Book Oracle RAC

Build Oracle 10g RAC Database On Solaris / Linux With VMWare
All rights reserved. No part of this publication may be reproduced, stored in retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of the author.
About the Book

Build Oracle 10g RAC Database on Solaris / Linux with VMWare covers all the important aspects of Implementing/Managing Oracle 10g RAC from installation to day-to-day maintenance. The book would be a very good starting point for someone who wants to know about Oracle RAC.
To my parents and my beloved country
Table of Contents
1. 2.
Introduction Implement Oracle RAC on Solaris 7 13
Task 1: Prepare System for Oracle RAC Section: A Create first Virtual Machine 13 Section: B Install Solaris 10 OS on first Virtual Machine 14 Section: C Create Shared Storage and Configure 15 Section: D Create and Configure the Second Virtual Machine 16 Section: E Add Ether Net Adapter for Private Network in Both Machine 16 Section: F Prepare Disk for OCR, Voting and ASM Storage 17 Task 2: Install Oracle Clusterware Section: A Check All Pre-request for Clusterware installation 20 Section: B Create Oracle User and Group Section: C Configure SSH Section: D Specify default gateway
22 22 23
Section: E Install Oracle Clusterware 24 Task 3: Install Oracle 10gR12 Software/binary Task 4: Configure Oracle Listener 24 Task 5: Create and Configure ASM Instance and DISK Groups By DBCA By MANUAL Tack 6: Create Database By DBCA By MANUAL Task 7: Setup Test Transport Failover 3. Implement 10gR2 Oracle RAC on Linux Section: A Hardware Requirements and Overview 58 Section: B Configure the First Virtual Machine Section: C Install and Configure Enterprise Linux on the First Virtual Machine 63 Section: D Create and Configure the Second Virtual Machine Section: E Configure Oracle Automatic Storage Management (ASM) 81 Section: F Configure Oracle Cluster File System (OCFS2) 82 Section: G Install Oracle Clusterware Section: H Install Oracle database binary 86 Section: I Configure Oracle Listener 86 Section: J Confihure ASM 86 Section: K Create Database 86 Section: L Setup Test Transport Failover 53 36 40 42 50 53 28
59 77
85
4. Convert 10gR2 Stand-alone database to Oracle RAC on Solaris

Method: Manually
87
Method: rconfig 95 Method: DBCA 98
5. Managing OCR and Voting Disk

100 6. Administering Cluster Ready Services (CRS) 106 7. Administrating Services 109 8. Managing UNDO, Temporary and redo logs in RAC Environment 112 9. De-Installing Oracle Real Application Clusters Software and Database 115 10. De-Installation RAC Component after Fail Installation 116 11. Adding a Node To 10gR2 RAC cluster 117 12. Removing a 10gR2 RAC cluster Node 120 13. RAC Load Balancing 122 14. RAC Failover Case Study 125 15. Oracle RAC Log Directory 127
Introduction
What is RAC? RAC stand for Real Application cluster. RAC provides cluster solution that ensures high availability of instance and load balancing. What is Cluster? A cluster is a set of 2 or more machine (nodes) that share resource to perform the same task. What is RAC Database? A RAC database is 2 or more instance running on a set of clusters nodes with all instances accessing a shared set of database files. Different between RAC and Non RAC database RAC stands for Real Application Clusters. It allows multiple nodes in a clustered system to mount and open a single database that resides on shared disk storage. Should a single system fail (node), the database service will still be available on the remaining nodes. A non-RAC database is only available on a single system. If that system fails, the database service will be down (single point of failure). Why use RAC? We can achieve benefit in the following ways:
High availability - If some nodes fail, the remainder of the nodes will still be available for processing requests. Speedup (increased transaction response time) - RAC normally adds some overhead. Scale-up (increased transaction volume) - RAC can be used to provide increased application scalability
What are Clusterware processes? When clusterware is started, the following process will be running: crsd Cluster Resource Services Daemon
cssd Cluster Synchronization Services Daemon evmd Event Manager Daemon Major component in 10G RAC (Hardware Level and Software Level) Here we discuss about major component in 10g RAC. We can categories 10g RAC component between Hardware and Software level. In Hardware level, three major components are Shared Disk Storage, Private network and Public Network. In RAC architecture, we need Shared Disk Storage because the database file, redo logs and control files (also OCR and VOTING Disk) must be accessible by each node. For Shared Disk configuration, we can use SCSI (just a Bunch of Disk), Storage Area network (SAN) and Network Attached Storage (NAS) Each node must be connected to all other node via private high-speed network in RAC architecture. Each node must be assign public IP address. We must assign a virtual IP address (VIP) for maintain high availability. At the time of node failure, the failed node's IP address can be reassigned to a surviving node to allow applications to continue accessing the database through the same IP address. In Software Level, also three major components are OS, Clusterware Software and Storage Mechanism we would like to use for the database (OCFS/ASM/RAW) At the time of RAC implementation, be ensure that we have install all necessary software package, setting kernel parameter, configure the Network, configure the disk device and create directory structure and also ensure Oracle clusterware have install on each cluster node. We can use following file storage mechanism for the oracle clusterware component and database componentt: For Clusteware Component: OCFS (Release 1 or 2) Raw devices Third party cluster filesystem such as GPFS or Veritas For RAC database storage: OCFS (Release 1 or 2) ASM
Raw devices Third party cluster filesystem such as GPFS or Veritas Hardware Requirements for Oracle Real Application Clusters Each node in a cluster requires the: External shared disks (All nodes connected with External Shared Storage) for storing OCR, Voting Disk file and Database files. One Network Ethernet Card for private connection. We assign IP address for each node to serve as the private Interconnect. Important: Private interconnect must be separate from the public network Private interconnect must be accessible on the same network interface on each node Must have a unique address on each node One Network Ethernet Card for public connection. We assign IP address for each node, to be used as the Virtual IP address for client Connections and for connection fail over
Storage Option for Oracle Real Application Clusters
We can use a file system (NFS) or raw device (partition) for Oracle Clusterware File. We can use ASM or raw device for database or recovery file storage but can not use raw devices for recovery. NOTE: Oracle recommends that you choose Automatic Storage Management (ASM) as the storage option for database and recovery files. You cannot use ASM to store Oracle Clusterware files, because these files must be accessible before any ASM instance starts.
If you intend to use ASM with RAC, and you are configuring a new ASM instance, then your system must meet the following conditions: All nodes on the cluster have the release 2 (10.2) version of Oracle Clusterware installed. Any existing ASM instance on any node in the cluster is shut down. If you do not have a storage option that provides external file redundancy, then you must configure at least three voting disk areas to provide voting disk redundancy.
Memory structure and background process in Oracle RAC Oracle RAC is composed of two or more database instances. They are composed of Memory structures and background processes same as the single instance database. Oracle RAC instances use two processes GES (Global Enqueue Service), GCS (Global Cache Service) that enable cache fusion. Oracle RAC instances are composed of following background processes: An Oracle RAC database has the same processes and memory structures as a single-instance Oracle database as well as additional process and memory structures that are specific to Oracle RAC. The list of Oracle RAC processes are: 1)LMSGlobal Cache Service Process 2)LMDGlobal Enqueue Service Daemon 3)LMONGlobal Enqueue Service Monitor 4)LCK0Instance Enqueue Process LMS: - Is called Global Cache Service process. This process manages the data sharing and data exchange in RAC Environment. Each cache blocks record in Global Cache Directory. LMON: - is the Lock monitor Process and is responsible for managing the Global Enqueue services. It maintains consistency of GCS memory in case of process death. LMD; - is the daemon process that manages enqueue manager service request for GCS. LCK0:- Is managing instance resource request and cross-instance call operation for shared resources.
10
DIAG: - is a lightweight daemon process for all the diagnostic need of an instance in an RAC environment. How Oracle RAC does works? Each instance will have its own set of buffer but will be able to request and receive data blocks currently held in another instance cache. In single instance environment, the buffer cache access by only one set of process. In RAC, the buffer cache of one node may contain data that is requested by another node. The management of data sharing and exchange in this environment is done by Global cache Services (GCS) When a block is transferred out of a local cache to another cache, the GRD is updated. Global Resource Directory (GRS) is the internal repository that record and store the current status of the data blocks. The GRD is managed by Global Cache Service (GCS) and Global Enqueue Services (GES). Following information is available in GRD. Data block identifiers Location of most current version Modes of data block (Null/Share/Exclusive) When one instance departs the cluster, the GRD portion of that instance needs to be redistributing to the surviving node. When a new instance enters the cluster, the GRD portions of the existing instance must be redistributed to create the GRD portion of the new instance.
What is Cache Coherency? Is an important part of RAC. Cache coherency is the technique of keeping multiple copies of a buffer consistent between different oracle instances of different nodes. Global Cache Management ensure that access to master copy of a data blocks in one buffer cache is coordinated with the copy of the block in another buffer cache. This ensure the most recent copy of a block in a buffer cache contains all changes that are mad to that block by any instance in the system regardless of wheather those changes have been commited on the transaction lavel.
11
Oracle 10g RAC Architecture (Graphical View)
Implement 10gR2 Oracle RAC on Solaris 10 with VMware

12
Important Tips Overview of the RAC database environment, which will be setup in bellow steps Host Nam e RAC1 RAC2 Databa se Name PROD PROD Instanc e Name PROD1 PROD2 Database File Storage ASM ASM OCR & CRS Voting Disk RAW RAW
Overview of Network Setup Details Node: RAC1 Node: RAC2 Network Adaptor pcn0 form public (IP: Network Adaptor pcn0 form public (ip: 192.168.0.111) 192.168.0.222) Network Adaptor pcn1 form public (IP: Network Adaptor pcn1 form public (IP: 192.168.0.11) 192.168.0.22) IP for VIP: 192.168.0.50 IP for VIP:192.168.0.60 Details c0d1s0 c0d1s1 c0d1s3 c0d1s4 c0d1s6 Overview of Shared Storage partition Size Uses 500 MB OCR 500 MB Voting Disk 3 GB ASM DISK 1 3 GB ASM DISK2 2 GB ASM DISK3 Task 1 (Prepare System for Oracle RAC) Section: A (Create first Virtual Machine) 1. Create the windows folders to house the first virtual machines and the shared storage. F:\>mkdir RAC1 F:\>mkdir SHARED-STORAGE 2. 3. 4. 5. 6. Open VMWARE Console and Click on New Virtual Machine and click next. Select Virtual Machine Configurations: Custom Select Gust Operation System: Sun Solaris and version solaris10 Type virtual Machine Name (RAC1) and Location (F:/RAC1). Set Access Right (Select Default Value)
13
7. Select Startup/Shutdown option (Select Default value) 8. Process Configuration (Select number of process: One) 9. Set memory for the virtual Machine (700MB) 10. Select Network Connection. Total 4 options (Select Host Only Network) User Bridge connection Use network address translation (NAT) Use host only network (Select Host Only Network) Do not use a network 11. Select I/O Adepter types (LSI Logic) 12. Select a Disk (create a new virtual disk) 13. Select a Disk type (Select: IDE) 14. Specify Disk Capacities. (8 GB) and Click on Finish Button. Now your virtual machine is ready to install Solaris 10 OS Section: B (Install Solaris 10 OS on first Virtual Machine) 1. Double Click on CD-ROM Devices. Select Use ISO image when you planning to install OS through ISO image otherwise Use physical drive. 2. Click on Start the Virtual Machine. 3. Select Solaris Interactive (Default) installation 4. Configure keyboard Layout screen appear (press F2) 5 Select a language English 6 welcome screens appear (Click Next) 7 Select network connectivity and (Click Next). 8. DHCP Screen appears select No and (Click Next) 9. Type Host Nome and (Click Next) ------RAC1 10. Type IP Address and (Click Next) ------192.168.0.111 11. Type NETMASK (Select default) and (Click Next) 12. Select No for Enable Ipv6 for pcn0 and (Click Next) 13. Select None for Default Route and (Click Next) 14. Select No for Enable Kerberos security and (Click Next) 15. Select None for Name Services and (Click Next) 16. NFSv4 Domain Name select Default and (Click Next) 17. Select Geographic Time Zones and (Click Next) 18. Continent and Country (INDIA) and (Click Next) 19. Accept the default date and time and (Click Next) 20. Enter ROOT password and (Click Next) 21. Select Yes for Enabling Remote Services and (Click Next) 22. Confirm Information and (Click Next) 23 Select default Install option and (Click Next) After that the system is being analyzed. Please wait screen appear and select the Type of Installation. Now Installer Install the OS Section: C (Create Shared Storage and Configure)
14
Create virtual disks for storage usage, which is shared by Both Machine 1. down the virtual Machine (RAC1) 2. Go to VMware Server Console. Click on Edit virtual machine settings. Virtual Machine Settings: Click on Add. Add Hardware Wizard: Click on Next. Hardware types: Select Hard Disk. Select a Disk: Disk: Select create a new virtual disk. Select a Disk Type: Virtual Disk Type: Select IDE (Recommended). Specify Disk Capacity: Disk capacity: Enter 10GB. Select Allocate all disk space now. Specify Disk File: Disk file: Enter F :\> SHARED-STORAGE\DISK1.vmdk. Click on Advanced Add Hardware Wizard: Virtual device node: Select IDE0.1. Mode: Select Independent, Persistent for all shared disks. 3. Click on Finish. Modify virtual machine configuration file. Additional parameters are required to enable disk sharing between the two virtual RAC nodes. Open the configuration file, F:\>SUNOS-1\ Solaris 10.vmx.vmx and add the bold parameters listed below. . . . priority.grabbed = "normal" priority.ungrabbed = "normal" disk.locking = "FALSE" diskLib.dataCacheMaxSize = "0" ide0:1.sharedBus = "virtual" ide0:0.redo = "" ethernet0.addressType = "generated ethernet0.connectionType = "hostonly" ide0:1.present = "TRUE" ide0:1.fileName = "E:\SHARED-DISK.vmdk" ide0:1.redo = "" checkpoint.vmState = ""
15
ide0:1.mode = "independent-persistent" ide0:1.deviceType= "disk" floppy0.present = "FALSE" Section: D (Create and Configure the Second Virtual Machine) 1. Create the windows folders to house the second virtual machines. E:\>mkdir RAC2 2. Shutdown the First Virtual Machine 3. Copy all the files from F:\RAC1 to E:\RAC2 4. Open VMware Server Console, press CTRL-O to open the second virtual machine, E:\RAC2\Solaris 10.vmx. 5. Rename the second virtual machine name from RAC1 to RAC2. Click on Start this virtual machine to start RAC2, leaving RAC1 powered off. RAC2 Virtual Machine: Select create a new identifier. 6. Log in as the root user and modify the network configuration. Follow below step for modifying Host Name and IP $ ifconfig <Ethernat> <new IP> (eg: $ifconfig pcn0 192.168.0.222) $ ifconfig <Ethernet> up (eg: $ifconfig pcn0 up) $ go to /etc/hosts file and change IP and host (eg 192.168.0.222 rac2 rac2) $ go to /etc/nodenames file and change host name (eg: RAC2) $ go to /etc/hostname.<Ethernet> (eg: /etc/hostname.pcn0) and change host name (eg: RAC2) 7. Restart the Second Virtual Machine. 8. Start the First virtual Machine 9. Verify all changes and Enjoy. Section: E (Add Ether Net Adapter for Private Network in Both Machine) 1. Put following entry in /etc/hosts file on Both Node. 192.168.0.111rac1 rac1 192.168.0.222rac2 rac2 192.168.0.11 rac1-priv 192.168.0.22 rac2-priv 192.168.0.50 rac1-vip 192.168.0.60 rac2-vip
rac1-priv rac2-priv rac1-vip rac2-vip
2. Power off both machines (Note: Complete step 3 to 8 on both node one by one) 3. Click on Edit virtual machine setting 4. Click Add button and select Ethernet Adapter 5. Select Host-Only Network type
16
6. Start the Both Machine and check network setting on both machines. 7. Go to RAC1 node: (Plumb Ethernet) $ifconfig pcn1 plumb 192.168.137.11 netmask 255.255.255.0 $ifconfig pcn1 up $ go to /etc/hostname.pcn1 and change host name (eg: rac1-priv) $ifconfig a (This will show all Ethernet Adapter) 8. Go to RAC2 node: (Plumb Ethernet) $ifconfig pcn1 plumb 192.168.137.222 netmask 255.255.255.0 $ifconfig pcn1 up $ go to /etc/hostname.pcn1 and change host name (eg: rac2-priv) $ifconfig a (This will show all Ethernet Adapter) 9. Power on both machine and verify your network setting Section: F (Prepare Disk for OCR, Voting and ASM Storage) 1. Run bellow mention command on both node $devfsadm 2. Log on to RAC1 node and complete bellow mention procedure. Here we will create following partition: c0d1s0 for OCR Disk c0d1s1 for Voting Disk c0s1s3 form ASM Disk c0s1s4 from ASM Disk c0s1s5 from ASM Disk 3. Complete bellow mention steps for creating partition $format
17
18
19
4. Set ownership of Disk (Note: This will be done after creating Oracle user and dba group) Default owner is root:sys needs to be changed to oracle:dba .We need to set oracle:dba ( here oracle is a Oracle owner and dba is a group) ownership of the disk. Check ownership of Disk. Execute following command for all slice as a root user. Example: $ ls -lhL /dev/rdsk/c0d1s0 crw-r----- 1 root sys 118, 64 Feb 16 02:10 /dev/rdsk/c0d1s0 $ chown oracle:dba /dev/rdsk/c0d1s0 Task 2 (Install Oracle Clusterware) Section: A (Check All Pre-request for Clusterware installation) Step 1: Check Hardware Requirement RAM should be at least 1GB How to Check? Use following command: $ /usr/sbin/prtconf | grep "Memory size" Swap Size: If RAM more then 2GB then Swap should be equal of RAM and If RAM between 1 to 2 gb then Swap should be 1.5 times of RAM How to Check? Use following command: # /usr/sbin/swap s or log in as a ROOT user and execute following command: $swap s or swap l How to ADD Swap Space? Method A 1. create a file system with desire space (eg: /c0d2s0) 2. Add the bellow mention line in /etc/vfstab file /dev/dsk/c1t0d0s3 --swap -no Method B 1. create a file system with desire space (eg: /c0d2s0) 2. Use the swap -a command to add additional swap area. $swap -a /dev/dsk/c1t0d0s3 Method C
20
Swap files can be used when you need to add swap space and do not have a free partition to use. To add a swap file, complete the following steps: 1. Create a 1G swap file named swapfile in the partition that have enough free space, for example /export/data directory. # mkfile 1000m /export/data/swapfile 2. Add the swap file to the systems swap space. # swap -a /export/data/swapfile 3. List the details of the modified system swap space with swap -l 4. List a summary of the modified system swap space with swap -s 5. To use a swap file when the system is subsequently rebooted, add an entry for the swap file in the /etc/vfstab file. /export/data/swapfile --swap -no Temp Space: Minimum 400 MB of disk space required in the /tmp directory How to check? Use following command: # df -k /tmp Check system architecture How to Check? Use following command: # /bin/isainfo kv Result should be: 64-bit SPARC installation: 64-bit sparcv9 kernel modules 32-bit x86 installation: 32-bit i386 kernel modules 64-bit x86 installation: 64-bit amd64 kernel modules Step 2: Check Required Packages How to check? Use following command: # pkginfo -i SUNWarc SUNWbtool SUNWhea SUNWlibm SUNWlibms SUNWsprot SUNWsprox SUNWtoo SUNWi1of SUNWi1cs SUNWi15cs SUNWxwfnt If a package that is required for your system architecture is not installed, then install it. Step 3: Check Patches
21
Refer to your operating system or software documentation for information about installing packages Section B (Create Oracle User and Group) Step 1 Create Oracle Users and Group (in all Nodes) Log in as a root and execute: # groupadd oinstall # groupadd dba # mkdir -p /export/home/oracle # useradd -d /export/home/oracle -g oinstall -G dba oracle # ac # passwd oracle New Password: Re-enter new Password: passwd: password successfully changed for oracle Section C (Configure SSH) Step 1 Create .SSH, and Create RSA Keys on Each Node Log in as an Oracle/CRS user and execute bellow steps $ mkdir ~/.ssh $ chmod 700 ~/.ssh $ /usr/bin/ssh-keygen -t rsa At the prompts: Accept the default location for the key file (press Enter). Enter and confirm a pass phrase unique for this installation user. Step 2 Add All Keys to a Common authorized_keys File On the primary node (RAC1), change directories to the .ssh directory. Then, add the RSA key to the authorized_keys file. $ cd .ssh $ cat id_rsa.pub >> authorized_keys $ ls In the .ssh directory, you should see the id_rsa.pub keys that you have created, and the file authorized_keys. Step 3 Copy authorized_key file to all cluster node (RAC2)
22
$ scp authorized_keys RAC2:/export/home/oracle/.ssh/ The authenticity of host 'rac2 (192.168.0.222)' can't be established. RSA key fingerprint is c0:1f:20:34:54:b2:cd:9f:42:f2:d6:25:36:2f:3e:db. Are you sure you want to continue connecting (yes/no)? yes Warning: Permanently added 'rac2,192.168.0.222' (RSA) to the list of known hosts. Password: authorized_keys 100% |*****************************| 221 00:00 Step 4 Log in on second node (RAC2) and Add the RSA keys for the second node (RAC2) to the authorized_keys file: $cd .ssh $cat id_rsa.pub >> authorized_keys Step 5 Copy authorized_key file to all cluster node (RAC2) $ scp authorized_keys RAC1:/export/home/oracle/.ssh/ Step 6 Enabling SSH User Equivalency on Cluster Member Nodes. On the system where you want to run OUI, log in as the ORACLE/CRS user and execute: $ $ $ $ ssh RAC1 date ssh RAC2 date ssh RAC1-PRIV date ssh RAC2-PRIV date Section D (Specify default gateway) specifying the default gateway to some dummy ip address (same subnet) in both of virtual machines. $ vi /etc/defaultrouter and add the IPADDRESS of your router. EXAMPLE: vi /etc/defaultrouter 192.168.0.100 reboot to test. or netstat -r (This will show the current routing table) Another Example: route add destination gateway Example:
23
# route add default 192.168.0.100 netstat r (This will now show your 'new' default gateway). Section E (Install Oracle Clusterware) Now Start the Oracle Cluster Ware Installation. Step 1 Start run installer. #./runInstaller If you see any error (PRKC-1044) regarding SSH. Please use bellow mention solution. Solution 1: you can use bello mention command. # ./runInstaller remoteshell /usr/bin/ssh Solution 2(permanent Solution): Complete bellow mentions steps. #go to /usr and make directory local and go to local directory and create directory bin #create symbolic link #ln s /usr/bin/ssh /usr/local/bin/ssh #ln s /usr/bin/scp /usr/local/bin/scp Step 2 Specify Inventory directory and credentials Step 3 Specify CRS Home and Path Details
24
Step 4 Installer Check pre-request. (This will show two warning, leave it)
Step 5 Specify cluster configuration. Click on Add button and specify public node name, private node name and virtual node name details for all RAC nodes.
25
Step 6 Specify network interface usage. At least one interface (pcn0) should be public.
Step 7 Specify OCR locations
26
Step 8 Specify Voting Disk locations (Choose External Redundancy /dev/rdsk/c0d1s1) Step 9 Click on Install button. Step 10 you need to execute scripts as root user, which will be appearing in screen.
27
Note: When you will execute ROOT.SH script on second node (RAC2), it will run VIPCA in silent mode and configure VIP. We can also run the following Command in Second node as root user. $CRS_HOME/bin/vipca for configuring VIP
Step 11Verify status of CRS services. Execute bellow command:

$/export/home/oracle/oracle/product/10.2.0/crs/bin/crs_stat t OR $ srvctl status nodeapps -n rac1 $ srvctl status nodeapps -n rac2 Task 3 (Install Oracle 10gR12 Software/binary) 2.5 GB of disk space should be required by Oracle Software 1.3 GB of disk space should be required by General Purpose Database. Here we have created two mount point /oracle and /database for oracle software and database.
28
Note: No specified operating system patches are required with Solaris 10 OS. Per-requisite Steps: Make sure that following software packages has been installed. SUNWarc SUNWbtool SUNWhea SUNWlibm SUNWlibms SUNWsprot SUNWtoo SUNWilof SUNWxwfnt SUNWilcs SUNWsprox SUNWil5cs we can verify that packages are installed or not by using following command: $pkginfo -i. Check following executable file must be presents in /usr/ccs/bin make ar il nm Checks swap space Swap space should be 512MB or Twice the size of RAM. Use following command to know about Physical Memory and Swap space: $ /usr/sbin/prtconf grep size $ /usr/sbin/swap l Need at least 400 MB of free space in /tmp directory. Check kernel parameter Set following kernel parameter in /etc/system file and reboot the server. set shmsys:shminfo_shmmax=4294967295
29
set shmsys:shminfo_shmmni=100 set semsys:seminfo_semmsl=256 set semsys:seminfo_semmni=100 Set X window enveiroment. Log in as a root with CDE (Common Desktop Environment Session) $ DISPLAY=:0.0 $export DISPLAY $xhost + $ su oracle $DISPLAY=:0.0 $export DISPLAY $/usr/openwin/bin/xclock Execute runInstaller Step 1 Log in as a ORACLE user and execute run installer. $./runInstaller If you see any error (PRKC-1044) regarding SSH. Please use mention solution. Solution 1: you can use bello mention command. # ./runInstaller remoteshell /usr/bin/ssh Solution 2: Complete bellow mentions steps. #go to /usr and make directory local and go to local directory and create directory bin #create symbolic link ( #ln s /usr/bin/ssh /usr/local/bin/ssh)
30
Click on Next Button Step 2 Select Installation type. Select Enterprise Edition and click on next button.
Step 3 Select Oracle home and Installation Path.
31
Step 4 Specify Hardware Cluster Installation Mode. Select Cluster installation and Select both RAC1, RAC2 nodes
32
Step 5 Select configuration option. Select software only.
33
Task 4 (Configure Oracle Listener) The Network Configuration Assistant (NETCA) should only be launched and configured on one node. At the end of the configuration process, the NETCA starts up the Oracle listener on both nodes. Step 1 Log in as a ORACLE user in RCA1 node and set following environment and invoke DBCA: $export ORACLE_HOME=/export/oracle/oracle/product/10.2.0/db $export PATH=$ORACLE_HOME/bin:$PATH $netca Step 2 Oracle Net Configuration Assistant: Real Application Clusters, Configuration: Select "Cluster configuration" Step 3 Oracle Net Configuration Assistant: TOPSNodes: Click "Select all nodes" Step 4 Oracle Net Configuration Assistant: Welcome:
34
Select "Listener configuration" Step 5 Oracle Net Configuration Assistant: Listener Configuration, Listener: Select "Add" Step 6 Oracle Net Configuration Assistant: Listener Configuration, Listener Name: Listener Name: LISTENER Step 7 Oracle Net Configuration Assistant: Listener Configuration, Select Protocols Selected Protocols: TCP Step 8 Oracle Net Configuration Assistant: Listener Configuration, TCP/IP Protocol: Select "Use the standard port number of 1521" Step 9 Oracle Net Configuration Assistant: Listener Configuration, More Listeners? Select "No" Step 10 Oracle Net Configuration Assistant: Listener Configuration Done: Click on "Next" Step 11 Oracle Net Configuration Assistant: Welcome Select "Naming Methods configuration" Click on "Next" Step 12 Oracle Net Configuration Assistant: Naming Methods Configuration, Select Naming: Select "Local Naming"
35
Step 13 Oracle Net Configuration Assistant: Naming Methods Configuration Done: Click on "Next" Step 14 Oracle Net Configuration Assistant: Welcome Click on "Finish" Step 15 Verify status of services. $/export/home/oracle/oracle/product/10.2.0/crs/bin/crs_stat t Tack 5 (Create and Configure ASM Instance and ASM DISK Groups) We can use two methods DBCA or Manual, for creating ASM Instance and Disk group. Here I will describe both methods. You can choose one of them. By DBCA Step 1 log in as a ORACLE user , set following environment and invoke DBCA : $export ORACLE_HOME=/export/oracle/oracle/product/10.2.0/db $export PATH=$ORACLE_HOME/bin:$PATH $dbca Step 2 Choose oracle RAC database configurations.
36
Step 3 Choose Configure Automatic Storage Management
Step 4 Choose BOTH NODE.
37
Step 5 Choose As per screen Shot
Step 6 Clicks on Create New.
38
Step 7 Create Disk Group as per screen shot.
Step 8 Choose As per screen Shot. Create Disk Group as per screen shot.
39
Step 9 oracle Automatically create ASM, ASM Disk group and mount the Instance and Disk group on both node. Step 10 Verify status of services. $/export/home/oracle/oracle/product/10.2.0/crs/bin/crs_stat t By MANUAL Step 1 Create parameter files (int+ASM1.ora) for ASM instance on the first node. Create init+ASM1.ora file for ASM Instance and put it default location. Cluser_database=true asm_diskgroups='DATA',RECOVERY #asm_diskstring='/dev/rdsk/c0d1s*' background_dump_dest=/export/home/oracle/oracle/admin/+ASM/bdump core_dump_dest==/export/home/oracle/oracle/admin/+ASM/cdump user_dump_dest==/export/home/oracle/oracle/admin/+ASM/udump instance_type=asm large_pool_size=16M remote_login_passwordfile=exclusive +ASM1.instance_number=1 +ASM2.instance_number=2 Step 2 Create password file for the ASM instance on first node. Using the orapwd utility, create an orapw+ASM1 file in $ORACLE_HOME/dbs on the second node. $orapwd file=orapwd+ASM1 password=sys entries=5 Step 3 Start the first ASM instance (+ASM1). $ export ORACLE_SID=+ASM1 $ sqlplus / as sysdba SQL*Plus: Release 10.2.0.2.0 - Production on Thu Oct 26 18:43:14 2009 Copyright (c) 1982, 2004, Oracle. All rights reserved. Connected to an idle instance. SQL> startup nomount ASM instance started Total System Global Area 125829120 bytes Fixed Size 769268 bytes
40
Variable Size 125059852 bytes Database Buffers 0 bytes Redo Buffers 0 bytes ORA-15110: no diskgroups mounted SQL> Step 4 create Disk Group SQL> create diskgroup DATA normal redundancy failgroup one disk '/dev/rdsk/c0d1s3' failgroup two disk '/dev/rdsk/c0d1s4';
Diskgroup created.
SQL> create diskgroup RECOVERY external redundancy disk '/dev/rdsk/c0d1s5'; Diskgroup created. Step 5 Create parameter files (int+ASM2.ora) for ASM instance on the second node. Create init+ASM1.ora file for ASM Instance and put it default location. Cluser_database=true asm_diskgroups='DATA',RECOVERY #asm_diskstring='/dev/rdsk/c0d1s*' background_dump_dest=/export/home/oracle/oracle/admin/+ASM/bdump core_dump_dest==/export/home/oracle/oracle/admin/+ASM/cdump user_dump_dest==/export/home/oracle/oracle/admin/+ASM/udump instance_type=asm large_pool_size=16M remote_login_passwordfile=exclusive +ASM1.instance_number=1 +ASM2.instance_number=2 Step 6 Create password file for the ASM instance on second node. Using the orapwd utility, create an orapw+ASM2 file in $ORACLE_HOME/dbs on the second node. $orapwd file=orapwd+ASM2 password=sys entries=5 Step 7 Start the second ASM instance (+ASM2). $ export ORACLE_SID=+ASM2 $ sqlplus / as sysdba SQL*Plus: Release 10.2.0.2.0 - Production on Thu Oct 26 18:43:14 2009
41
Copyright (c) 1982, 2004, Oracle. All rights reserved. Connected to an idle instance. SQL> startup nomount ASM instance started Total System Global Area 125829120 bytes Fixed Size 769268 bytes Variable Size 125059852 bytes Database Buffers 0 bytes Redo Buffers 0 bytes ORA-15110: no diskgroups mounted SQL> Step 8 Register the ASM instances with CRS. For higher availability, register the ASM instances with the CRS. When registered, the CRS should detect any failed instances and automatically attempt to start up the instances. The CRS should also automatically start up the instances when the servers are rebooted. On node RCA1: $ srvctl add asm -n rac1 -i +ASM1 -o /export/home/oracle/oracle/product/10.2.0/db $ srvctl start asm -n rac1 $ srvctl status asm -n rac1 ASM instance +ASM1 is running on node rac1. On node RAC2: $ srvctl add asm -n rac2 -i +ASM2 -o /export/home/oracle/oracle/product/10.2.0/db $ srvctl start asm -n rac2 $ srvctl status asm -n rac2 ASM instance +ASM2 is running on node rac2. Tack 6 (Create Database) We can use two methods DBCA or Manual, for creating RAC Enable Database. Here I will describe both methods. You can choose one of them. By DBCA Step 1 log in as a ORACLE user , set following environment and invoke DBCA : $export ORACLE_HOME=/export/oracle/oracle/product/10.2.0/db
42
$export PATH=$ORACLE_HOME/bin:$PATH $dbca Step 2 Follow Screen Shot
Step 3 Follow Screen Shot
43
44
45
46
47
48
49
Step 17 After creating Database. Verify the Services. $/export/home/oracle/oracle/product/10.2.0/crs/bin/crs_stat t By MANUAL Step 1 log in on first node (RAC1) as oracle user and create parameter file (initPROD1.ora) in default location with following parameter: *.audit_file_dest='/export/home/oracle/oracle/product/10.2.0/db/admin/PROD/ad ump' *.background_dump_dest='/export/home/oracle/oracle/product/10.2.0/db/admin /PROD/bdump' #*.cluster_database=false *.compatible='10.2.0.2.0' *.core_dump_dest='/export/home/oracle/oracle/product/10.2.0/db/admin/PROD/ cdump' *.db_block_size=8192 *.db_create_file_dest='+DATA' *.db_domain='' *.db_file_multiblock_read_count=16 *.db_name='PROD' *.sga_target = 250M *.job_queue_processes=10 *.log_checkpoints_to_alert=TRUE *.pga_aggregate_target=100M *.processes=500 *.remote_listener='LISTENERS_PROD'
50
*.remote_login_passwordfile='exclusive' *.sessions=200 *.undo_management='AUTO' *.undo_tablespace='UNDOTBS1' *.user_dump_dest='/export/home/oracle/oracle/product/10.2.0/db/admin/PROD/ udump' #PROD1.instance_name = PROD1 #*.control_file=+DATA Step 2 Create password file for the PROD1 instance on first node. Using the orapwd utility, create an orapw+PROD1 file in $ORACLE_HOME/dbs on the second node. $orapwd file=orapwd+PROD1 password=sys entries=5 Step 3 Now go to $ORACLE_HOME/network/admin and edit tnsnames.ora to add entries for your instances, the database and the listeners. The listeners_prod entry has to look like this: LISTENERS_PROD = (ADDRESS_LIST = (ADDRESS = (PROTOCOL = TCP)(HOST = rac1-vip)(PORT = 1521)) (ADDRESS = (PROTOCOL = TCP)(HOST = rac2-vip)(PORT = 1521)) ) Step 4 Start the instance and create the database on first node (RAC1) $export ORACLE_HOME=/export/home/oracle/oracle/product/10.2.0/db $export PATH=$ORACLE_HOME/bin:$PATH $export ORACLE_SID=PROD1 SQL> CREATE DATABASE PROD MAXINSTANCES 8 MAXLOGHISTORY 100 MAXLOGFILES 64 MAXLOGMEMBERS 3 MAXDATAFILES 150 DATAFILE SIZE 300M AUTOEXTEND ON NEXT 10240K MAXSIZE 1024M EXTENT MANAGEMENT LOCAL SYSAUX DATAFILE SIZE 200M AUTOEXTEND ON NEXT 10240K MAXSIZE 800M DEFAULT TEMPORARY TABLESPACE TEMP TEMPFILE SIZE 200M AUTOEXTEND ON NEXT 10M MAXSIZE 1000M EXTENT MANAGEMENT LOCAL
51
UNDO TABLESPACE UNDOTBS1 DATAFILE SIZE 200M AUTOEXTEND ON NEXT 10M MAXSIZE 1000M CHARACTER SET WE8ISO8859P1 NATIONAL CHARACTER SET AL16UTF16 LOGFILE GROUP 1 SIZE 50M, GROUP 2 SIZE 50M, GROUP 3 SIZE 50M ; Step 5 Run following scripts: SQL>@$ORACLE_HOME/rdbsm/admin/catalog.sql SQL>@$ORACLE_HOME/rdbsm/admin/catproc.sql SQL>@$ORACLE_HOME/rdbsm/admin/catclust.sql Note: at this point Database created; next we will convert into RAC. Step 6 Edit initPROD1.ora and add the cluster parameters: *.cluster_database_instances=2 *.cluster_database=true PROD1.instance_number=1 PROD2.instance_number=2 PROD2.thread=2 PROD1.thread=1 PROD1.undo_tablespace='UNDOTBS1' PROD2.undo_tablespace='UNDOTBS2' PROD1.instance_name = PROD1 PROD1.instance_name = PROD2 Step 7 Copy initPROD1.ora and tnsnames.ora file to node2 (RAC2) and rename initPROD1.ora to initPROD2.ora. Step 8 Now create the second undo tablespace: SQL> Create undo tablespace undotbs2 datafile size 200M; Tablespace created Step 9 Create the second instance's redo log thread: SQL> alter database add logfile thread 2 group 4 size 50M, group 5 size 50M, group 6 size 50M; Database altered
52
Step 10 Issue a "shutdown immediate" now on RAC1, and then start the instance. It should come up ok, Step 11 Now activate the 2nd redo log thread: SQL> alter database enable public thread 2; Step 12 Finally start your second instance. You have a proper RAC database up and running now. Check v$active_instances to make sure both appear in that view. Step 13 Shut down both instances next and register the database and its instances with Clusterware as oracle: $ srvctl add databaese -d PROD -o /export/home/oracle/oracle/product/10.2.0/db $ srvctl add instance -d PROD -i PROD1 -n rac1 $ srvctl add instance -d PROD -i PROD2 -n rac2 Step 14 Check crs_stat to see if that worked and start the database: $ srvctl start database -d PROD Task 7 (Setup Test Transport Failover) Oracle TAF enables any failed database connections to reconnect to another node within the cluster. The failover is transparent to the user. Oracle reexecutes the query on the failed over instance and continues to display the remaining results to the user. Step 1 Create a new database service. Lets begin by creating a new service called CRM. Database services can be created using either DBCA or the srvctl utility. Here you will use DBCA to create the CRM service on PROD1. Service Name :CRM Database Name:PROD Preferred Instance : PROD1 Available Instance : PROD2 TAF Policy :BASIC Log in as a oracle user in RAC1 node and execute:
53
rac1-> dbca Welcome: Select Oracle Real Application Clusters database. Operations: Select Services Management. List of cluster databases: Click on Next. Database Services: Click on Add. Add a Service: Enter CRM. Select prod1 as the Preferred instance. Select prod2 as the Available instance. TAF Policy: Select Basic. Click on Finish. Database Configuration Assistant: Click on No to exit. Note: The Database Configuration Assistant creates the CRM service name entry in tnsnames.ora file. Step 2 Check Service name String SQL> connect system/oracle@prod1 Connected. SQL> show parameter service NAME TYPE VALUE ----------------------------- ----------- -----------------------service_names string prod, CRM SQL> connect system/oracle@prod2 Connected. SQL> show parameter service NAME TYPE VALUE --------------------------------------- -----------------------service_names string prod Step 3 Connect the first session using the CRM service. If the returned output of failover_type and failover_mode is 'NONE', verify that the CRM service is configured correctly in tnsnames.ora. SQL> connect system/oracle@crm Connected. SQL> select instance_number instance#, instance_name, host_name, status
54
from v$instance; INSTANCE# ---------1 INSTANCE_NAME ---------------prod1 HOST_NAME STATUS --------------------- -----------rac1 OPEN
SQL> select failover_type, failover_method, failed_over from v$session where username='SYSTEM'; FAILOVER_TYPE ------------SELECT FAILOVER_METHOD --------------BASIC FAILED_OVER ---------------NO
Step 4 Shut down the instance from another session. Connect as the sys user on CRM instance and shut down the instance. rac1-> export ORACLE_SID=prod1 rac1-> sqlplus / as sysdba SQL> select instance_number instance#, instance_name, host_name, status from v$instance; INSTANCE# ---------1 INSTANCE_NAME ---------------prod1 HOST_NAME STATUS -------------------------------rac1 OPEN
SQL> shutdown abort; ORACLE instance shut down. Verify that the session has failed over. From the same CRM session you opened previously, execute the queries below to verify that the session has failed over to another instance. SQL> select instance_number instance#, instance_name, host_name, status from v$instance; INSTANCE# ---------2 INSTANCE_NAME ---------------prod2 HOST_NAME STATUS --------------------- -----------rac2 OPEN
SQL> select failover_type, failover_method, failed_over from v$session where username='SYSTEM'; FAILOVER_TYPE FAILOVER_METHOD FAILED_OVER
55
------------SELECT
--------------BASIC
---------------YES
Step 5 Relocate the CRM service back to the preferred instance. After PROD1 is brought back up, the CRM service does not automatically relocate back to the preferred instance. You have to manually relocate the service to prod1. rac1-> export ORACLE_SID=prod1 rac1-> sqlplus / as sysdba SQL> startup ORACLE instance started. Total System Global Area 209715200 bytes Fixed Size 1218556 bytes Variable Size 104859652 bytes Database Buffers 100663296 bytes Redo Buffers 2973696 bytes Database mounted. Database opened. SQL> show parameter service NAME TYPE VALUE --------------------------------------- -----------------------service_names string prod rac2-> export ORACLE_SID=prod2 rac2-> sqlplus / as sysdba SQL> show parameter service NAME TYPE VALUE --------------------------------------- -----------------------service_names string prod, CRM rac1-> srvctl relocate service -d prod -s crm -i prod2 -t prod1 SQL> connect system/oracle@prod1 Connected. SQL> show parameter service
56
NAME TYPE VALUE --------------------------------------- -----------------------service_names string prod, CRM SQL> connect system/oracle@devdb2 Connected. SQL> show parameter service NAME TYPE VALUE ----------------------------- ----------- -----------------------service_names string prod
Implement 10gR2 Oracle RAC on Linux with VMware

Here, you will learn how to install and configure two nodes on running Oracle RAC 10g Release 2 on Enterprise Linux and VMware Server. Note that this guide is intended for educational/evaluation purposes only; neither Oracle nor any other vendor will support this configuration. The guide is structured into the following sections: 1. 2. 3. 4. 5. 6. 7. Hardware Requirements and Overview Configure the First Virtual Machine Configure Enterprise Linux on the First Virtual Machine Create and Configure the Second Virtual Machine Configure Oracle Automatic Storage Management (ASM) Configure Oracle Cluster File System (OCFS2) Install Oracle Clusterware 8. Install Oracle Database 10g Release 2 9. Explore the RAC Database Environment
57
10. Test Transparent Application Failover (TAF) 11. Database Backup and Recovery 12. Explore Oracle Enterprise Manager (OEM) Database Console 13. Common Issues Section: A Hardware Requirements and Overview Allocate a minimum of 700MB of memory to each virtual machine; reserve a minimum of 30GB of disk space for all the virtual machines. Host Name indiandb a OS Windows XP Professional Service Pack 2 (32-bit) Processor Intel Pentium 4 Memory 2 GB Disk 250 GB,
An overview of guest operating system environment: Host Name RAC1 RAC2 OS Enterprise Linux 4 (32-bit) Enterprise Linux 4 (32-bit) Processor 1 1 Memory 700 MB 700 MB
An overview of the virtual disk layout: Virtual Disk on Host Virtual Disk on Gust /dev/sda1 /dev/sda2 /dev/sda3 Virtual Devic e Node SCSI 0:0 Size (MB) 20 Descriptio n / mountpoin t Swap space Oracle binaries OCFS2 disk ASM disk group 1 ASM disk group 1
d:\vm\rac\localdisk.vmdk
d:\vm\rac\sharedstorage\ocfs2disk.vm dk d:\vm\rac\sharedstorage\asmdisk1.vm dk d:\vm\rac\sharedstorage\asmdisk2.vm dk
/dev/sdb /dev/sdc /dev/sdd
SCSI 1:0 SCSI 1:1 SCSI 1:2
512 307 2 307 2
58
d:\vm\rac\sharedstorage\asmdisk3.vm dk
/dev/sde
SCSI 1:3
204 8
ASM flash recovery area
An overview of the RAC database environment: Host Name RAC1 RAC2 ASM Instance Name +ASM1 +ASM2 RAC Instance Name devdb1 devdb2 Database Name devdb devdb Database File Storage ASM ASM OCR & Voting Disk OCFS2 OCFS2
Section B Configure the First Virtual Machine To create and configure the first virtual machine, you will add virtual hardware devices such as disks and processors. Before proceeding with the install, create the windows folders to house the virtual machines and the shared storage. D:\>mkdir vm\rac\rac1 D:\>mkdir vm\rac\rac2 D:\>mkdir vm\rac\sharedstorage Double-click on the VMware Server icon on your desktop to bring up the application: 1. Press CTRL-N to create a new virtual machine. 2. New Virtual Machine Wizard: Click on Next. 3. Select the Appropriate Configuration: a. Virtual machine configuration: Select Custom. 4. Select a Guest Operating System: a. Guest operating system: Select Linux. b. Version: Select Red Hat Enterprise Linux 4. 4. Name the Virtual Machine: a. Virtual machine name: Enter rac1. b. Location: Enter d:\vm\rac\rac1. 5. Set Access Rights: a. Access rights: Select Make this virtual machine private. 6. Startup / Shutdown Options: a. Virtual machine account: Select User that powers on the virtual machine. 7. Processor Configuration: a. Processors: Select One. 8. Memory for the Virtual Machine:
59
a. Memory: Select 700MB. 9. Network Type: a. Network connection: Select Use bridged networking. 10. Select I/O Adapter Types: a. I/O adapter types: Select LSI Logic. 11. Select a Disk: a. Disk: Select Create a new virtual disk. 12. Select a Disk Type: a. Virtual Disk Type: Select SCSI (Recommended). 13. Specify Disk Capacity: a. Disk capacity: Enter 20GB. b. Deselect Allocate all disk space now. To save space, you do not have to allocate all the disk space now. 14. Specify Disk File: a. Disk file: Enter localdisk.vmdk. b. Click on Finish. 15. Repeat steps 16 to 24 to create four virtual SCSI hard disks ocfs2disk.vmdk (512MB), asmdisk1.vmdk (3GB), asmdisk2.vmdk (3GB), and asmdisk3.vmdk (2GB). 16. VMware Server Console: Click on Edit virtual machine settings. 17. Virtual Machine Settings: Click on Add. 18. Add Hardware Wizard: Click on Next. 19. Hardware Type: a. Hardware types: Select Hard Disk. 19. Select a Disk: a. Disk: Select Create a new virtual disk. 20. Select a Disk Type: a. Virtual Disk Type: Select SCSI (Recommended). 21. Specify Disk Capacity: a. Disk capacity: Enter 0.5GB. b. Select Allocate all disk space now. You do not have to allocate all the disk space if you want to save space. For performance reason, you will pre-allocate all the disk space for each of the virtual shared disk. If the size of the shared disks were to grow rapidly especially during Oracle database creation or when the database is under heavy DML activity, the virtual machines may hang intermittently for a brief period or crash in a few rare occasions. 23. Specify Disk File: a. Disk file: Enter d:\vm\rac\sharedstorage\ocfs2disk.vmdk. b. Click on Advanced. 23. Add Hardware Wizard: a. Virtual device node: Select SCSI 1:0. b. Mode: Select Independent, Persistent for all shared disks. c. Click on Finish.
60
Finally, add an additional virtual network card for the private interconnects and remove the floppy drive, if any. 25. VMware Server Console: Click on Edit virtual machine settings. 26. Virtual Machine Settings: Click on Add. 27. Add Hardware Wizard: Click on Next. 28. Hardware Type: a. Hardware types: Ethernet Adapter. 28. Network Type: a. Host-only: A private network shared with the host b. Click on Finish. 29. Virtual Machine Settings: a. Select Floppy and click on Remove. 30. Virtual Machine Settings: Click on OK.
Modify virtual machine configuration file. Additional parameters are required to enable disk sharing between the two virtual RAC nodes. Open the configuration file, d:\vm\rac\rac1\Red Hat Enterprise Linux 4.vmx and add the bold parameters listed below. config.version = "8" virtualHW.version = "4" scsi0.present = "TRUE" scsi0.virtualDev = "lsilogic" memsize = "700" scsi0:0.present = "TRUE"
61
scsi0:0.fileName = "localdisk.vmdk" ide1:0.present = "TRUE" ide1:0.fileName = "auto detect" ide1:0.deviceType = "cdrom-raw" floppy0.fileName = "A:" Ethernet0.present = "TRUE" displayName = "rac1" guestOS = "rhel4" priority.grabbed = "normal" priority.ungrabbed = "normal" disk.locking = "FALSE" diskLib.dataCacheMaxSize = "0" scsi1.sharedBus = "virtual" scsi1.present = "TRUE" scsi1:0.present = "TRUE" scsi1:0.fileName = "D:\vm\rac\sharedstorage\ocfs2disk.vmdk" scsi1:0.mode = "independent-persistent" scsi1:0.deviceType = "disk" scsi1:1.present = "TRUE" scsi1:1.fileName = "D:\vm\rac\sharedstorage\asmdisk1.vmdk" scsi1:1.mode = "independent-persistent" scsi1:1.deviceType = "disk" scsi1:2.present = "TRUE" scsi1:2.fileName = "D:\vm\rac\sharedstorage\asmdisk2.vmdk" scsi1:2.mode = "independent-persistent" scsi1:2.deviceType = "disk" scsi1:3.present = "TRUE" scsi1:3.fileName = "D:\vm\rac\sharedstorage\asmdisk3.vmdk" scsi1:3.mode = "independent-persistent" scsi1:3.deviceType = "disk" scsi1.virtualDev = "lsilogic" ide1:0.autodetect = "TRUE" floppy0.present = "FALSE" Ethernet1.present = "TRUE" Ethernet1.connectionType = "hostonly"
62
Section C Install and Configure Enterprise Linux on the First Virtual Machine 1. On your VMware Server Console, double-click on the CD-ROM device on the right panel and select the ISO image for disk 1, Enterprise-R4-U4-i386disc1.iso. VMware Server console: Click on Start this virtual machine. Hit Enter to install in graphical mode. Skip the media test and start the installation. Welcome to enterprise Linux: Click on Next. Language Selection: <select your language preference>. Keyboard Configuration: <select your keyboard preference>. Installation Type: Custom. Disk Partitioning Setup: Manually partition with Disk Druid. Warning: Click on Yes to initialize each of the device sda, sdb, sdc, sdd, and sde. 10. Disk Setup: Allocate disk space on sda drive by double-clicking on /dev/sda free space for the mount points (/ and /u01) and swap space. You will configure the rest of the drives for OCFS2 and ASM later. Add Partition: Mount Point: / File System Type: ext3 Start Cylinder: 1 End Cylinder: 910 File System Type: Swap Start Cylinder: 911 End Cylinder: 1170 Mount Point: /u01 File System Type: ext3 Start Cylinder: 1171 End Cylinder: 2610
2.
3.
4.
5.
6. 7. 8. 9.
63
11. Boot Loader Configuration: Select only the default /dev/sda1 and leave the rest unchecked 12. Boot Loader Configuration: Select only the default /dev/sda1 and leave the rest unchecked. 13. Network Configuration: A Network Devices Select and edit eth0 1. De-select Configure Using DHCP. 2. Select Activate on boot. 3. IP Address: Enter 192.168.2.131. 4. Netmask: Enter 255.255.255.0. Select and edit eth1 1. De-select Configure Using DHCP. 2. Select Activate on boot. 3. IP Address: Enter 10.10.10.31. 4. Netmask: Enter 255.255.255.0. B Hostname Select manually and enter rac1.mycorpdomain.com. C Miscellaneous Settings Gateway: Enter 192.168.2.1. Primary DNS: <optional>
64
Secondary DNS: <optional> 13. Firewall Configuration: a) Select No Firewall. If firewall is enabled, you may encounter an error, mount.ocfs2: Transport endpoint is not connected while mounting when you attempt to mount ocfs2 file system later in the set up. b) Enable SELinux?: Active. 14. Warning No Firewall: Click on Proceed. 15. Additional Language Support: <select the desired language>. 16. Time Zone Selection: <select your time zone> 17. Set Root Password: <enter your root password> 18. Package Group Selection: a. Select X Window System. b. Select GNOME Desktop Environment. Select Editors. Click on Details and select your preferred text editor. c. d. Select Graphical Internet. e. Select Text-based Internet. f. Select Office/Productivity. g. Select Sound and Video. h. Select Graphics. i. Select Server Configuration Tools. j. Select FTP Server. Select Legacy Network Server. Click on Details. 1. Select rsh-server. 2. Select telnet-server. k. l. Select Development Tools. m. Select Legacy Software Development. n. Select Administration Tools. Select System Tools. Click on Details. Select the following packages in addition to the default selected packages. Select ocfs-2-2.6.9-42.0.0.0.1EL (driver for UP kernel), or select ocfs-2-2.6.9-42.0.0.0.1ELsmp (driver for SMP kernel). 1. 2. Select ocfs2-tools. 3. Select ocfs2console. Select oracle oracleasm-2.6.9-42.0.0.0.1EL (driver for UP kernel) or select oracleasm-2.6.9-42.0.0.0.1ELsmp (driver for SMP kernel). 4. 5. Select sysstat.
65
o. p. Select Printing Support. 18. 19. About to Install: Click on Next. 20. Required Install Media: Click on Continue. Change CD-ROM: On your VMware Server Console, press CTRL-D to bring up the Virtual Machine Settings. Click on the CD-ROM device and select the ISO image for disk 2, Enterprise-R4-U4-i386-disc2.iso, followed by the ISO image for disk 3, Enterprise-R4-U4-i386-disc3.iso. 21. At the end of the installation: On your VMware Server Console, press CTRL-D to bring up the Virtual Machine Settings. Click on the CD-ROM device and select Use physical drive. a. b. Click on Reboot. 23. Welcome: Click on Next. 24. License Agreement: Select Yes, I agree to the License Agreement. 25. Date and Time: Set the date and time. 26. Display: <select your desired resolution>. 27. System User: Leave the entries blank and click on Next. 28. Additional CDs: Click on Next. 29. Finish Setup: Click on Next.
66
Congratulations, you have just installed Enterprise Linux on VMware Server! Install VMware Tools. VMware Tools is required to synchronize the time between the host and guest machines. On the VMware Console, log in as the root user, 1. 2. 3. 4. Click on VM and then select Install VMware Tools. rac1 Virtual Machine: Click on Install. Double-click on the VMware Tools icon on your desktop. cdrom: Double-click on VMwareTools-1.0.1-29996.i386.rpm.
67
5. Completed System Preparation: Click on Continue. 6. Open up a terminal and execute vmware-config-tools.pl. Enter the desired display size. Synchronize Guest OS time with Host OS. When installing the Oracle Clusterware and Oracle Database software, the Oracle installer will initially install the software on the local node and then remotely copies the software to the remote node. If the date and time of both RAC nodes are not synchronized, you will likely receive errors similar to the one below. "/bin/tar: ./inventory/Components21/oracle.ordim.server/10.2.0.1.0: time stamp 2006-11-04 06:24:04 is 25 s in the future" To ensure a successful Oracle RAC installation, the time on the virtual machines has to synchronize with the host machine. Perform the steps below to synchronize the time as the root user.
1. Execute vmware-toolbox to bring up the VMware Tools Properties

window. Under the Options tab, select Time synchronization between the virtual machine and the host operating system. You should find the tools.syncTime = "TRUE" parameter appended to the virtual machine configuration file, d:\vm\rac\rac1\Red Hat Enterprise Linux 4.vmx.
68
2. Edit /boot/grub/grub.conf and add the options, "clock=pit nosmp noapic nolapic" to the line that reads kernel /boot/. You have added the options to both kernels. You are only required to make the change to your specific kernel. #boot=/dev/sda default=0 timeout=5 splashimage=(hd0,0)/boot/grub/splash.xpm.gz hiddenmenu title Enterprise (2.6.9-42.0.0.0.1.ELsmp) root (hd0,0) kernel /boot/vmlinuz-2.6.9-42.0.0.0.1.ELsmp ro root=LABEL=/ rhgb quiet clock=pit nosmp noapic nolapic initrd /boot/initrd-2.6.9-42.0.0.0.1.ELsmp.img title Enterprise-up (2.6.9-42.0.0.0.1.EL) root (hd0,0) kernel /boot/vmlinuz-2.6.9-42.0.0.0.1.EL ro root=LABEL=/ rhgb quiet clock=pit nosmp noapic nolapic initrd /boot/initrd-2.6.9-42.0.0.0.1.EL.img 3. Reboot rac1. # reboot Create the oracle user. As the root user, execute # groupadd oinstall # groupadd dba # mkdir -p /export/home/oracle /ocfs # useradd -d /export/home/oracle -g oinstall -G dba -s /bin/ksh oracle # chown oracle:dba /export/home/oracle /u01 # passwd oracle New Password: Re-enter new Password: passwd: password successfully changed for oracle Create the oracle user environment file. /export/home/oracle/.profile export PS1="`/bin/hostname -s`-> " export EDITOR=vi export ORACLE_SID=devdb1 export ORACLE_BASE=/u01/app/oracle export ORACLE_HOME=$ORACLE_BASE/product/10.2.0/db_1 export ORA_CRS_HOME=$ORACLE_BASE/product/10.2.0/crs_1 export LD_LIBRARY_PATH=$ORACLE_HOME/lib
69
export PATH=$ORACLE_HOME/bin:$ORA_CRS_HOME/bin:/bin: /usr/bin:/usr/sbin:/usr/local/bin:/usr/X11R6/bin umask 022 Create the filesystem directory structure. As the oracle user, execute rac1-> mkdir -p $ORACLE_BASE/admin rac1-> mkdir -p $ORACLE_HOME rac1-> mkdir -p $ORA_CRS_HOME rac1-> mkdir -p /u01/oradata/devdb Increase the shell limits for the Oracle user. Use a text editor and add the lines listed below to /etc/security/limits.conf, /etc/pam.d/login, and /etc/profile. Additional information can be obtained from the documentation. /etc/security/limits.conf oracle soft nproc 2047 oracle hard nproc 16384 oracle soft nofile 1024 oracle hard nofile 65536 /etc/pam.d/login session required /lib/security/pam_limits.so /etc/profile if [ $USER = "oracle" ]; then if [ $SHELL = "/bin/ksh" ]; then ulimit -p 16384 ulimit -n 65536 else ulimit -u 16384 -n 65536 fi fi Install Enterprise Linux software packages. The following additional packages are required for Oracle software installation. If you have installed the 64-bit version of Enterprise Linux, the installer should have already installed these packages. libaio-0.3.105-2.i386.rpm openmotif21-2.1.30-11.RHEL4.6.i386.rpm
Extract the packages from the ISO CDs and execute the command below as the root user. # ls libaio-0.3.105-2.i386.rpm openmotif21-2.1.30-11.RHEL4.6.i386.rpm
70
# # rpm -Uvh *.rpm warning: libaio-0.3.105-2.i386.rpm: V3 DSA signature: NOKEY, key ID b38a8516 Preparing... ########################################### [100%] 1:openmotif21 ########################################### [ 50%] 2:libaio ########################################### [100%] Configure the kernel parameters. Use a text editor and add the lines listed below to /etc/sysctl.conf. To make the changes effective immediately, execute /sbin/sysctl p. # more /etc/sysctl.conf kernel.shmall = 2097152 kernel.shmmax = 2147483648 kernel.shmmni = 4096 kernel.sem = 250 32000 100 128 fs.file-max = 65536 net.ipv4.ip_local_port_range = 1024 65000 net.core.rmem_default = 1048576 net.core.rmem_max = 1048576 net.core.wmem_default = 262144 net.core.wmem_max = 262144 Modify the /etc/hosts file. # more /etc/hosts 127.0.0.1 localhost 192.168.2.131 rac1.mycorpdomain.com rac1 192.168.2.31 rac1-vip.mycorpdomain.com rac1-vip 10.10.10.31 rac1-priv.mycorpdomain.com rac1-priv 192.168.2.132 rac2.mycorpdomain.com rac2 192.168.2.32 rac2-vip.mycorpdomain.com rac2-vip 10.10.10.32 rac2-priv.mycorpdomain.com rac2-priv Configure the hangcheck timer kernel module. The hangcheck timer kernel module monitors the system's health and restarts a failing RAC node. It uses two parameters, hangcheck_tick (defines the system checks frequency) and hangcheck_margin (defines the maximum hang delay before a RAC node is reset), to determine if a node is failing. Add the following line in /etc/modprobe.conf to set the hangcheck kernel module parameters. /etc/modprobe.conf options hangcheck-timer hangcheck_tick=30 hangcheck_margin=180
71
To load the module immediately, execute "modprobe -v hangcheck-timer". Create disk partitions for OCFS2 and Oracle ASM. Prepare a set of raw disks for OCFS2 (/dev/sdb), and for Oracle ASM (/dev/sdc, /dev/sdd, /dev/sde). On rac1, as the root user, execute # fdisk /dev/sdb Command (m for help): n Command action e extended p primary partition (1-4) p Partition number (1-4): 1 First cylinder (1-512, default 1): Using default value 1 Last cylinder or +size or +sizeM or +sizeK (1-512, default 512): Using default value 512 Command (m for help): w The partition table has been altered! Calling ioctl() to re-read partition table. Syncing disks. # fdisk /dev/sdc Command (m for help): n Command action e extended p primary partition (1-4) p Partition number (1-4): 1 First cylinder (1-391, default 1): Using default value 1 Last cylinder or +size or +sizeM or +sizeK (1-391, default 391): Using default value 391 Command (m for help): w The partition table has been altered! Calling ioctl() to re-read partition table. Syncing disks. # fdisk /dev/sdd Command (m for help): n Command action e extended p primary partition (1-4)
72
p Partition number (1-4): 1 First cylinder (1-391, default 1): Using default value 1 Last cylinder or +size or +sizeM or +sizeK (1-391, default 391): Using default value 391 Command (m for help): w The partition table has been altered! Calling ioctl() to re-read partition table. Syncing disks. # fdisk /dev/sde Command (m for help): n Command action e extended p primary partition (1-4) p Partition number (1-4): 1 First cylinder (1-261, default 1): Using default value 1 Last cylinder or +size or +sizeM or +sizeK (1-261, default 261): Using default value 261 Command (m for help): w The partition table has been altered! Calling ioctl() to re-read partition table. Syncing disks. # fdisk -l Disk /dev/sda: 21.4 GB, 21474836480 bytes 255 heads, 63 sectors/track, 2610 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Device Boot Start End Blocks Id System /dev/sda1 * 1 910 7309543+ 83 Linux /dev/sda2 911 1170 2088450 82 Linux swap /dev/sda3 1171 2610 11566800 83 Linux Disk /dev/sdb: 536 MB, 536870912 bytes 64 heads, 32 sectors/track, 512 cylinders Units = cylinders of 2048 * 512 = 1048576 bytes Device Boot Start End Blocks Id System /dev/sdb1 1 512 524272 83 Linux Disk /dev/sdc: 3221 MB, 3221225472 bytes 255 heads, 63 sectors/track, 391 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Device Boot Start End Blocks Id System /dev/sdc1 1 391 3140676 83 Linux Disk /dev/sdd: 3221 MB, 3221225472 bytes
73
255 heads, 63 sectors/track, 391 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Device Boot Start End Blocks Id System /dev/sdd1 1 391 3140676 83 Linux Disk /dev/sde: 2147 MB, 2147483648 bytes 255 heads, 63 sectors/track, 261 cylinders Units = cylinders of 16065 * 512 = 8225280 bytes Device Boot Start End Blocks Id System /dev/sde1 1 261 2096451 83 Linux Install oracleasmlib package. Download the ASM library from OTN and install the ASM RPM as the root user. # rpm -Uvh oracleasmlib-2.0.2-1.i386.rpm Preparing... ########################################### [100%] 1:oracleasmlib ########################################### [100%] At this stage, you should already have the following ASM packages installed. [root@rac1 swdl]# rpm -qa | grep oracleasm oracleasm-support-2.0.3-2 oracleasm-2.6.9-42.0.0.0.1.ELsmp-2.0.3-2 oracleasmlib-2.0.2-1 Map raw devices for ASM disks. A raw device mapping is required only if you are planning on creating ASM disks using standard Linux I/O. An alternative to creating ASM disks is to use the ASM library driver provided by Oracle. You will configure ASM disks using ASM library driver later. Perform the following tasks to map the raw devices to the shared partitions created earlier. The raw devices have to bind with the block devices each time a cluster node boots. Add the following lines in /etc/sysconfig/rawdevices. /etc/sysconfig/rawdevices /dev/raw/raw1 /dev/sdc1 /dev/raw/raw2 /dev/sdd1 /dev/raw/raw3 /dev/sde1 To make the mapping effective immediately, execute the following commands as the root user: Assigning devices:
74
/dev/raw/raw1 --> /dev/sdc1 /dev/raw/raw1: bound to major 8, minor 33 /dev/raw/raw2 --> /dev/sdd1 /dev/raw/raw2: bound to major 8, minor 49 /dev/raw/raw3 --> /dev/sde1 /dev/raw/raw3: bound to major 8, minor 65 done # chown oracle:dba /dev/raw/raw[1-3] # chmod 660 /dev/raw/raw[1-3] # ls -lat /dev/raw/raw* crw-rw---- 1 oracle dba 162, 3 Nov 4 07:04 /dev/raw/raw3 crw-rw---- 1 oracle dba 162, 2 Nov 4 07:04 /dev/raw/raw2 crw-rw---- 1 oracle dba 162, 1 Nov 4 07:04 /dev/raw/raw1 As the oracle user, execute rac1-> ln -sf /dev/raw/raw1 /u01/oradata/devdb/asmdisk1 rac1-> ln -sf /dev/raw/raw2 /u01/oradata/devdb/asmdisk2 rac1-> ln -sf /dev/raw/raw3 /u01/oradata/devdb/asmdisk3 Modify /etc/udev/permissions.d/50-udev.permissions. Raw devices are remapped on boot. The ownership of the raw devices will change to the root user by default upon boot. ASM will have problem accessing the shared partitions if the ownership is not the oracle user. Comment the original line, raw/*:root:disk:0660 in /etc/udev/permissions.d/50-udev.permissions and add a new line, raw/*:oracle:dba:0660. /etc/udev/permissions.d/50-udev.permissions # raw devices ram*:root:disk:0660 #raw/*:root:disk:0660 raw/*:oracle:dba:0660 Section D Create and Configure the Second Virtual Machine To create the second virtual machine, simply shut down the first virtual machine, copy all the files in d:\vm\rac\rac1 to d:\vm\rac\rac2 and perform a few configuration changes Modify network configuration. 1. As the root user on rac1, # shutdown h now 2. On your host system, copy all the files in rac1 folder to rac2. D:\>copy d:\vm\rac\rac1 d:\vm\rac\rac2
75
3. On your VMware Server Console, press CTRL-O to open the second virtual machine, d:\rac\rac2\Red Hat Enterprise Linux 4.vmx. 4. VMware Server console: Rename the virtual machine name from rac1 to rac2. Right-click on the new rac1 tab you have just opened and select Settings. Select the Options tab. 1. Virtual machine name: Enter rac2.
Click on Start this virtual machine to start rac2, leaving rac1 powered off. rac2 Virtaul Machine: Select Create a new identifier.
5. Log in as the root user and execute system-config-network to modify the network configuration. IP Address: Double-click on each of the Ethernet devices and use the table below to make the necessary changes.
76
Device eth0 eth1
IP Address 192.168.2.132 10.10.10.32
Subnet mask 255.255.255.0 255.255.255.0
Default gateway address 192.168.2.1 <leave empty>
MAC Address: Navigate to the Hardware Device tab and probe for a new MAC address for each of the Ethernet device. Hostname and DNS: Use the table below to make the necessary changes to the entries in the DNS tab and press CTRL-S to save
Hostname
Primary DNS
Secondary DNS
DNS search path Accepts the default or leave it empty.
rac2.mycorpdomain.com Enter your Enter your DNS DNS IP IP address or address or leave it empty. leave it empty.
Finally, activate each of the Ethernet device. Modify /etc/hosts. Add the following entry in /etc/hosts. 127.0.0.1 localhost
VIPCA will attempt to use the loopback address later during the Oracle clusterware software installation. Modify /export/home/oracle/.profile. Replace the value of ORACLE_SID with devdb2. Establish user equivalence with SSH. During the Cluster Ready Services (CRS) and RAC installation, the Oracle Universal Installer (OUI) has to be able to copy the software as oracle to all RAC nodes without being prompted for a password. In Oracle 10g, this can be accomplished using ssh instead of rsh. To establish user equivalence, generate the user's public and private keys as the oracle user on both nodes. Power on rac1 and perform the following tasks
77
on both nodes. On rac1, rac1-> mkdir ~/.ssh rac1-> chmod 700 ~/.ssh rac1-> ssh-keygen -t rsa Generating public/private rsa key pair. Enter file in which to save the key (/export/home/oracle/.ssh/id_rsa): Enter passphrase (empty for no passphrase): Enter same passphrase again: Your identification has been saved in /export/home/oracle/.ssh/id_rsa. Your public key has been saved in /export/home/oracle/.ssh/id_rsa.pub. The key fingerprint is: 87:54:4f:92:ba:ed:7b:51:5d:1d:59:5b:f9:44:da:b6 oracle@rac1.mycorpdomain.com rac1-> ssh-keygen -t dsa Generating public/private dsa key pair. Enter file in which to save the key (/export/home/oracle/.ssh/id_dsa): Enter passphrase (empty for no passphrase): Enter same passphrase again: Your identification has been saved in /export/home/oracle/.ssh/id_dsa. Your public key has been saved in /export/home/oracle/.ssh/id_dsa.pub. The key fingerprint is: 31:76:96:e6:fc:b7:25:04:fd:70:42:04:1f:fc:9a:26 oracle@rac1.mycorpdomain.com
On rac2, rac2-> mkdir ~/.ssh rac2-> chmod 700 ~/.ssh rac2-> ssh-keygen -t rsa Generating public/private rsa key pair. Enter file in which to save the key (/export/home/oracle/.ssh/id_rsa): Enter passphrase (empty for no passphrase): Enter same passphrase again: Your identification has been saved in /export/home/oracle/.ssh/id_rsa. Your public key has been saved in /export/home/oracle/.ssh/id_rsa.pub. The key fingerprint is: 29:5a:35:ac:0a:03:2c:38:22:3c:95:5d:68:aa:56:66 oracle@rac2.mycorpdomain.com rac2-> ssh-keygen -t dsa Generating public/private dsa key pair. Enter file in which to save the key (/export/home/oracle/.ssh/id_dsa): Enter passphrase (empty for no passphrase):
78
Enter same passphrase again: Your identification has been saved in /export/home/oracle/.ssh/id_dsa. Your public key has been saved in /export/home/oracle/.ssh/id_dsa.pub. The key fingerprint is: 4c:b2:5a:8d:56:0f:dc:7b:bc:e0:cd:3b:8e:b9:5c:7c oracle@rac2.mycorpdomain.com On rac1, rac1-> cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys rac1-> cat ~/.ssh/id_dsa.pub >> ~/.ssh/authorized_keys rac1-> ssh rac2 cat ~/.ssh/id_rsa.pub >> ~/.ssh/authorized_keys The authenticity of host 'rac2 (192.168.2.132)' can't be established. RSA key fingerprint is 63:d3:52:d4:4d:e2:cb:ac:8d:4a:66:9f:f1:ab:28:1f. Are you sure you want to continue connecting (yes/no)? yes Warning: Permanently added 'rac2,192.168.2.132' (RSA) to the list of known hosts. oracle@rac2's password: rac1-> ssh rac2 cat ~/.ssh/id_dsa.pub >> ~/.ssh/authorized_keys oracle@rac2's password: rac1-> scp ~/.ssh/authorized_keys rac2:~/.ssh/authorized_keys oracle@rac2's password: authorized_keys 100% 1716 1.7KB/s 00:00 Test the connection on each node. Verify that you are not prompted for password when you run the following the second time. ssh rac1 date ssh rac2 date ssh rac1-priv date ssh rac2-priv date ssh rac1.mycorpdomain.com date ssh rac2.mycorpdomain.com date ssh rac1-priv.mycorpdomain.com date ssh rac2-priv.mycorpdomain.com date Section E Configure Oracle Automatic Storage Management (ASM) Oracle ASM is tightly integrated with Oracle Database and works with Oracles suite of data management tools. It simplifies database storage management and provides the performance of raw disk I/O. Configure ASMLib. Configure the ASMLib as the root user on both nodes. # /etc/init.d/oracleasm configure Configuring the Oracle ASM library driver. This will configure the on-boot properties of the Oracle ASM library
79
driver. The following questions will determine whether the driver is loaded on boot and what permissions it will have. The current values will be shown in brackets ('[]'). Hitting without typing an answer will keep that current value. Ctrl-C will abort. Default user to own the driver interface []: oracle Default group to own the driver interface []: dba Start Oracle ASM library driver on boot (y/n) [n]: y Fix permissions of Oracle ASM disks on boot (y/n) [y]: y Writing Oracle ASM library driver configuration: [ OK ] Loading module "oracleasm": [ OK ] Mounting ASMlib driver filesystem: [ OK ] Scanning system for ASM disks: [ OK ] Create ASM disks. Create the ASM disks on any one node as the root user. # /etc/init.d/oracleasm createdisk VOL1 /dev/sdc1 Marking disk "/dev/sdc1" as an ASM disk: [ OK ] # /etc/init.d/oracleasm createdisk VOL2 /dev/sdd1 Marking disk "/dev/sdd1" as an ASM disk: [ OK ] # /etc/init.d/oracleasm createdisk VOL3 /dev/sde1 Marking disk "/dev/sde1" as an ASM disk: [ OK ] Verify that the ASM disks are visible from every node. # /etc/init.d/oracleasm scandisks Scanning system for ASM disks: [ OK ] # /etc/init.d/oracleasm listdisks VOL1 VOL2 VOL3 Section F Configure Oracle Cluster File System (OCFS2) OCFS2 is a general-purpose cluster file system developed by Oracle and integrated with the Enterprise Linux kernel. It enables all nodes to share files concurrently on the cluster file system and thus eliminates the need to manage raw devices. Here you will house the OCR and Voting Disk in the OCFS2 file system. Additional information on OCFS2 can be obtained from OCFS2 Users Guide. You should already have the OCFS2 RPMs installed during the Enterprise Linux installation. Verifythat the RPMs have been installed on both nodes. rac1-> rpm -qa | grep ocfs ocfs2-tools-1.2.2-2 ocfs2console-1.2.2-2 ocfs2-2.6.9-42.0.0.0.1.ELsmp-1.2.3-2
80
Create the OCFS2 configuration file. As the root user on rac1, execute # ocfs2console 1. 2. 3. 4. OCFS2 Console: Select Cluster, Configure Nodes. The cluster stack has been started: Click on Close. Node Configuration: Click on Add. Add Node: Add the following nodes and then click on Apply. Name: rac1 IP Address: 192.168.2.131 IP Port: 7777 Name: rac2 IP Address: 192.168.2.132 IP Port: 7777 5. Verify the generated configuration file. # more /etc/ocfs2/cluster.conf node: ip_port = 7777 ip_address = 192.168.2.131 number = 0 name = rac1 cluster = ocfs2 node: ip_port = 7777 ip_address = 192.168.2.132 number = 1 name = rac2 cluster = ocfs2 cluster: node_count = 2 name = ocfs2 6. Propagate the configuration file to rac2. You can rerun the steps above on rac2 to generate the configuration file or select Cluster, Propagate onfiguration on the OCFS2 Console on rac1 to propagate the configuration file to rac2. Configure the O2CB driver. O2CB is a set of clustering services that manages the communication between the nodes and the cluster file system. Below is a description of the individual services: NM: Node Manager that keep track of all the nodes in the cluster.conf HB: Heartbeat service that issues up/down notifications when nodes join or leave the cluster TCP: Handles communication between the nodes
81
DLM: Distributed lock manager that keeps track of all locks, its owners, and status CONFIGFS: User space driven configuration file system mounted at /config DLMFS: User space interface to the kernel space DLM
Perform the procedure below on both nodes to configure O2CB to start on boot. When prompted for a value for the heartbeat dead threshold, you have to specify a value higher than 7 to prevent the nodes from crashing due to the slow IDE disk drive. The heartbeat dead threshold is a variable used to calculate the fence time. Fence time (seconds) = (heartbeat dead threshold -1) * 2 A fence time of 120 seconds works well in our environment. The value of heartbeat dead threshold should be the same on both nodes. As the root user, execute # /etc/init.d/o2cb unload Stopping O2CB cluster ocfs2: OK Unmounting ocfs2_dlmfs filesystem: OK Unloading module "ocfs2_dlmfs": OK Unmounting configfs filesystem: OK Unloading module "configfs": OK # /etc/init.d/o2cb configure Configuring the O2CB driver. This will configure the on-boot properties of the O2CB driver. The following questions will determine whether the driver is loaded on boot. The current values will be shown in brackets ('[]'). Hitting without typing an answer will keep that current value. Ctrl-C will abort. Load O2CB driver on boot (y/n) [y]: y Cluster to start on boot (Enter "none" to clear) [ocfs2]: Specify heartbeat dead threshold (>=7) [7]: 61 Writing O2CB configuration: OK Loading module "configfs": OK Mounting configfs filesystem at /config: OK Loading module "ocfs2_nodemanager": OK Loading module "ocfs2_dlm": OK Loading module "ocfs2_dlmfs": OK Mounting ocfs2_dlmfs filesystem at /dlm: OK Starting O2CB cluster ocfs2: OK
82
Format the file system. Before proceeding with formatting and mounting the file system, verify that O2CB is online on both nodes; O2CB heartbeat is currently inactive because the file system is not mounted. # /etc/init.d/o2cb status Module "configfs": Loaded Filesystem "configfs": Mounted Module "ocfs2_nodemanager": Loaded Module "ocfs2_dlm": Loaded Module "ocfs2_dlmfs": Loaded Filesystem "ocfs2_dlmfs": Mounted Checking O2CB cluster ocfs2: Online Checking O2CB heartbeat: Not active You are only required to format the file system on one node. As the root user on rac1, execute # ocfs2console 1. OCFS2 Console: Select Tasks, Format. 2. Format: Available devices: /dev/sdb1 Volume label: oracle Cluster size: Auto Number of node slots: 4 Block size: Auto 6. OCFS2 Console: CTRL-Q to quit. Mount the file system. To mount the file system, execute the command below on both nodes. # mount -t ocfs2 -o datavolume,nointr /dev/sdb1 /ocfs To mount the file system on boot, add the following line in /etc/fstab on both nodes. /etc/fstab /dev/sdb1 /ocfs ocfs2 _netdev,datavolume,nointr 0 0 Create Oracle Clusterware directory. Create the directory in OCFS2 file system where the OCR and Voting Disk will reside. On rac1, # mkdir /ocfs/clusterware # chown -R oracle:dba /ocfs You have completed the set up of OCFS2. Verify that you can read and write files on the shared cluster file system from both nodes.
83
Section G Install Oracle Clusterware Go to page 20 and Check All Pre-request for Clusterware installation Go to page 24 and Install Oracle Clusterware Section H Install Oracle database binary Go to page 28 and Install Oracle Database Binary. Section I Configure Oracle Listener Go to page 34 and Configure Oracle Listener Section J Create and Configure ASM Instance and ASM DISK Groups Go to page 36 and create/Configure Oracle ASM Section K Create Database Go to page 42and create Database Section L Setup Test Transport Failover Go to page 53and create Database
84
Convert 10gR2 Stand-alone database to Oracle RAC

There are multiple ways in Oracle 10gR2 to convert a single instance database into a clustered database. We can use the following methods to do this: - Manually - DBCA - Oracle Enterprise Manager - rconfig utility (new with 10gR2 there is utility called rconfig which is located in $ORACLE_HOME/bin.) Here we will describe all methods to do Conversion stand-alone database to RCA. Method: Manually Here's an overview of our single-instance database environment before converting to RAC: Host Name Instance Database Database File Name Name Storage RAC1 PROD1 PROD OS file system And an overview of the RAC database environment: Host Name RAC1 RAC2 Instance Name PROD1 PROD2 Database Name PROD PROD Database File Storage ASM ASM OCR/Voting Disk RAW RAW
Task 1 (Create Shared Storage and Configure) Go to Page 15 Task 2 (Create and Configure the Second Virtual Machine) Go to page 16 Task 3 (Add Ether Net Adapter for Private Network in Both Machine) Go to page 16
85
Task 4 (Prepare Disk for OCR, Voting and ASM Storage) Go to page 17 Task 5 (Install Oracle Clusterware) Go to page 24 Task 6 (Install Oracle 10gR12 Software/binary) Go to page 28 Task 7 (Configure Oracle Listener) Go to page 34 Task 8 (Create and Configure ASM Instance and ASM DISK Groups) Go to page 36 Task 9 (Mount and open database with new Oracle Home) Step 1 copy init parameter (initPROD1.ora) file from <OLD_ORACLE_HOME/dbs> to <NEW_ORACLE_HOME/dbs on node1. Step 1 Set environment path. $ export ORACLE_HOME=<SET_PATH_NEW_ORACLE_HOME> $export PATH=$ORACLE_HOME/bin:$PATH $export ORACLE_SID=PROD1 #sqlplus /nolog SQL> conn / as sysdba SQL> startup Task 10 (Migrate Database to RAC) We must use RMAN to migrate the data files to ASM disk groups. All data files will be migrated to the newly created disk group, DATA. The redo logs and control files are created in DATA. Step 1 Migrate Datafile & controle filr to ASM. SQL> connect sys/sys@prod1 as sysdba Connected. SQL> alter system set db_create_file_dest=+DATA; System altered. SQL> alter system set control_files='+DATA/cf1.dbf' scope=spfile; System altered. SQL> shutdown immediate; $ rman target /
86
RMAN> startup nomount; Oracle instance started Total System Global Area 419430400 bytes Fixed Size 779416 bytes Variable Size 128981864 bytes Database Buffers 289406976 bytes Redo Buffers 262144 bytes RMAN> restore controlfile from '/oracle/oradata/prod1/control01.ctl'; RMAN> alter database mount; database mounted released channel: ORA_DISK_1 RMAN> backup as copy database format '+DATA'; RMAN> switch database to copy; datafile 1 switched to datafile copy "+DATA/prod1/datafile/system.257.1" datafile 2 switched to datafile copy "+DATA/prod1/datafile/undotbs1.259.1" datafile 3 switched to datafile copy "+DATA/prod1/datafile/sysaux.258.1" datafile 4 switched to datafile copy "+DATA/prod1/datafile/users.260.1" RMAN> alter database open; database opened RMAN> exit SQL> connect sys/sys@prod1 as sysdba Connected. SQL> select tablespace_name, file_name from dba_data_files; TABLESPACE --------------------USERS SYSAUX UNDOTBS1 SYSTEM FILE_NAME ----------------------------------------+DATA/prod1/datafile/users.260.1 +DATA/prod1/datafile/sysaux.258.1 +DATA/prod1/datafile/undotbs1.259.1 +DATA/prod1/datafile/system.257.1
Step 2 Migrate temp tablespace to ASM. SQL> alter tablespace temp add tempfile size 100M;
87
Tablespace altered. SQL> select file_name from dba_temp_files; FILE_NAME ------------------------------------+DATA/prod1/tempfile/temp.264.3 Step 3 Migrate redo logs to ASM. Drop existing redo logs and recreate them in ASM disk groups, DATA. SQL> alter system set db_create_online_log_dest_1='+DATA'; System altered. SQL> alter system set db_create_online_log_dest_2='+DATA'; System altered. SQL> select group#, member from v$logfile; GROUP# MEMBER -----------------------------------------------1 /oracle/oradata/prod1/redo01.log 2 /oracle/oradata/prod1/redo02.log SQL> alter database add logfile group 3 size 10M; Database altered. SQL> alter system switch logfile; System altered. SQL> alter database drop logfile group 1; Database altered. SQL> alter database add logfile group 1 size 100M; Database altered. SQL> alter database drop logfile group 2;
88
Database altered. SQL> alter database add logfile group 2 size 100M; Database altered. SQL> alter system switch logfile; System altered. SQL> alter database drop logfile group 3; Database altered. SQL> select group#, member from v$logfile; GROUP# MEMBER -----------------------------------------------------1 +DATA/prod1/onlinelog/group_1.265.3 1 +DATA/prod1/onlinelog/group_1.257.1 2 +DATA/prod1/onlinelog/group_2.266.3 2 +DATA/prod1/onlinelog/group_2.258.1 Step 4 Add additional control file. If an additional control file is required for redundancy, you can create it in ASM as you would on any other filesystem. SQL> connect sys/sys@prod1 as sysdba Connected to an idle instance. SQL> startup mount ORACLE instance started. Total System Global Area 419430400 bytes Fixed Size 779416 bytes Variable Size 128981864 bytes Database Buffers 289406976 bytes Redo Buffers 262144 bytes Database mounted. SQL> alter database backup controlfile to '+RECOVERY/cf2.dbf'; Database altered.
89
SQL> alter system set control_files='+DATA/cf1.dbf ','+RECOVERY/cf2.dbf' scope=spfile; System altered. SQL> shutdown immediate; ORA-01109: database not open Database dismounted. ORACLE instance shut down. SQL> startup ORACLE instance started. Total System Global Area 419430400 bytes Fixed Size 779416 bytes Variable Size 128981864 bytes Database Buffers 289406976 bytes Redo Buffers 262144 bytes Database mounted. Database opened. SQL> select name from v$controlfile; NAME --------------------------------------+DATA/cf1.dbf +RECOVERY/cf2.dbf After successfully migrating all the data files over to ASM, the old data files are no longer needed and can be removed. Your single-instance database is now running on ASM! Task 11 (Register the ASM instances with CRS) For higher availability, register the ASM instances under the CRS framework. When registered, the CRS should detect any failed instances and automatically attempt to start up the instances. The CRS should also automatically start up the instances when the servers are rebooted. On node1 (RAC1): $ srvctl add asm -n rac1 -i +ASM1 -o <ORCALE_HOME> $ srvctl start asm -n rac1 $ srvctl status asm -n rac1 ASM instance +ASM1 is running on node rac1.
90
On node 2(RAC2): $ srvctl add asm -n rac2 -i +ASM2 -o <ORCALE_HOME> $ srvctl start asm -n rac2 $ srvctl status asm -n rac2 ASM instance +ASM2 is running on node rac2.
Task 12 (Add RAC specific parameters in pfile on node 1 (RAC1) Modify the initPROD1.ora file on node 1 (RAC1) and copy this file to default location of node 2 (RAC2). Add and modify the following parameters: *.cluster_database_instances=2 *.cluster_database=true *.remote_listener='LISTENERS_PROD prod1.thread=1 prod1.instance_number=1 prod1.undo_tablespace='UNDOTBS1' prod2.thread=2 prod2.instance_number=2 prod2.undo_tablespace='UNDOTBS2' Task 12 (Create RAC Data Dictionary Views) Create the RAC data dictionary views on the first RAC instance. SQL> !echo $ORACLE_SID prod1 SQL> spool /tmp/catclust.log SQL> @$ORACLE_HOME/rdbms/admin/catclust ... ... ... SQL> spool off SQL> shutdown immediate; Task 13 (Register the RAC instances with CRS) On node 1 (RAC1): $ srvctl add database -d prod -o $ORACLE_HOME $ srvctl add instance -d prod -i prod1 -n rac1
91
$ srvctl add instance -d prod -i prod2 -n rac2 $ srvctl start instance -d prod -i prod1 Task 14 (Create redo logs for the second RAC instance) SQL> connect sys/sys@prod1a as sysdba Connected. SQL> alter database add logfile thread 2 group 3 size 100M; SQL> alter database add logfile thread 2 group 4 size 100M; SQL> select group#, member from v$logfile; GROUP# MEMBER --------------- ---------------------------------------1 +DATA/prod/onlinelog/group_1.265.3 1 +DATA/prod/onlinelog/group_1.257.1 2 +DATA/prod/onlinelog/group_2.266.3 2 +DATA/prod/onlinelog/group_2.258.1 3 +DATA/prod/onlinelog/group_3.268.1 3 +DATA/prod/onlinelog/group_3.259.1 4 +DATA/prod/onlinelog/group_4.269.1 4 +DATA/prod/onlinelog/group_4.260.1 8 rows selected. SQL> alter database enable thread 2; Database altered. Task 15 (Create undo tablespace for the second RAC instance) SQL> create undo tablespace UNDOTBS2 datafile size 200M; SQL> select tablespace_name, file_name from dba_data_files where tablespace_name=UNDOTBS2; TABLESPACE --------------------UNDOTBS2 FILE_NAME -------------------------------------+DATA/prod/datafile/undotbs2.270.1 Task 16 (Start up the second RAC instance) $ srvctl start instance -d prod -i prod1b $ crs_stat -t $ srvctl status database -d prod
92
Instance prod1a is running on node rac1 Instance prod1b is running on node rac2 $ srvctl stop database -d prod $ srvctl start database -d prod $ sqlplus system/system@prod1 SQL> select instance_number instance#, instance_name, host_name, status from gv$instance; INSTANCE# INSTANCE_NAME HOST_NAME 1 prod1 rac1 2 prod2 rac2 STATUS OPEN
OPEN
Task 17 (Setup Test Transport Failover) Go to page 53 Congratulations, you have converted your single-instance database to RAC. Method: rconfig New rconfig utility with 10gR2, which is located in $ORACLE_HOME/bin. It uses a file called $ORACLE_HOME/assistants/rconfig/sampleXMLs/ConvertToRAC.xml which you need to edit in order to adjust some parameters as needed and then save it under a new name. This project is structured into following steps: 1. Installing Oracle Cluster Software 2. installing Oracle Database 3. Copy the sample ConvertToRAC.xml file from $ORACLE_HOME/assistants/rconfig/sampleXMLs directory to any temp directory 4. Modify the Sample XML file and verify if it is correct 5. Modify the Sample XML file and execute the sample XML file using rconfig utility Here's an overview of our single-instance database environment before converting to RAC: Host Name Instance Database Database File Name Name Storage RAC1 PROD1 PROD OS file system And an overview of the RAC database environment:
93
Host Name RAC1 RAC2
Instance Name PROD1 PROD2
Database Name PROD PROD
Database File Storage ASM ASM
OCR/Voting Disk RAW RAW
Task 1 (Create Shared Storage and Configure) Go to Page 15 Task 2 (Create and Configure the Second Virtual Machine) Go to page 16 Task 3 (Add Ether Net Adapter for Private Network in Both Machine) Go to page 16 Task 4 (Prepare Disk for OCR, Voting and ASM Storage) Go to page 17 Task 5 (Install Oracle Clusterware) Go to page 24 Task 6 (Install Oracle 10gR12 Software/binary) Go to page 28 Task 8 (Create and Configure ASM Instance and ASM DISK Groups) Go to page 36 Task 10 (Migrate Database to RAC) Step 1 COPY THE SAMPLE XML FILE to TEMP localtion. $cp /export/home/oracle/oracle/10.2.0/db/assistants/rconfig/sampleXMLs/ConvertTo RAC.xml /temp Step2 MODIFY THE copy SAMPLE FILE Modify bellow mention parameter: 1. Convert verify="ONLY" before carrying out the actual conversion. This will perform a test run to validate parameters and flag any incorrect settings or issues that need to be resolved before the final conversion takes place. It is very important and below are few examples that shows that I am missing some information and so it is corrected. Convert verify="YES": rconfig performs checks to ensure that the prerequisites for single-instance to RAC conversion have been met before it starts conversion Convert verify="NO": rconfig does not perform prerequisite checks, and starts conversion
94
Convert verify="ONLY" rconfig only performs prerequisite checks; it does not start conversion after completing prerequisite checks 2. Specify 'SourceDBHome' variable in ConvertToRAC.xml as Non- RAC oracle home (e.g. $OLD_ORACLE_HOME path). As I am using same RAC Oracle HOME for non-RAC instance and so specifying the same RAC $ORACLE_HOME 3. Specify 'TagetDBHome' variable in ConvertToRAC.xml as RAC oracle home (e.g. $ORACLE_HOME path). 4. Specify the Database SID [PROD] 5. Specify the SYS password for Database instance 6. Specify the SYS password for ASM instance 7. Remove any additional Node and specify the Correct Node Name 8. Specify the Instance Prefix you want to use like PROD in our case 9. Specify any change in Network Port. 10. Specify the ASM DiskGroup that will be used 11. execute the rconfig and use the following syntax to run the rconfig command: $ rconfig <path to rconfig xml file > This rconfig run will: Migrate the database to ASM storage (Only if ASM is specified as storage option in the configuration XML file above) Create database instances on all nodes in the cluster Configure listener and NetService entries Configure and register CRS resources Start the instances on all nodes in the cluster Task 17 (Setup Test Transport Failover) Congratulations, you have converted your single-instance database to RAC. Method: DBCA
95
We can use DBCA to convert from single-instance Oracle databases to RAC. DBCA automates the configuration of the control file attributes, creates the undo tablespaces and the redo logs, and makes the initialization parameter file entries for cluster-enabled environments. It also configures Oracle Net Services, Oracle Clusterware resources, and the configuration for RAC database management for use by Oracle Enterprise Manager or the SRVCTL utility. Task 1 (Create Shared Storage and Configure) Go to Page 15 Task 2 (Create and Configure the Second Virtual Machine) Go to page 16 Task 3 (Add Ether Net Adapter for Private Network in Both Machine) Go to page 16 Task 4 (Prepare Disk for OCR, Voting and ASM Storage) Go to page 17 Task 5 (Install Oracle Clusterware) Go to page 24 Task 6 (Install Oracle 10gR12 Software/binary) Go to page 28 Task 8 (Create and Configure ASM Instance and ASM DISK Groups) Go to page 36 Step 1 create Current database template
Go to Standalone database server and start DBCA. At the Welcome page, click next. On the Operations page, select Manage Templates, and click Next. On the Template Management page, select Create a database template and From an existing database (structure as well as data), and click Next. On the Source Database page, select the database name in the Database instance field, and click Next. On the Template Properties page, enter a name for your template in the Name field. By default, the template files are generated in the directory ORACLE_HOME/assistants/dbca/templates and click next. On the Location of Database Related Files page, choose Convert the file location to use OMF structure, and click Finish.
DBCA generates two files: a database structure file (template_name.dbc), database preconfigured image file (template_name.dfb) and database control file (Template_Name.ctrl)
96
Step 2 Copies Template file to Target System. Copy template_name.dbf,template _name.ctrl and template_name.dfb to Target System fefault location. Step 3 Run the DBCA on Oracle RAC node (target System) On the DBCA Template Selection page, use the template that you copied to a temporary location.
Managing OCR and Voting Disk

OCR: Oracle Cluster registry (OCR) store the cluster configuration information and database configuration information like cluster node list, cluster database instance to node mapping and CRS application resource profile. OCR location is specified during CRS installation. OCR.loc file indicate the OCR device location. OCR.loc file located in /etc/oracle on linux system and /var/opt/oracle on Solaris system.
97
We create OCR in shared disk storage that must be accessible to all cluster nodes.
The daemon OCSSd manages the configuration info in OCR and maintains the changes to cluster in the registry.
How to check health of OCR? We just use ocrcheck utility. How to take Backup of OCR? There are two methods of backup. The first method uses automatically and the second method uses manually created logical OCR export files. Automatically: Oracle automatically takes a backup of OCR to default location on every four hours. Oracle always retains the last three backup copies of the OCR. Default location is $CRS_HOME/cdata/cluster_name where cluster_name is the name of your cluster. We can change backup default location of OCR by using ocrconfig command. (Example: $ ocrconfig backuploc <location> The CRSD process that also creates and retains an OCR backup for each full day and at the end of each week. We cannot customize the backup frequencies or the number of files that Oracle retains. Manually: We can take export backup of OCR after making changes by using oraconfig command. (Example: ocrconfig export <location>) How to Recovering the OCR? In event of a failure, before you attempt to restore the OCR, ensure that the OCR is unavailable. Run the following command to check the status of the OCR: ocrcheck If this command does not display the message 'Device/File integrity check succeeded' for at least one copy of the OCR, then both the primary OCR and the OCR mirror have failed. You must restore the OCR from a backup. Restoring the OCR from Automatically Generated OCR Backups: Step 1 Identify the available OCR backups using the ocrconfig command:
98
# ocrconfig -showbackup Step 2 Review the contents of the backup using the following ocrdump command, where file_name is the name of the OCR backup file: $ ocrdump -backupfile file_name Step 3 As the root user, stop Oracle Clusterware on all the nodes in your Oracle RAC cluster by executing the following command: # crsctl stop crs Step 4 Repeat this command on each node in your Oracle RAC cluster. Step 5 As the root user, restore the OCR by applying an OCR backup file. # ocrconfig -restore file_name Step 6 As the root user, restart Oracle Clusterware on all the nodes in your cluster by restarting each node, or by running the following command: # crsctl start crs Repeat this command on each node in your Oracle RAC cluster. Step 7 Use the Cluster Verify Utility (CVU) to verify the OCR integrity. Run the following command, where the -n all argument retrieves a list of all the cluster nodes that are configured as part of your cluster: $ cluvfy comp ocr -n all [-verbose] Recovering the OCR from an OCR Export File: We use the ocrconfig -import command to restore the OCR Step 1 Log in as a Root User. Stop oracle Clusterware on all nodes. Step 2 Restore the OCR data by importing the contents of the OCR export file using the following command, where: ocrconfig -import <file_location> Step 3 Start oracle Clusterware on all nodes. crsctl start crs
99
Step 4 Use the CVU to verify the OCR integrity. cluvfy comp ocr -n all [-verbose] How to Adding an OCR Location You can add an OCR location after an upgrade or after completing the Oracle RAC installation. If you already mirror the OCR, then you do not need to add an OCR location; Oracle Clusterware automatically manages two OCRs when you configure normal redundancy for the OCR. Oracle RAC environments do not support more than two OCRs, a primary OCR and a secondary OCR. Run the following command to add an OCR location: ocrconfig -replace ocr <disk location> Run the following command to add an OCR mirror location: ocrconfig -replace ocrmirror <disk_location> How to Replacing an OCR If you need to change the location of an existing OCR, or change the location of a failed OCR to the location of a working one, you can use the following procedure as long as one OCR file remains online. Step 1 Use the OCRCHECK utility to verify that a copy of the OCR other than the one you are going to replace is online using the following command: ocrcheck Step 2 Verify that Oracle Clusterware is running on the node on which the you are going to perform the replace operation using the following command: crsctl check crs Step 3 Run the following command to replace the OCR: ocrconfig -replace ocr <destination_location> Run the following command to replace an OCR mirror location: ocrconfig -replace ocrmirror destination_file How to Repairing an Oracle Cluster Registry Configuration on a Local Node
100
You may need to repair an OCR configuration on a particular node if your OCR configuration changes while that node is stopped. For example, you may need to repair the OCR on a node that was shut down while you were adding, replacing, or removing an OCR. To repair an OCR configuration, run the following command on the node on which you have stopped the Oracle Clusterware daemon: ocrconfig repair ocrmirror device_name How to Removing an Oracle Cluster Registry To remove an OCR location, at least one OCR must be online. You can remove an OCR location to reduce OCR-related overhead or to stop mirroring your OCR because you moved your the OCR to a redundant storage system, such as a redundant array of independent disks (RAID). To remove an OCR location from your Oracle RAC environment: Step 1 Use the OCRCHECK utility to ensure that at least one OCR other than the OCR that you are removing is online. ocrcheck Step 2 Run the following command on any node in the cluster to remove one copy of the OCR: ocrconfig -replace ocr Step 3 This command updates the OCR configuration on all the nodes on which Oracle Clusterware is running. Voting Disks The voting disk records node membership information. A node must be able to access more than half of the voting disks at any time. Backing up Voting Disks The node membership information does not usually change; you do not need to back up the voting disk every day. However, back up the voting disks at the following times: After installation After adding nodes to or deleting nodes from the cluster After performing voting disk add or delete operations
101
How to take backup? dd if=/dev/rdsk/c0d1s1 of=/tmp/voting.dmp When you use the dd command for making backups of the voting disk, the backup can be performed while the Cluster Ready Services (CRS) process is active; you do not need to stop the crsd.bin process before taking a backup of the voting disk. Recovering Voting Disks If a voting disk is damaged, and no longer usable by Oracle Clusterware, you can recover the voting disk if you have a backup file. Run the following command to recover a voting disk where backup_file_name is the name of the voting disk backup file and voting_disk_name is the name of the active voting disk: dd if=backup_file_name of=voting_disk_name Adding and Removing Voting Disks You can dynamically add and remove voting disks after installing Oracle RAC. Do this using the following commands where path is the fully qualified path for the additional voting disk. Run the following command as the root user to add a voting disk: crsctl add css votedisk path Run the following command as the root user to remove a voting disk: crsctl delete css votedisk path crsctl query css votedisk
102
Administering Cluster Ready Services (CRS)

We use Cluster Control Utility CRSCTL to perform administrative operation of oracle clusterware. It is located in $CRS_HOME/bin and must be executed by the root user. 1. To check the current state of all oracle clusterware daemon: $ ./crsctl check crs CSS appears healthy CRS appears healthy EVM appears healthy 2. You can also check the state of individual oracle clusterware daemon: $ ./crsctl check cssd CSS appears healthy $ ./crsctl check crsd CRS appears healthy
103
$ ./crsctl check evmd EVM appears healthy 3. To start oracle clusterware $ ./crsctl start crs Attempting to start CRS stack The CRS stack will be started shortly 4. To stop oracle clusterware $ ./crsctl stop crs Stopping resources. Successfully stopped CRS resources Stopping CSSD. Shutting down CSS daemon. Shutdown request successfully issued. 5. To disable oracle clusterware: # ./crsctl disable crs 6. To enable oracle clusterware: $ ./crsctl enable crs 7. To list the module for debugging in CSS $ ./crsctl lsmodules css The following are the CSS modules :: CSSD COMMCRS COMMNS 8. CRS_STAT: It reports the current state of resources configured in the OCR. $ ./crs_stat -t Name Type Target State Host ora.C1.inst application ONLINE ONLINE rac1 ora.C2.inst application ONLINE ONLINE rac2 ora.AC1.srv application ONLINE ONLINE rac1 ora.RAC.abc.cs application ONLINE ONLINE rac1 ora.RAC.db application ONLINE ONLINE rac2
104
ora.AC1.srv ora.ice2.cs ora.AC1.srv
application application application
ONLINE ONLINE ONLINE
ONLINE ONLINE ONLINE
rac1 rac1 rac1
9. CRS_STOP: This command used to stop resource or cluster member. $ ./crs_stop ora.rac1.ons Attempting to stop òra.rac1.ons` on member `rac1` Stop of òra.rac1.ons` on member `rac1` succeeded. 10. CRS_START: This command used to start resource or cluster member. $ ./crs_start ora.rac1.ons Attempting to start òra.rac1.ons` on member `rac1` Start of òra.rac1.ons` on member `rac1` succeeded. 11. OCRCHECK : It verifies the integrity of the OCR. $ ./ocrcheck Status of Oracle Cluster Registry is as follows : Version : 2 Total space (kbytes) : 5237072 Used space (kbytes) : 9360 Available space (kbytes) : 5227712 ID : 794527192 Device/File Name : /apps/oracle/oradata/ocr Device/File integrity check succeeded Cluster registry integrity check succeeded 12 Check that all nodes have joined the cluster. $./olsnodes 13 Oracle interface Configuration (oifcfg) $./oifcfg getif Oracle Interface Configure Utility This command should cluster_interconnect. return values for global public and global
If the command does not return a value for global cluster_interconnect, enter the following command to delete and set the desired interface: $ ./oifcfg delif global
105
$./oifcfg setif global <interface name >/<subnet>:public $./oifcfg setif global <interface name >/<subnet>:cluster_interconnect\ Cluster Name Check Utility This utility orints the cluster name information: $./cemutlo n -w
Administrating Services
The following tools are available for administrating services. DBCA OEM DBMS_SERVICE Server Control Utility Here we will discuss only about Server Control Utility (SRVCTL). We can use SRVCTL to Add, start, stop, and Enable, Disable and remove instances and services. Command Syntax: Srvctl add The SRVCTL add command add configuration in the OCR. Add Database: - srvctl add database -d <DB NAME> -o <ORACLE HOME> Add Instance: - srvctl add instance -d <DB NAME> -i <Instance Name> -n <Node Name> Add nodeapps: - srvctl add nodeapps -n <Node Name> -o <Oracle Home> -A <VIP>/255.255.255.0
106
Add asm: - srvctl add asm -n <Node Name> -i <ASM Instance> -o <ORACLE Home> srvctl config The SRVCTL config command displays the configuration stored in the OCR. Config database: - srvctl config database -d <Db Name> config nodeapps:- srvctl config nodeapps -n <node name> config asm:- srvctl config asm -n <node name> config listener:- srvctl config listener -n <node name> srvctl start Start Start Start Start Start database:- srvctl start database -d <Db name> -o open instance:- srvctl start instance -d <db name> -i <instance name> nodeapps:- srvctl start nodeapps -n <node name> asm:- srvctl start asm -n <node name> -i <ASM instance> listener:- srvctl start listener -n <node name>
srvctl stop Stop database: - srvctl stop database -d <db name> Stop instance: - srvctl stop instance -d <dba name> -i <instance name> Stop nodeapps: - srvctl stop nodeapps -n <node name> Stop asm: - srvctl stop asm -n <node name> -i <ASM instance> Stop listener:- srvctl stop listener -n <node name> srvctl status srvctl status srvctl status srvctl status srvctl status srvctl status srvctl status srvctl status srvctl status database database -d <db name> -v instance instance -d <db name> -i <instance name> -v nodeapps nodeapps -n <node name> asm asm -n <node name>
srvctl remove srvctl remove database srvctl remove database -d <db name> srvctl remove instance srvctl remove instance -d <db name> -i <instance name> srvctl remove nodeapps srvctl remove nodeapps -n <node name> srvctl remove asm
107
srvctl remove asm -n <node name> -i <ASM Instance> srvctl remove listener -n node1 -l lsnr01
AlterNet method to remove listener by using crs_unregister 1. $ crs_stat | grep NAME\= | grep lsnr NAME=ora.rac1.LISTENER_RAC1.lsnr NAME=ora.rac2.LISTENER_RAC2.lsnr then $ crs_unregister ora.rac1.LISTENER_RAC1.lsnr $ crs_unregister ora.rac2.LISTENER_RAC2.lsnr
srvctl enable srvctl enable database srvctl enable database -d <db name> srvctl enable instance srvctl enable instance -d <db name> -i <instance name> srvctl enable asm srvctl enable asm -n <node name> -i <ASM Instance> srvctl disable srvctl disable database srvctl disable database -d <db name> srvctl disable instance srvctl disable instance -d <db name> -i <instance name> srvctl disable asm srvctl disable asm -n <node name> -i <asm instance>
108
Managing UNDO, Temporary and Redo logs in RAC Environment

Managing UNDO in RAC Environment In the oracle RAC Environment, each instance store transaction undo data in its dedicated undo Tablespace. We can set the undo Tablespace for each instance by setting the undo_tablespace parameter and undo_management to be the same across all the instances. Example: <Node 1 instance name>.undo_tablespace=undo_tbs1 <Node 2 instance name>.undo_tablesapce=undo_tbs2 Managing Temporary Tablespace In an RAC environment, a user will always use the same assigned temporary Tablespace irrespective of the instance being used. Each instance creates a temporary segment in the temporary Tablespace it is using. If an instance is running a big sort operation requires a large temporary tablesapce , it can reclaim the space used by other instances temporary segments in that tablesapce. Main Point: All instance share the same temporary Tablespace Size should be at least equal to the concurrent maximum requirement of the entire instance. Administrating Online redologs
109
Each instance has exclusive write access to its own online redolog files. An instance can read another instance current on line redologs file to perform instance recovery it that instance has terminated abnormally. Online redologs file needs to be located on a shared storage device and can not be on a local node. How to Enable Archiving in the RAC Environment Step 1 Log in Node1. Step 2 set cluster_database=false in parameter file. Step 3 shut down all the instances. Srvctl stop database d <db name> Step 4 mount the database SQL> startup mount Step 5 Enable Archiving SQL> alter database archivelog; Step 6 Change cluser_database=true in parameter file Step 7 Shutdown the instance SQL> shutdown immediate Step 8 Start all the instance. Scrvctl start database d <db_name> How to Enable Archiving in the RAC Environment Step 1 log in node 1 Step 2 Verify that the database is running in Archive log mode. Step 3 Set parameter cluster_database =false SQL> alter system set cluster_database=falce scope=spfile sid=prod1 Step 4 set parameter DB_RECOVERY_FILE_DEST_SIZE and DB_RECOVERYU_FILE_DEST SQL> alter system set DB_RECOVERY_FILE_DEST_SIZE=200M scope=spfile; SQL> alter system set DB_RECOVERY_FILE_DEST=/dev/rdsk/c0d3s1 scope=spfile;
110
Step 5 Shut down all instance. # srvctl stop database d <db_name> Step 6 Mount the database SQL> statup mount Step 7 Enable the flashback. SQL> alter database flashback on; Step 8 Set parameter cluster_database = true SQL> alter system set cluster_database=falce scope=spfile sid=prod1 Step 9 Shutdown instance SQL> shutdown Step 10 Start all instance $ srvctl start database d >db_name>
111
De-Installing Oracle Real Application Clusters Software and Database

Step 1 Examine the Oratab file (/var/opt/oracle/oratab in Solaris) to identify the instance dependencies on this oracle home. Step 2 Start DBCA, selects Oracle Real Application Clusters Database, select Delete a database, and selects the database that you want to delete. Repeat this step to delete all databases. Steps 3 connect with ASM instance, run the following command: a) Check disk group which are using by ASM instance: SQL> select * from V$ASM_DISKGROUP; b) Drop the disk group. SQL> drop <diskgroup diskgroup_name> including contents; c) Shut down ASM on all RAC nodes, and verify that all ASM instances are stopped. e) To remove the ASM entry from the OCR, run the following command for all nodes on which this Oracle home exists: srvctl remove asm -n nodename Step 4 Remove oracle home (On All Node) Go to $ORACLE_HOME/oui/bin and execute ./runInstaller. Start OUI, and remove any existing Oracle Database 10g with RAC software by selecting Deinstall Products and selecting the Oracle home that you want to remove. Step 5 Remove oratab entries for the deleted Oracle home databases. Step 6 De-Installing Oracle Clusterware
112
a) Go to $CRS_HOME/install directory and execute rootdelete.sh script (On All node) # /export/home/oracle/product/10.2.0/crs/install/rootdelete.sh This will disable the Oracle Clusterware applications that are running on the cluster node. b) Run the script CRS_home/install/rootdeinstall.sh on a local node to remove the OCR. c) Go to $CRS_HOME/oui/bin and execute. /runInstaller. Start OUI. In the Welcome page, click Deinstall Products to display the list of installed products. Select the Oracle Clusterware home to de-install (Execute on All Node)
De-Installation RAC Component after Fail Installation

Step 1 Go to $CRS_HOME/install directory and execute rootdelete.sh script (On All node) # /export/home/oracle/product/10.2.0/crs/install/rootdelete.sh This will disable the Oracle Clusterware applications that are running on the cluster node. Step 2 Go to $CRS_HOME/install directory and execute rootdeinstall.sh on a local node to remove the OCR. (Only One Node) Step 3 Start OUI and Select the Oracle Clusterware home to de-install.
113
Adding a Node To 10gR2 RAC cluster

Step 1 Preparing Access to the New Node Adding the public and private node names for the new node to the /etc/hosts file on the existing nodes, RCA1, RAC2 and also new Node RAC3 Verifying the new node can be accessed (using the ping command) from the existing nodes Configure SSH on new node. Running the following command on either docrac1 or docrac2 to verify the new node has been properly configured: cluvfy stage -pre crsinst -n docrac3 Step 2 Adding an Oracle Clusterware Home to a New Node (RAC3) Using OUI Log in as a oracle use on node 1 (RAC1) Go to $ORACLE_CRS_HOME/oui/bin and execute ./addNode.sh and click next. The Oracle Universal Installer (OUI) displays the Node Selection Page Enter the node that you want to add and verify the entries that OUI displays on the Summary Page click Next. Monitor the progress of the copy crs home to new node and verify the total size of the CRS directory Verify crs is started on new node and nodeapps are started except for listener, and then exit OUI. Obtain the remote port identifier, which you need to know for the next step, by running the following command on the existing node from the $ORACLE_CRS_HOME/opmn/conf directory: $ cat $ORA_CRS_HOME/opmn/conf/ons.config localport=6113 remoteport=6201 loglevel=3 useocr=on From the $ORACLE_CRS_HOME/bin directory on an existing node(RAC1), run the Oracle Notification Service (RACGONS) utility as in the following example where remote_port is the port number from the previous step and node3 is the name of the node that you are adding:
114
. /racgons add_config New_Node:Remote_Port Example: $. /racgons add_config rac3:6201 Step 3 Adding an Oracle Home to a New Node Using OUI 1. Log in as a oracle use on node 1 (RAC1) 2. Go to ORACLE_HOME/oui/bin and execute ./addNode.sh and click next. 3. When OUI displays the Node Selection Page, select the node to be added and click Next. 4. Verify the entries that OUI displays on the Summary Page and click Next. Step 4 reconfigure listener on new node Log in as a oracle use on node 1 (RAC1) Start the Oracle Net Configuration Assistant by entering netca at the system prompt from the $ORACLE_HOME/bin directory. Select Listener configuration, and click Next. Select Add to create a new Listener, then click Next. Accept the default value of LISTENER for the Listener name by clicking Next. Choose TCP and move it to the Selected Protocols area, then click Next. Choose Use the standard port number of 1521, then clicks next. Select Cluster configuration for the type of configuration to perform, then click Next. Select the name of the node you are adding, for example RAC3, then click Next. NETCA creates a Listener using the configuration information provided. You can now exit NETCA. Step 5 add ASM instance on new node 1. Add ASM entry in /var/opt/oracle/oratab (on Solaris) and /etc/oratab (on Linux) on new node +ASM3:/export/home/oracle/oracle/product/10.2.0/db_1:N 2. Rename init+ASM1 to init+ASM3 3. Create admin directories for ASM instance 4. Add new ASM instance on all RAC nodes +ASM3.instance_number=3
115
6. Add ASM to cluster

srvctl add asm -n rac3 /export/home/oracle/oracle/product/10.2.0/db_1 6. Start asm instance srvctl start asm -n rac3 Step 6 Creating a Database Instance Log in as a Oracle user on New Node (RAC3) Start DBCA by entering dbca at the system prompt from the $ORACLE_HOME/bin directory. Select Oracle Real Application Clusters database, and then click Next. Select Instance Management, and then click Next. Select Add an Instance, then click Next. In the List of Cluster Databases window, select the active Oracle RAC database to which you want to add an instance, for example PROD. Enter the user name and password for the database user that has SYSDBA privileges. Click Next. You will see a list of the existing instance. Click Next, and on the following screen enter PROD3 as the instance name and choose RAC3 as the node name. This will create a database instance PROD3 (on rac3) , click next in the Database storage screen. During creation, you will be asked whether the ASm instance should be extended to RAC3, choose yes. You should now have a new cluster database instance and ASM instance running on the new node. After you terminate your DBCA session, you should run the following command to verify the administrative privileges on the new node and obtain detailed information about these privileges: CRS_home/bin/cluvfy comp admprv -o db_config -d oracle_home -n rac3 -verbose -i +ASM3 o
116
Removing a 10gR2 RAC cluster Node

Step 1 Deleting the Database Instance Log in as a Oracle user on New Node (RAC3) Start DBCA by entering dbca at the system prompt from the $ORACLE_HOME/bin directory. Select Oracle Real Application Clusters database, and then click Next. Select Instance Management, and then click Next. Select Delete an Instance, then click Next. Select the database name, then choose the instance to delete and confirm the deletion. Step 2 Clean up the ASM Log on node 1 (RAC1) stop ASM services for node 3 (RAC3) $srvctl stop asm n rac3 remove the ASM service $srvctl remove asm n rac4 Remove ASM Directory Structure of node to be deleted Edit /var/opt/oracle/oratab (on Solaris) /etc/oratab (on Linux) and remove ASM instance references. Step 4 Remove the Listener from Node to be deleted Log in as a Oracle user on New Node (RAC3) Start NETCA by entering dbca at the system prompt from the $ORACLE_HOME/bin directory. Choose Cluster Management Choose Listener Choose Remove Confirm delectation of LISTENER Step 5 removing the Node from the Database 1. Log in as a Oracle user on New Node (RAC3) Go to $ORACLE_HOME/oui/bin and Execute :
117
./runInstaller updateNodeList ORACLE_HOME=$ORACLE_HOME CLUSTER_NODES=(rac3) local $./runInstaller Choose th deinstall product and select the dbhome. This will remove the database home software and leave behind only some files and directories. Log in as a oracle user on NODE1 (RAC1) Go to to $ORACLE_HOME/oui/bin and Execute: ./runInstaller updateNodeList ORACLE_HOME=$ORACLE_HOME CLUSTER_NODES=(rac1,rac2) Step 6 Removing RAC3 from the Clusterware Log in on RAC1 node Obtain the remote port identifier:, by running the following command on the existing node from the $ORACLE_CRS_HOME/opmn/conf directory: $ cat $ORA_CRS_HOME/opmn/conf/ons.config localport=6113 remoteport=6201 loglevel=3 useocr=on Run the following command $ORACLE_CRS_HOME/bin/racgone remove_config rac3:6201 log in as a root user on RAC3 and execute: $ORACLE_CRS_HOME/install $./rootdelete.sh run following from RAC1 node: $ORACLE_CRS_HOME/bin/olsnodes n run the following from RAC3 Node: $cd $ORACLE_CRS_HOME/oui/bin $./runInstaller updateNodeList ORACLE_HOME=$ORACLE_CRS_HOME CLUSTER_NODES=(rac3) CRS=TRUE local $./runInstaller Choose Deinstall Software and remove the CRS_HOME Run the following from Node 1 (RAC1) $cd $ORACLE_CRS_HOME/oui/bin $ ./runInstaller updateNodeList ORACLE_HOME=$ORACLE_CRS_HOME CLUSTER_NODES={rac1,rac2} CRS=TRUE
118
RAC Load Balancing

In my opinion, there are 2 types of load balancing in a 10g RAC. Client Side Load Balancing Server Side Load Balancing
Client Side Load Balancing In Client Side Load Balancing Method, when a user attempt to connect to the database , the connection distribute across several listeners and oracle database randomly select an address in the address list and connects to that node listeners. Client Side load Balancing is configured by adding LOAD_BALANCE=ON in tnsnames.ora file. How to Configure? Suppose, you have a 2-node 10g RAC cluster, and for load balancing, you have the following entry in my tnsnames.ora file. PROD= (description= (load_balance=on) (address=(protocol=tcp)(host=rac1-vip)(port=1521)) (address=(protocol=tcp)(host=rac2-vip)(port=1521)) (connect_data= (service_name=PROD) ) ) Testing Client Side Load Balancing Behavior It is quite simple to setup Oracle Net tracing on the client, to test and show whether Client Side Load Balancing is working properly. This type of load balancing has nothing to do with balancing server side load. It is to do with balancing load across the listeners. If you want to trace the Client Side Load Balancing behavior from the client side, so You enable Oracle net tracing by adding the following lines in sqlnet.ora and you should fine trace files for each physical connection made under $ORACLE_HOME/network/trace. TRACE_LEVEL_CLIENT = USER TRACE_FILE_CLIENT = SQLTRC
119
Step 1 Log on any Client Machine (here we user UNIX flavor Client) Step 2 open $ORACLE_HOME/network/admin/sqlnet.ora file and add following line. TRACE_LEVEL_CLIENT = USER TRACE_FILE_CLIENT = SQLTRC Step 3 Executes bellow mention bash shell scripts. for a in {1..1000} do echo $a sqlplus -s system/manager@PROD<<EOF EOF done exit 0 Step 4 Examining the traces file We can use a combination of grep, and wc to see the result of load balancing. Each connection produced 1 trace files, so we have 1000 trace files for 1000 connections. $ ls -l *.trc |wc -l 1000 $ grep nsc2addr *.trc | grep load_balance |grep rac1-vip |wc -l 498 $ grep nsc2addr *.trc | grep load_balance |grep rac2-vip |wc -l 502 We can see over 1000 connections, 498 were made to rac1-vip listener, whilst 502 were made to rac2-vip listener. It is a fairly even distribution. Server Side Load Balancing In Server Side Load Balancing Method, when users make a connection request, the listener directs connection request to the best instance providing by each database instance PMON process. To implement server-side load balancing, listeners must be configured on all node and the REMOTE_LISTENERS initialization parameter must be added to the databases PFILE or SPFILE so that the database knows to search out the value provided in that parameter in the database servers TNSNAMES.ORA
120
If you want server side load balancing, REMOTE_LISTENER should point to all listeners on all nodes, otherwise don't set REMOTE_LISTENER. The purpose of REMOTE_LISTENER is to connect all instances with all listeners so the instances can propagate their load balance advisories to all listeners. if you connect to a listener, this listener uses the advisories to decide who should service your connect. if the listener decides its local instance(s) are least loaded and should service your connect it passes your connect to the local instance. if the node you connected to is overloaded, the listener can use TNS redirect to redirect your connect a less loaded instance. How to Setup Server Side Load Balancing Requirements: New entries in every client's TNSNAMES.ORA file for the new alias New entries in the TNSNAMES.ORA file of every node in the cluster to include the REMOTE_LISTENER setting The addition of *.REMOTE_LISTENER parameter to all nodes in cluster to force each node's Listener to register with each other
Step1 Add bellow entries to each server's TNSNAMES.ORA file to enable Server-Side Load Balancing: LISTENERS_PROD = (ADDRESS_LIST = (ADDRESS = (PROTOCOL = TCP)(HOST = rac1-vip)(PORT = 1521)) (ADDRESS = (PROTOCOL = TCP)(HOST = rac2-vip)(PORT = 1521)) ) Step 2 Run this command to add the REMOTE_LISTENERS initialization parameter to the common SPFILE for all nodes in the RAC clustered database: ALTER SYSTEM SET REMOTE_LISTENER = LISTENERS_PROD SID='*' SCOPE=BOTH;
121
RAC Failover Case Study

Case: Test the failover instance (PROD1) of a connected oracle user in the event of an insert. Action: 1. Connect to any oracle instance. C:> sqlplus /nolog SQL> Conn scott/tiger@PROD 2. Check connected instance SQL> select instance_name from v$instance; INSTANCE_NAME ---------------PROD1 3. Create Table TEST. Sql> create table test (Col1 number); 4. insert over 100,000 line of data into the table sql>declare x number; begin for x in 1..100000 loop insert into test values (x); end loop; end;
5. Telnet into the node that the SCOTT user connected.

6. Terminate the PROD1 instance. C:> telnet 10.1.1.170 bash-3.00$ ps -ef | grep smon oraprod 25829 16727 0 16:25:18 pts/1 0:00 grep smon oraprod 1645 1 0 15:53:10 ? 0:01 ora_smon_PROD1 oraprod 6631 1 0 20:39:52 ? 0:04 asm_smon_+ASM1 kill PROD1 SMON Proceee ( $ kill -9 1645) 7. Check user session should be hang
122
8. now start instance (PROD1) $ /srvctl start instance -d prod -i prod1
9. When instance UP. User terminal show a Error: ERROR at line 1: ORA25402: transaction must roll back 10. rollback the transaction and continue work. Conclusion: The insert transaction should be affected and should not failover to the next node. Case: Test the failover instance (PROD1) of a connected oracle user in the event of Delete and Update. The Delete/Update transaction should be affected and should not failover to the next node. Case: Test the failover instance (PROD1) of a connected oracle user in the event of Select. The Select Statement should not be affected and failover to the next node without any disruption. Case: To test the failover of the private interconnect Action:
1. Pull out one of the interconnect wire and do a ping to the other node via
the private address. 2. The ping should be successful with return messages Case: To test the failover of the network card of the public IP address Action:
1. Plug out one of the cable connecting the node to the public network.
2. Ping from the observation client to the node 3. The ping from the client should be able to go thru
Oracle RAC Log Directory 123
Each component in the CRS stack has its respective directories created under the CRS home: The Cluster Ready Services Daemon (crsd) Log Files $CRS home/log/hostname/crsd Oracle Cluster Registry (OCR) Log Files $CRS Home/log/hostname/client Cluster Synchronization Services (CSS) Log Files $CRS Home/log/hostname/cssd Event Manager (EVM) Log Files $CRS Home/log/hostname/evmd RACG Log Files $ORACLE_HOME/log/hostname/racg
124

Book Oracle RAC

Uploaded by

Copyright:

Available Formats

Book Oracle RAC

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Book Oracle RAC

Uploaded by

Copyright:

Available Formats

Build Oracle 10g RAC Database On Solaris / Linux With VMWare

About the Book

To my parents and my beloved country

4. Convert 10gR2 Stand-alone database to Oracle RAC on Solaris

Method: rconfig 95 Method: DBCA 98

5. Managing OCR and Voting Disk

Storage Option for Oracle Real Application Clusters

Oracle 10g RAC Architecture (Graphical View)

Implement 10gR2 Oracle RAC on Solaris 10 with VMware

rac1-priv rac2-priv rac1-vip rac2-vip

Step 7 Specify OCR locations

Step 11Verify status of CRS services. Execute bellow command:

Step 3 Select Oracle home and Installation Path.

Step 5 Select configuration option. Select software only.

Step 3 Choose Configure Automatic Storage Management

Step 4 Choose BOTH NODE.

Step 5 Choose As per screen Shot

Step 6 Clicks on Create New.

Step 7 Create Disk Group as per screen shot.

$export PATH=$ORACLE_HOME/bin:$PATH $dbca Step 2 Follow Screen Shot

Step 3 Follow Screen Shot

Step 4 Follow Screen Shot

Step 5 Follow Screen Shot

Step 6 Follow Screen Shot

Step 7 Follow Screen Shot

Step 8 Follow Screen Shot

Step 9 Follow Screen Shot

Step 10 Follow Screen Shot

Step 11 Follow Screen Shot

Step 12 Follow Screen Shot

Step 13 Follow Screen Shot

Step 14 Follow Screen Shot

Step 15 Follow Screen Shot

Step 16 Follow Screen Shot

Implement 10gR2 Oracle RAC on Linux with VMware

d:\vm\rac\sharedstorage\ocfs2disk.vm dk d:\vm\rac\sharedstorage\asmdisk1.vm dk d:\vm\rac\sharedstorage\asmdisk2.vm dk

/dev/sdb /dev/sdc /dev/sdd

SCSI 1:0 SCSI 1:1 SCSI 1:2

512 307 2 307 2

ASM flash recovery area

1. Execute vmware-toolbox to bring up the VMware Tools Properties

Device eth0 eth1

IP Address 192.168.2.132 10.10.10.32

Subnet mask 255.255.255.0 255.255.255.0

Default gateway address 192.168.2.1 <leave empty>

DNS search path Accepts the default or leave it empty.

Convert 10gR2 Stand-alone database to Oracle RAC

Host Name RAC1 RAC2

Instance Name PROD1 PROD2

Database Name PROD PROD

Database File Storage ASM ASM

OCR/Voting Disk RAW RAW

Managing OCR and Voting Disk

Administering Cluster Ready Services (CRS)

ora.AC1.srv ora.ice2.cs ora.AC1.srv

application application application

ONLINE ONLINE ONLINE

ONLINE ONLINE ONLINE

rac1 rac1 rac1

Managing UNDO, Temporary and Redo logs in RAC Environment