| 1: | What is the name of the background process that writes the online redo log files out to the archive logs when the database is running in ARCHIVELOG mode? A. ARCn B. DBWn C. LGWR D. CKPT
|
| A1: | Answer A is correct. Archiver (or ARCn) is responsible for archiving the redo log files. Answer B is incorrect because Database Writer (DBWn) is responsible for writing the dirty buffers out to the data files. Answer C is incorrect because Log Writer (LGWR) is responsible for taking care of the online redo log files but not the archived redo log files. Answer D is incorrect because checkpoint has nothing to do with archive logs but is key in telling DBWn when to write out the dirty buffers. |
| 2: | What is the name of the single parameter that you set in Oracle 9i to control the frequency of checkpoints in your instance? A. FAST_START_MTTR_TARGET B. LOG_CHECKPOINT_INTERVAL C. LOG_CHECKPOINT_TIMEOUT D. LOG_CHECKPOINT_TARGET
|
| A2: | Answer A is correct. The single parameter that you would set in the initialization file is FAST_START_MTTR_TARGET. Answer B is incorrect because LOG_CHECKPOINT_INTERVAL is part of what makes up FAST_START_MTTR_TARGET and was one of the components that controlled checkpointing frequency in Oracle 8i. Answer C is incorrect because LOG_CHECKPOINT_TIMEOUT is part of what makes up FAST_START_MTTR_TARGET and was one of the components that controlled checkpointing frequency in Oracle 8i. Answer D is incorrect because it is a nonexistent parameter. |
| 3: | How can you reduce the frequency of checkpoints and log switches in your database without impacting the initialization files? A. Decrease the size of the online redo log files. B. Increase the size of the online redo log files. C. Increase the number of LGWR processes running. D. Increase the number of ARCn processes running.
|
| A3: | Answer B is correct. If you want to reduce the frequency of log switches and checkpointing without any impact to the initialization files, you would need to increase the size of the online redo log files. Answer A is incorrect; decreasing the size of the online redo log files would cause them to log switch more often. Answer C is incorrect because there is no way to increase the number of log writer processes that are running on an instance. Answer D is incorrect because archiver processes have no effect on the frequency of log switches or checkpointing. |
| 4: | Oracle recommends that you multiplex your redo log members in each online redo log group. What is the reason for this recommendation? A. So that DBWn can write more efficiently B. To eliminate contention between LGWR and ARCn in archiving C. To eliminate a possible single point of failure D. So that there isn't contention between writing to the online redo log files and the control files
|
| A4: | Answer C is correct. You want to multiplex redo log group members for the same reason that you want to multiplex control files: to limit the places where you have a single point of failure in the database and instance. Answer A is incorrect because multiplexing redo log files has no net effect on database writer processes. Answer B is incorrect because eliminating contention between LGWR and ARCn is not a reason to multiplex. If contention will occur, it will occur regardless of how many copies of the redo logs are available. Answer D is incorrect because log writer is the process that writes to the online redo log files and not to the control files, so there is a limited possibility of having contention between the two. |
| 5: | You want to use OMF for maintaining and naming your online redo log files. What is the name of the initialization parameter that allows you to do this? |
| A5: | Answer A is correct. LOG_ARCHIVE_DEST_n is the initialization parameter that controls the location and naming of the online redo log files (as well as the control files) in an OMF managed database. Answer B is incorrect. FAST_START_MTTR_TARGET controls how often checkpoints occur, not where the redo log files are written. Answer C is incorrect. LOG_CHECKPOINT_INTERVAL is one of the Oracle 8i parameters that controlled checkpointing. Answer D is incorrect because LOG_FILE_DEST_n is not an Oracle 9i parameter. Answer E is incorrect because LOG_CHECKPOINT_TARGET is one of the Oracle 8i parameters that controls checkpointing frequency. |
| 6: | By default, checkpoints happen at least as often as what? [check all that apply] A. DBWn writes dirty buffers out to files. B. Redo log switches. C. Redo logs are archived. D. As the DBA forces the checkpoint.
|
| A6: | Answers B and D are correct. Checkpoints happen at least as often as redo log switches and at least as often as the DBA manually forces the checkpointing. Answer A is incorrect, technically, because checkpointing causes DBWn to write its dirty buffers out to file. Answer C is incorrect because there is no direct correlation between the frequency of checkpointing and archiving. |
| 7: | Which data dictionary view gives you information on the history of your online redo log files? A. V$LOG B. V$LOGHIST C. V$REDOLOG D. V$LOGFILE E. V$LOG_HISTORY
|
| A7: | Answer B is correct. You can find information on the history of your online redo log files in the V$LOGHIST view. Answer A is incorrect; V$LOG contains information from the control file on the redo log files. Answer C is incorrect because V$REDOLOG is not a V$ view in the Oracle 9i data dictionary. Answer D is incorrect because V$LOGFILE provides the current status of the redo log files in the instance. Answer E is incorrect; V$LOG_HISTORY gives you a history of the archived redo log files. |
| 8: | You want to limit contention on the different log members in your online redo log file groups. Which would be the best way (and the Oracle recommended way) to customize your configuration? A. Put the online redo log members on the same disk to reduce I/O contention. B. Put the online redo log members on different disks to reduce I/O contention. C. Put all the redo log members on the same disk as the archived redo logs to reduce contention. D. Have only redo log members in each online redo log group to reduce contention.
|
| A8: | Answer B is correct. To limit potential I/O contention you need to put the different redo log members on different disks or disk groups. Answer A is incorrect; putting the members on the same disk increases the chance of contention. Answer C is incorrect because putting not only all the redo log members on the same disk but also all the archive logs on the same disk as the redo log members increases the chance of I/O contention even further. Answer D is technically incorrect because Oracle recommends that you multiplex online redo log files. |
| 9: | When is a valid time to be running in ARCHIVELOG mode? A. When you want to be able to take backups while the database is running. B. When you want to save time and overhead by only running in ARCHIVELOG mode at night when the system is quiet. C. When you are running a data warehouse and the only updates are taking place as bulk loads. D. When your database is high volume and you don't want to impact users by adding background processes to the instance needlessly.
|
| A9: | Answer A is correct. One of the primary reasons to run your database in ARCHIVELOG mode is to be able to take online backups (or hot backups) of your database. Answer B is incorrect because archiving only when there is limited activity defeats the purpose for running in this manner. Answer C is incorrect because you can reload the data warehouse, and it is one place where running in NOARCHIVELOG mode is fairly safe. Answer D is incorrect because, if you have a high-volume database, you would likely want to be running in ARCHIVELOG mode so that you could recover any information in case of a failure. |
| 10: | What is the command to manually archive the online redo log files? A. ALTER SYSTEM ARCHIVE LOG ALL; B. ALTER DATABASE ARCHIVE LOG ALL; C. ALTER REDO_LOGS ARCHIVE LOG ALL; D. ALTER SYSTEM ARCHIVE LOG ALL REDO_LOGS;
|
| A10: | Answer A is correct; to manually archive the redo log files, you enter the ALTER SYSTEM ARCHIVE LOG ALL; command. |
