Exam Code: 1Z0-048
Exam Name: Oracle Database 10g R2: Administering RAC
Which optional parameter of the DESCRIPTION clause of a TNS entry can you use to randomize the access to available listeners?
You are working as a DBA in a RAC environment that is available for testing. The RAC environment has all the database files stored in the OCFS file system. You plan to install ASM and move all the database files to ASM to achieve better performance. During the installation, you selected redundancy level as normal for the disk group being added. What is the reason for selecting this redundancy level?
A. You want to use hardware mirroring.
B. You do not want to have any mirroring in the disk group.
C. You have two failure groups defined, so you want to use two-way mirroring.
D. You have three failure groups defined, so you want to use three-way mirroring.
You are running a two-instance database with six redo log groups defined and decide to add a third thread to support a third database instance on the third node of the cluster. Which three commands would you issue to achieve this using command line administration? (Choose three.)
A. ALTER SYSTEM ENABLE THREAD 3;
B. ALTER DATABASE ENABLE THREAD 3;
C. ALTER SYSTEM ADD LOGFILE THREAD 3 GROUP 8;
D. ALTER SYSTEM ADD LOGFILE THREAD 3 GROUP 7;
E. ALTER DATABASE ADD LOGFILE THREAD 3 GROUP 7;
F. ALTER DATABASE ADD LOGFILE THREAD 3 GROUP 8;
Which initialization parameter would you use for each RAC instance to make sure that each listener on each RAC node is aware of the load of all the others?
You have a well-tuned, non-RAC single instance Oracle database with several services, all of which perform perfectly. You plan to migrate the database to RAC in the next two weeks and all file systems would be shared using Oracle Cluster File System. After the migration, you want to perform a test to monitor RAC-specific tuning issues that would not have occurred on a non-RAC database. Which six areas would you monitor to determine whether there are RAC-specific performance problems after migration? (Choose six.)
A. Interconnect latency
B. Instance recovery time
C. Services with applications that use external tables
D. Services with applications that truncate large tables
E. Services with applications using local context variables
F. Services with query-intensive applications relying on full table scans
G. Services with insert-intensive applications that use Oracle sequences for index keys
H. Services with applications that have occasional queries, using index access on modestly sized tables
I. Services with insert-intensive applications that generate their own keys without Oracle sequences
After a power failure, you noticed that one of your cluster database instances did not start automatically, whereas all other instances came up successfully. What would you do to determine the precise cause for the instance startup failure? (Choose three.)
A. Check the resource state by using the CRS_STAT command.
B. Check for OCR corruptions by using the OCRCHECK command.
C. Check the alert.log file that belongs to the instance that did not start.
D. Check the alert.log file that belongs to the Oracle Clusterware software.
E. Check if the Oracle Clusterware is functional by using the CRSCTL command.
F. Reboot the node where the failing instance resides to check if the problem is persistent.
G. Enable extra debugging in the Oracle Clusterware software by using the CRSCTL command.
View the Exhibit, which displays the cluster database performance page after the same workload has been executed three times on your cluster without varying any non-Oracle workloads on the cluster. What are the two ways by which you can tell that the corresponding application was better tuned during the second and third execution? (Choose two.)
A. There is less paging each time.
B. The run queue length is going down each time.
C. It takes less time to execute the workload each time.
D. The number of transactions per second is getting higher each time.
E. The database throughput graphic is below the red dashed line during the last execution.
Which three storage types are suitable as a flash recovery area for an Oracle 10g RAC database? (Choose three.)
A. Raw volume
B. ASM disk group
C. Local file system
D. Cluster file system
E. Shared NFS directory
After the cluster database has been successfully created, several postinstallation tasks should be completed. Identify three of them. (Choose three.)
A. Create the ASM instance.
B. Back up the root.sh script.
C. Initialize the cluster backup procedure.
D. Set up additional required user accounts.
E. Verify the Enterprise Manager/Oracle Cluster Registry configuration by running srvctl config database -d db_name.
Oracle Cluster Registry (OCR) can be recovered on a UNIX platform using which two commands? (Choose two.)
A. ocrconfig -import /u01/logical_ocr/yesterday.ocr where yesterday.ocr is an automatically generated OCR backup
B. ocrconfig -import /u01/logical_ocr/yesterday.ocr_exp where yesterday.ocr_exp is a logical backup created using the ocrconfig -export command
C. cp /app/oracle/product/10.1.0/crs_1/cdata/dbclust01/day.ocr /ocfs/OCR/crs.dbf where /ocfs/OCR/crs.dbf is the cluster registry file
D. ocrconfig -restore /app/oracle/product/10.1.0/crs_1/cdata/dbclust01/day.ocr where day.ocr is an automatically generated OCR backup
Which two initialization parameters must be set to start up an ASM instance in a RAC environment? (Choose two.)
Your company has a combination of high availability and scalability requirements that need to be addressed, and RAC has been discussed as a solution. There are a mix of “off-the-shelf” software products and homegrown applications currently in use in your environment that run on single-instance Oracle databases. You have been asked by the IT director to provide an impact analysis report to help assess the impact on applications when migrating to RAC. Which three are among the areas that should be examined for performance impact and application functionality impact or for both? (Choose three.)
A. applications that use large parallel full table scans
B. applications that truncate tables in a data warehouse process
C. three-tier applications that use application global context variables
D. client/server type applications that use application local context variables
E. applications that access small tables using the same index occasionally
F. applications that use SQL profiles created by the ATO optimizer in the SQL Tuning Advisor
You decide to implement a Data Guard configuration to support a disaster recovery plan for your RAC database. Your primary database has all shared files and the flash recovery area is stored on ASM disk groups. Which three options exist for storing the data guard broker configuration files at the primary location? (Choose three.)
A. can be in any shared ASM disk group
B. can be stored on shared Raw Devices
C. must be stored on Oracle Cluster File System if a cluster file system is used
D. may be stored on any supported cluster file system if a cluster file system is used
E. must be in the same shared ASM disk group as the database data files if the broker configuration files are stored in ASM disk groups
F. must be in the same shared ASM disk group as the Flash Recovery Area if the broker configuration files are stored in ASM disk groups.
Your company has certain standard practices for technology and you are configuring ASM disk groups for your four-node RAC cluster. The company guidelines suggest that there is one disk group for the flash recovery area that is shared by all databases and a separate disk group for the database area of all databases. Which three exceptions exist that may require you to create additional disk groups? (Choose three.)
A. The retention policy for backups are different for different databases.
B. There are large nonpartitioned tables accessed frequently using indexes.
C. The disk farm has different disk models with different characteristics and ages.
D. Some databases operate in ARCHIVELOGMODE and others in NOARCHIVELOGMODE.
E. Some database instances are shut down every night while others are always running.
F. Applications using the databases have different Service-Level Agreements for response time and throughput.
G. There are partitioned tables where the older partitions are in read-only tablespaces and seldom accessed, but the more recent partitions are frequently accessed.
You have attended a user group presentation on RAC and based on what you have learned you feel that your company would benefit from having the RAC option to meet its Oracle database requirements in several ways. Your manager does not know much about RAC and asks you to create a presentation to the team explaining RAC, and whether it should be implemented in your installation. Which three statements explain the benefits and implications of RAC? (Choose three.)
A. RAC alone would not help protect from human errors.
B. RAC would increase availability in the event of data failures.
C. RAC would increase availability in the event of a network failure.
D. RAC alone would increase availability when doing release upgrades.
E. RAC would increase scalability for all application workloads in all cases.
F. RAC would increase availability in the event of node and instance failures.
G. RAC used in conjunction with parallel execution may provide speedup of DSS queries.
H. RAC used in conjunction with parallel execution may provide for speedup of OLTP workload processes.
In a RAC environment, you see a steep increase for enq:HW – contention and gc current grant wait events. What would you do to fix this issue?
A. Increase the cache size of sequences.
B. Reduce the time of transactions by committing more often.
C. Use larger uniform extent sizes for objects.
D. Reduce the number of updates causing locks on tables accessed by multiple instances.
After migrating from a 10g R1 cluster to 10g R2, you decide to multiplex your voting disk. Which are the two ways in which you can accomplish this task? (Choose two.)
A. You should not use symbolic links.
B. You can add additional voting disks online.
C. You must add additional voting disks offline.
D. You may use the VIPCA utility to add new voting disks.
E. You may use the command:
crsctl add css votedisk <new voting disk path>
You are about to build a new Oracle Clusterware-based cluster and you want to make the installation process as smooth as possible. You decide to use the Cluster Verification Utility (CLUVFY) to check whether or not your system administrator has set up the hardware and system software correctly. Which two statements about CLUVFY are true? (Choose two.)
A. You can run CLUVFY only from a privileged operating system account.
B. You can run CLUVFY directly from the Oracle Clusterware distribution media.
C. CLUVFY cannot be used until the Oracle Clusterware is installed successfully.
D. After the Oracle Clusterware installation, you can run CLUVFY from the Oracle Home that holds the CRS binaries.
E. You have to install CLUVFY on all cluster nodes in order to use CLUVFY on other nodes than the one from which you perform the installation.
F. To use CLUVFY during the Oracle Clusterware installation, you have to download CLUVFY from the Oracle Technology Network (OTN) Web site and install it before you can use it.
Several steps must be followed to successfully add a new node to your cluster database. Assuming that a UNIX platform is used, which of the three steps below are correct? (Choose three.)
A. Add instances using OUI.
B. Reconfigure listeners for the new node with netca.
C. Configure the OS and hardware for the new node.
D. Add the node to the cluster using the addnode.sh script run from ORA_CRS_HOME.
E. Add the RAC software to the new node using addnode.sh from ORA_CRS_HOME.
Which three statements regarding backup of the voting disk are true? (Choose three.)
A. On UNIX platforms, the voting disk can be backed up with the dd command.
B. A backup of the voting disk should be taken whenever an RMAN catalog reset is done.
C. A backup of the voting disk should be taken whenever a new node is added to the cluster.
D. A backup of the voting disk should be taken whenever an existing node is removed from the cluster.
E. Whenever a voting disk backup is made, the Clusterware software should be stopped on all nodes in the cluster.