktops / X window system 
Application / Internet Browser
Clicking Next will start the installation.
When finished we click on Reboot to complete.
Once the server is started, we follow the installation as shown in the following screens.
In the next screen do not create user, those that we will need we will be creating them later.
On the next screen, enter the address of your NTP server.
Completed all the above will have the login of our server. You must login with the root account.
PART IV - Oracle Clusterware Preinstallation Tasks
We start the session with the root user, all the following steps are done with this user.
It is important to highlight the following points when configuring the interfaces for the cluster:
  • The eth0 public interface must be on a different subnet than the eth1 private interface.
  • the public and private interfaces must be the same on both nodes.
  • The SCAN addresses must be resolved by all cluster nodes via DNS by the oralab-cluster name.
  • The SCAN addresses must be on the same subnet as the published interface.
With the previous points in clear, proceed to configure each interface with the following values
Interface: eth0 
IP: 192.168.1.151 
Netmask: 255.255.255.0 
Gateway: 192.168.1.1
Interface: eth1 
IP: 192.168.2.152 
Netmask: 255.255.255.0 
Gateway: 0.0.0.0
Restart network services
service network reload
For the following steps I will be using the Putty (just for convenience).
We disable Linux Security by editing the / etc / sysconfig / selinux file and add disable as a sample.
vi / etc / sysconfig / selinux
We will create the necessary users and groups.
groupadd asadmin 
groupadd asmoper 
groupadd asmdba 
useradd -g oinstall -G asmadmin, asmdba, dba, asmoper grid 
usermod -g oinstall -G asmdba, dba oracle 
passwd grid 
passwd oracle
We create the necessary directories and assign the permissions
mkdir -p / u01 / app / grid 
mkdir -p /u01/app/11.2.3/grid 
mkdir -p /u01/app/oracle/product/11.2.3/db1 
chown -R grid: oinstall / u01 / app 
chown -R oracle: oinstall / u01 / app / oracle 
chmod 775 / u01 / app
We edit the profiles of the users oracle and grid with the following values.
#VARIABLES ORACLE USER
export ORACLE_SID = orcl1 
export ORACLE_BASE = / u01 / app / oracle 
export ORACLE_HOME = $ ORACLE_BASE / product / 11.2.3 / db1 
export ORACLEUNQ_NAME = orcl1 
export ORACLE_HOSTNAME = oralab1.localdomain 
export PATH = $ PATH: $ ORACLE_HOME / bin
#VARIABLES GRID USER 
export ORACLE_SID = + ASM1 
export ORACLE_BASE = / u01 / app / grid 
export ORACLE_HOME = / u01 / app / 11.2.3 / grid 
export ORACLEUNQ_NAME = + ASM1 
export ORACLE_HOSTNAME = oralab1.localdomain 
export PATH = $ PATH: $ ORACLE_HOME bin
We edit the host file with the following values.
vi / etc / hosts
#ip PUBLICAS 
192.168.1.151 oralab1.localdomain oralab1 # eth0 
192.168.1.156 oralab2.localdomain oralab2 
#ip DEPRIVED 
192.168.2.152 oralab1-priv.localdomain oralab1-priv # eth1 
192.168.2.157 oralab2-priv.localdomain oralab2-priv 
#ip VIRTUALES 
192.168 .1.153 oralab1-vip.localdomain oralab1-vip 
192.168.1.158 oralab2-vip.localdomain oralab2-vip
We edit the / etc / sysconfig / ntpd file and add the x flag
vi / etc / sysconfig / ntpd
We restart our ntp
Edit user limits add the following values
grid soft nofile 1024 
grid hard nofile 65536 
grid soft nproc 2047 
grid hard nproc 16384 
grid soft stack 10240 
grid hard stack 32768
vi / etc / security / limits.conf
We turn off our virtual machine and proceed to create the disks that will be used by ASM. In total we will create 5 discs of about 3GB in size.
You can create them by running the following statements on the host machine.
VBoxManage createhd -filename oralab11.vdi -size 3072 -format VDI -variant Fixed 
VBoxManage createhd -filename oralab12.vdi -size 3072 -format VDI -variant Fixed 
VBoxManage createhd -filename oralab13.vdi -size 3072 -format VDI -variant Fixed 
VBoxManage createhd -filename oralab14.vdi -size 3072 -format VDI -variant Fixed 
VBoxManage createhd -filename oralab15.vdi -size 3072 -format VDI -variant Fixed
VBoxManage storageattach oralab1 -storagectl "SATA Controller" -port 1 -device 0 -type hdd -medium oralab11.vdi -mtype shareable 
VBoxManage storageattach oralab1 -storagectl "SATA Controller" -port 2 -device 0 -type hdd -medium oralab12.vdi - mtype shareable 
VBoxManage storageattach oralab1 -storagectl "SATA Controller" -port 3 -device 0 -type hdd -medium oralab13.vdi -mtype shareable 
VBoxManage storageattach oralab1 -storagectl "SATA Controller" -port 4 -device 0 -type hdd -medium oralab14. vdi -mtype shareable 
VBoxManage storageattach oralab1 -storagectl "SATA Controller" -port 5 -device 0 -type hdd -medium oralab15.vdi -mtype shareable
VBoxManage modifyhd oralab11.vdi -type shareable 
VBoxManage modifyhd oralab12.vdi -type shareable 
VBoxManage modifyhd oralab13.vdi -type shareable 
VBoxManage modifyhd oralab14.vdi -type shareable 
VBoxManage modifyhd oralab15.vdi -type shareable
Once created and attached to the VM will look like this
We start our VM and partition each of the disks. For this we execute the following.
fdisk / dev / sdb 
options n, p, 1,1, w
fdisk / dev / sdc 
options n, p, 1,1, w
fdisk / dev / sdd 
options n, p, 1,1, w
fdisk / dev / sde 
options n, p, 1,1, w
fdisk / dev / sdf 
options n, p, 1,1, w
More or less it would look like this
After partitioning the disks we execute a ls / dev / sd * to verify the correct creation of the partitions
It's time to configure asmlib and create ASM disks. Run the following
/etc/init.d/oracleasm configure
And respond according to the following screen
Create the disks with the following commands
oracleasm createdisk disk1 / dev / sdb1 
oracleasm createdisk disk2 / dev / sdc1 
oracleasm createdisk disk3 / dev / sdd1 
oracleasm createdisk disk4 / dev / sde1 
oracleasm createdisk disk5 / dev / sdf1
Verify creation
oracleasm listdisks
So far we are ready with the pre-installation of clusterware on node1. What follows is to create our second virtual machine which will be our second node.
PART V - Creation of the second virtual machine (Node 2)
To simplify the creation of our second node, we will proceed to clone the main disk of the oralab1 virtual machine and then create a new virtual machine with that disk.
To create the disk clone run the following statement on the host machine.
VBoxManage clonehd "oralab1.vdi" "oralab2.vdi"
Once the disk is cloned, we create the new virtual machine using it.
Once created, we add our second network interface.
Now we execute the following statements to cross out the shared disks.
VBoxManage storageattach oralab2 -storagectl "SATA Controller" -port 1 -device 0 -type hdd -medium oralab11.vdi -mtype shareable 
VBoxManage storageattach oralab2 -storagectl "SATA Controller" -port 2 -device 0 -type hdd -medium oralab12.vdi - mtype shareable 
VBoxManage storageattach oralab2 -storagectl "SATA Controller" -port 3 -device 0 -type hdd -medium oralab13.vdi -mtype shareable 
VBoxManage storageattach oralab2 -storagectl "SATA Controller" -port 4 -device 0 -type hdd -medium oralab14. vdi -mtype shareable 
VBoxManage storageattach oralab2 -storagectl "SATA Controller" -port 5 -device 0 -type hdd -medium oralab15.vdi -mtype shareable
They should look like this:
We start our second virtual machine and change our network interfaces as follows
Restart network services
We edit the user profiles as we did before taking into account the new values
We then verify that the disks are present on this node by executing the following command.
oracleasm listdisks
This should be seen
Turn on node1 and do the following tests
From node1
ping oralab2 
ping oralab2-vip 
ping oralab2-priv
From node2
ping oralab1 
ping oralab1-vip 
ping oralab1-priv
If all the pings respond we will be ready to proceed with the installation of the grid infrastructure.
PART VI - Installing Oracle Clusterware
When you decompress the clusterware mean, a folder called grid will be created. There we can start the installation by calling the runInstaller script.
Before starting the installation of Oracle Clusteware it is important to verify that all the prerequisites are fulfilled. With cluster verification utility we can verify not only the entire installation, but also individual components of the cluster.
To verify all the components of the cluster before starting the installation we only execute the following command from the root of the installation.
./ runcluvfy.sh stage -pre crsinst -n oralab1, oralab2
Note: There are other options that we can pass to cluvfy, which I just passed is enough.
Start the installer with the statement:
./runInstaller
Wait for it to load and continue
Select the type of installation for cluster
Select the type of installation typical
In the next screen we will have to edit certain things. The first thing is to add the nodes that will contain the cluster.
Click on add and enter the public and virtual addresses (VIP) of node 2
Once the secondary node is added, it is time to establish the trust relationships between both nodes.
Clicking on SSH Connectivity
Once we complete all the above we continue and enter the information of our storage.
Select ASM and enter the password that the ASM will have
In the next screen we will have to select the disks that will form the disk group that will house our Voting disk and OCR. We will create a diskgroup called DG_DATA with external redundancy.
You can create it as shown in the following screen
Note: If the disks do not appear initially, change the path of discovery path to / dev / oracleasm / disks
In the next screen we will see the route of the inventory
The next screen is run again cluvfy to identify possible faults
If everything is correct in the configuration you will see the summary of the installation.
When you click click install
Once certain tasks are completed, the installer will ask us to run some scripts on all the nodes that make up the cluster.
It would look something like this
It is important to preserve the order of execution. First run script 1 on node2 (oralab1) and then on node2 (oralab2), when you complete execute script 2 on node1 (oralab1) and then on node2. At the end of the execution of each of the nodes are seen screens as well
Node1
Node2
When you finish it looks something like this
We already have our cluster ready.
Check the status of all components running
crsctl status resource -t
As we see all the components are running, I mean ASM instances, listener, scan, etc.
PART VII - Installing the Database Software.
We will perform a software-only installation to then create our RAC database.
We will start unpacking the installation media
Once the files are unzipped we proceed to start the installer
Follow the installation as shown on the screens
When you are completing the installation you will be prompted to execute a script on each node. Run them in indicated order
PART VIII - Creation of the RAC database.
With the software installed, we proceed to create our rac database.
Let's call dbca
Select the type of database.
Select admin-managed and select the nodes of the cluster.
Select the password for the accounts.
For storage select the previously created diskgroup.
Setting the ASM password
As this installation is demonstration we will not activate archiving nor the FRA.
Review status and configuration
srvctl status database -d orcl 
srvctl config database -d orcl
Now with the grid user.
I hope you serve them.
Related article:

Simple Oracle 10g Rac setup

Comments

Popular posts from this blog

Oracle DBMS SCHEDULER Examples

How to find the server is whether standby (slave) or primary(master) in Postgresql replication ?

7 Steps to configure BDR replication in postgresql

How to Get Table Size, Database Size, Indexes Size, schema Size, Tablespace Size, column Size in PostgreSQL Database

How to Enable/Disable autovacuum on PostgreSQL