Chapter07Instance & Media Recovery Structures

After completing this lesson,you should be able to do the following:

• Describe the Oracle processes,memory structures,and files relating to recovery
• Identify the importance of checkpoints,redo log files,and archived log files
• Describe ways to tune instance recovery

Types of Checkpoints

• Full checkpoint
  • All dirty are written
  • SHUTDOWN NORMAL,IMMEDIATE,OR TRANSACTIONAL
  • ALTER SYSTEM CHECKPOINT
• Incremental checkpoint(Fast-Start checkpoint)
  • Periodic writes
  • Only write the oldest blocks
• Partial checkpoint
  • Dirty buffers beloging to the tablespace
  • ALTER TABLESPACE BEGIN BACKUP
  • ALTER TABLESPACE tablespace OFFLINE NORMAL

Oracle 的两个或多个control file，其中一个损坏掉了，Oracle就会罢工；redo log files 中损坏掉了部分，只要有可用的，Oracle就不会罢工；

Multiplexed Control Fiels

• Control File Function
  • The control file is a binary file that describes the structure of the database.It must be available for writing by the oracle server whenever the database is mounted or open.Its default name is operating system-dependent.Without this file,the database cannot be mounted and recovery or re-creation of the control file will be required.The recommended configuration is a minimun of two control files on different disk to minimize the impact of a loss of one control file.
• Control File contents
  • Database name
  • Time stamp of database creation
  • Synchronization infromation(checkpoint and log sequence information) needed for recovery
  • Names and location of datafiles and redo log files
  • Archiving mode of the database
  • Current log sequence number
  • Recovery Manager backup meta data

Phases for Instance Recovery

1. Datafiles out-of-synch
2. Roll-forward(redo)
3. Commited and non-commited data in files
4. Roll back(undo)
5. Commited data in files

Crash Recovery Phases

• Unlike media recovery,Oracle performs crash recovery and instance recovery automatically after an instance failure.Crash and isntance recovery recover a database to its transaction-consistent state just before instance failure.By definition,crash recovery is the recovery of a database in a single-insance configuration or an Oracle Real Application Clusters configuration in which all instance have crashed.In contrast,instance recovery is the recovery of one fiiled instance by a live instance in an Oracle Real Application Clursters configuration.
• Crash Recovery Phases

1. The datafiles are not synchronized.
2. During the crash recovery or roll forward phase,all of the changes recorded in the redo log files since the last checkpoint are reapplied to the datafiles.This phase also regenerates undo or rollback data.
3. The datafiles now contain committed and perhaps uncommitted changes.The database is opened.
4. During the transaction recovery or rollback phase,any changes that were ont actually committed are rolled back.
5. The datafiles now contain only commited changes to the database.

EngLish corner

head crash
accidentally erased
strategy
if available,apply archived redo log files to recover data committed since the last backup.
evolutionary process
transaction volume that affects desired frequency of backups
volume 
frequency 
desired 
concurrence
malfunction
evaluate potential failures in your environment
develop a strategy dictated by business,operational,and techical requirements
体系结构
architecure
asynchronous
synchronous
attempts to 
alert
multiplexed redo log files
guard against failure
guard