Disk failure
A disk failure is the most common form of database failure. With the cost of RAID solutions (including mirroring and parity solutions such as RAID-1 and RAID-5) coming down, most sites have moved to a hardware solution to prevent a single disk failure from impacting the database. If your site does not use these solutions, then at some point in your career you will have to perform a recovery from a disk failure.
We recommend multiple control files, separate disks for archive files, and mirroring redo logs either with redo log groups or operating system mirroring. With mirrored redo logs and multiple control files on separate disks, no single disk failure will cause a loss of the database. If you have only one control file or only one redo member per group, then losing the disk containing the control file or the active redo log would be catastrophic to the database. Placing archive files on a separate disk from the redo logs (if mirroring is not used) at least ensures that you can recover the database up to the point of the last archive file.
In the case of a disk failure, you have the following options:
Restore the datafile to a new disk, rename it using the ALTER DATABASE RENAME DATAFILE command, and recover the file using archive files and online redo logs.
If you are not running in archivelog mode, you can still use the tablespace point-in-time recovery procedure from a cold backup.
Recover the entire database using a cold backup.
Place the tablespace containing the missing ...
Get Oracle Database Administration: The Essential Refe now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
