Do you know what's required to be an LUW database administrator? Read on.

Editor's note: This article is an excerpt from the book DB2 10.1/10.5 for Linux, UNIX, and Windows Database Administration (Exams 611 and 311): Certification Study Guide.

Q1. Which command sets the automatic instance restart every time the server is restarted?

1. db2set DB2AUTOSTART=ON
2. UPDATE DBM CFG USING START_STOP_TIME 60
3. db2iauto –on <InstanceName>
4. UPDATE DBM CFG USING AUTORESTART ON

Q2. Which command can you use to set the threshold rule for processor usage?

1. db2fodc -detect
2. db2pd –detect
3. db2top –detect
4. db2support –detect

Q3. Which command shows the best possible way of dropping column C1 from system-period temporal table T1?

A.   LOCK TABLE t1 IN EXCLUSIVE MODE; ALTER TABLE t1 DROP VERSIONING;

ALTER TABLE t1 DROP COLUMN C1; ALTER TABLE t1_history DROP COLUMN C1;

ALTER TABLE t1 ADD VERSIONING USE HISTORY TABLE t1_history; COMMIT;

B.   LOCK TABLE t1 IN EXCLUSIVE MODE; ALTER TABLE t1 DROP COLUMN C1;

ALTER TABLE t1_history DROP COLUMN C1; COMMIT;

C.   LOCK TABLE t1 IN EXCLUSIVE MODE; ALTER TABLE t1 DROP VERSIONING;

ALTER TABLE t1 DROP COLUMN C1; ALTER TABLE t1 ADD VERSIONING USE

HISTORY TABLE t1_history; COMMIT;

D.   LOCK TABLE t1 IN EXCLUSIVE MODE; ALTER TABLE t1 DROP VERSIONING;

ALTER TABLE t1_history DROP COLUMN C1; ALTER TABLE t1 ADD VERSIONING

USE HISTORY TABLE t1_history; COMMIT;

Q4. Which step is necessary during a redirected restore operation when you have storage groups defined in the database?

A.   RESTORE DATABASE … REDIRECT

SET STOGROUP PATHS FOR

RESTORE DATABASE … CONTINUE

B.   RESTORE DATABASE … REDIRECT

SET TABLESPACE CONTAINER FOR

RESTORE DATABASE … CONTINUE

C.   RESTORE DATABASE … REDIRECT

SET TABLESPACE CONTAINER FOR … AUTOMATIC STORAGE

RESTORE DATABASE … CONTINUE

D.  RESTORE DATABASE … REDIRECT

SET STOGROUP PATHS FOR … AUTOMATIC STORAGE

RESTORE DATABASE … CONTINUE

 Q5. What is I/O fencing?

1. Each member of a pureScale cluster gets a quorum of the disk storage space.
2. pureScale uses a stage area for page disputes between members.
3. Cluster services serializes I/O to the DB2 catalog to prevent locking conflicts.
4. DB2 ensures that it is not possible for the failed member to modify shared data on disk
5. None of the above

Q6. In pureScale, how do you remove a storage path from a storage group?

1. Use ALTER STOGROUP DROP <storage path>, then execute the ALTER TABLESPACE <name> REBALANCE statements for each table space assigned to that storage group.
2. Create a new storage group by using CREATE STOGROUP without the storage path; then for each table space assigned to that storage group, run ALTER TABLESPACE USING STOGROUP <new storage group>, followed by ALTER TABLESPACE <name> REBALANCE.
3. Use a redirected restore command with SET STOGROUP PATHS for <STOGROUPNAME> ON <storage paths>.
4. Create a new storage group by using CREATE STOGROUP without the storage path. Use the ADMIN_MOVE_TABLE utility to move all the tables and indexes to the new table spaces. Drop all the old table spaces by using DROP TABLESPACE and then DROP STOGROUP <old storage group>. If the old storage group was set as the default, you must run ALTER STOGROUP <new storage group> set as the default before you can drop the old storage group.
5. Either C or D
6. Either A or B
7. None of the above
8. All of the above

Q7. Which statement about insert time clustering tables (ITCs) is true?

1. ITC tables cannot be range partitioned tables.
2. ITC table records are clustered based only on timestamp columns.
3. The records in the ITC table are clustered according to the time they are inserted into the table.
4. ITC table records are clustered based on a cluster index created on the timestamp column.

Answers

Q1. The correct answer is C. The command db2iauto –on <instancename> enables the automatic start of an instance after each server restart. The START_STOP_TIME DBM CFG specifies the time, in minutes, within which all database partition servers must respond to a START DBM or a STOP DBM command. It is also used as the timeout value during ADD DBPARTITIONNUM and DROP DBPARTITIONNUM operations. The AUTORESTART DBM CFG parameter determines whether the database manager automatically initiates crash recovery when a user connects to a database that had previously terminated abnormally. If this configuration parameter is not set, the user must issue an explicit restart database command before he or she can connect to the database. The DB2AUTORESTART is an incorrect parameter to DB2 registry setting.

Q2. The correct answer is A. You can use the db2fodc –detect command to collect data when the environment reaches certain thresholds. The -detect parameter is compatible with the following main data collection parameters:

• -cpu
• -memory
• -connections
• -hadr
• -hang
• -perf
• -cpl
• -preupgrade

To collect information about the activity consuming more than 90 percent of CPU for more than 40 seconds (triggercount * interval) for the monitoring duration of five hours, execute this command:

db2fodc -cpu basic -detect us_sy">=90" sleeptime="30" iteration="10"

interval="10" triggercount="4" duration="5"

Q3. The correct answer is A. When you try dropping a column on a system-period temporal table, you will receive an error similar to the following:

SQL0196N Column "C1" in "T1" cannot be dropped. Reason code ="5".

The reason code 5 indicates the column cannot be dropped because it is a column in a system-period temporal table. The solution is to follow these steps:

• Alter the system-period temporal table to drop versioning. This breaks the link between the tables.
• Alter the system-period temporal table to drop the column.
• Alter the table that was the history table to drop the column.
• Alter the former system-period temporal table to add versioning. This reestablishes the link between the tables.

Q4. The correct answer is A. From DB2 10.1 onward, when you define storage groups, you will have to set the storage path for the storage groups during the redirected restore operation instead of setting the container paths for table spaces as in previous versions.

Q5. The correct answer is D. Before you recover a failed member, DB2 cluster services ensures that it is not possible for the failed member to modify share data on disk. I/O fencing requires SCSI-3 PR commands to be enabled, which is not possible on virtual disks.

Q6. The correct answer is E. In pureScale, you cannot alter table spaces to a different storage group or drop a storage path from a storage group. Both C and D will work; however, C will require a service disruption to complete, but it will work in all cases. D can be performed online, but not all tables are supported by the ADMIN_MOVE_TABLE utility.

Q7. The correct answer is C. ITC tables provide an effective way of maintaining data clustering based on the insert time with easier space management. The characteristics of ITC tables are as follows:

• ITC table records are clustered based on their insert time.
• ITC tables have similar characteristics to MDC tables—they use block-based data allocation and a dimension block index. However, the data organization scheme is based on an implicitly created virtual dimension (record insert time) instead of explicitly stated dimension columns.
• ITC tables are created by using the CREATE TABLE ... ORGANIZE BY INSERT TIME clause.
• ITC tables’ space reclamation can be done by executing the REORG TABLE ... RECLAIM EXTENTS command.
• ITC tables can be range partitioned tables.

Learn more with the book DB2 10.1/10.5 for Linux, UNIX, and Windows Database Administration (Exams 611 and 311): Certification Study Guide.