A database is running In ARCHIVBXXMS mode. It has two online redo log groups and each group
has one member. A LGWR Input/output (I/O) fells due to permanent media failure that has
resulted In the loss of redo log file and the LWGR terminates causing the instance to crash.
The steps to recover from the loss of a current redo log group member in the random order are as
follow.
1) Restore the corrupted redo log group.
2) Restore from a whole database backup.
3) Perform incomplete recovery.
4) Relocate by renaming the member of the damaged online redo log group to a new location.
5) Open the database with the RESETLOGS option.
6) Restart the database instance.
7) Issue a checkpoint and clear the log.
Identify the option with the correct sequential steps to accomplish the task efficiently.
A.
1, 3, 4, and 5
B.
7, 3, 4. and 5
C.
2, 3, 4, and 5
D.
7, 4, 3. and 5
E.
Only 6 is required
Explanation:
http://docs.oracle.com/cd/B10501_01/server.920/a96572/recoscenarios.htm#11771
“If the lost redo log group is CURRENT, you must restore, perform cancel-based incomplete
recovery, and open the database using the RESETLOGS option. A redo log group with a
CURRENT status indicates that LGWR is currently writing to it. To recover from this failure, you
must restore the database from a whole backup and perform cancel-based incomplete recovery.
Then, you should open the database using the RESETLOGS option.”
So something is wrong with the answer.
C is correct answer.
for sure, you cannot checkpoint if its already crashed, however you can try 🙂