Skip to content

Archive for

Dec 6

RMAN 1.2 Backup Types

Physical and Logical Backups

  1. Copy of Datafile using an OS utility such as cp command.
  2. You can use this file to restore the database contents if you lose the disk file.
  3. Copy of the files like Datafiles, control files and redo log files.
  4. logical backups by using Oracle Data pump Export tool wherein you copy the definitions and contents all of the database’s logical componets such as table and so on.
  5. Its not complete backup and recovery solution.

Whole and Partial backups

  1. Backup of the entire database.
  2. Partial backup –> backup of a tablespace or datafile in a database.
  3. Datafile backup —> only a single OS file.
  4. Tablespace backup –> all the datafiles that are part of the tablespace.

Online and Offline Backups

  1. RMAN support both online ( HOT ) and offline ( COLD ) backups.
  2. An online backup is always inconsistent during a recovery , the application of the necessary archived log file will make the backup consistent.

Full and Incremental Backups

  1. It will contain complete backups of all the datafiles.
  2. It contain only the changed data blocks in the datafiles.
  3. You can make incremental backups only with the help of RMAN.

Consistent and Inconsistent Backups

What is SCN?

  1. SCN is an Oracle assigned number that indicates a committed version of a database.
  2. Its quite possible that different datafiles in the database might have a different SCN at any given point in time.
  3. If the SCN across all the datafiles are synchronized, it means that the data across the datafiles comes from a single point of time and thus, is consistent.
  4. During each checkpoint, the server makes all the datafiles SCN consistent w.r.t an identical SCN.
  5. It updates control file  with that SCN information.
  6. This synchronization of the SCN gives you a consistent backup of your database.

Inconsistent Backup

  1. If you backup your database while it’s running, you may end up with backups of the various data files at various time points and different SCN.
  2. This means your database are inconsistent, since the SCN aren’t identical across all the datafiles.
  3. If you’re using inconsistent backups, you must use the archive redo log files to make the data current and synchronize the SCN across the datafiles.
  4. If you’re using the RMAN offline backup, the database must be mounted before you can actually perform the RMAN backup. RMAN needs to update the target database control file.

Dec 6

RMAN 1.1 Types of Failure

Statement Failures

  1. When a program attempts to enter invalid data into an Oracle Table.
  2. Long data insertion job or data import job to fail midway between there is no more room to put the data in.
  3. Proper privileges to perform a task.

User process Failure

  1. abnormal disconnect or performing a terminal program error and losing the session connection.
  2. DBA not much work to do here.
  3. BG process rollback the uncommitted transaction  changes to the data and releases the locks.

Instance Failure

  1. Your database comes down such as Hardware , a power failure and an emergency shutdown procedure.
  2. An instance shutdown when the key Oracle bg process such as PMON shutdown because of an error condition.
  3. Check Alert log and trace files.
  4. Just restarting the Database instance using by STARTUP Command.
  5. The database was clearly shutdown and the database files aren’t synchronized.
  6. Oracle will perform an automatic instance or crash recovery at this point.
  7. Automatically perform a rollback uncommitted transactions by using the data from undo segments and roll forward committed changes it in the online redo logs.
  8. Don’t need to any sort of backup when restarting the database.

Network Failure

  1. Net listener, NIC and network connection has failed.
  2. DBA must configure the multiple network cards.

User Error

  1. Wrongly deleting  data from table and dropping a table you can use FLASHBACK feature.
  2. If the transactions not completed yet, rollback statement.
  3. Oracle LOGMINER also comes on handy situation like this.

Media Failure

  1. It occur when you lose a disk or a disk  controller fails.
  2. Examples of media failure  i) Head crash II) File corruption III)Overwriting or deleting of a datafile.
  3. Any one of the multiplexed control files are deleted or lost because of disk failure you must restore the missing control file from an existing control file.
  4. Datafiles, undo table space is deleted or lost because of a disk failure.If you lose one of these files, the instance may shutdown or may not sutdown in such case

         sql> shutdown abort;           Then

         sql>startup mount

          Restore the datafiles and recover it.

  1.         An entire redo log group lost. if you have atleast one member of the redo log group , your database instance can continue to operating normally.
  2. Restore the log file by copying one of  the other members of the same group.