INSTANCE and CRASH RECOVERY

1.type of checkpoint

   full checkpoint     alter system checkpoint

   incrimental checkpoint(logfile switch) alter system switch logfile.

   partial  checkpoint    alter tablespace backup/online/offline

 

2.phases of instance recovery 

   1.chek asydatafiles.

   2.cache rocovery(roll forward withredo),open database. 

   3.transaction recovery(roll back uncommited).

   4.commited datas in the datafiles.

 

3.RBA:every three second, checks record the RBA from the oldest entry in the checkopint quene in the control file.only in the event of a log switch,check wirte this formation to the headers of the datafiles.

logfile switch does not force all dirty buffers to be written to disk.

 

4.Crash recovery is used to recover from a failure either when a single-instance
database crashes or all instances of an Oracle Real Application Clusters database
crashes. Instance recovery refers to the case where a surviving instance recovers a
failed instance in an Oracle Real Application Clusters database.

 

5.tunning the duration of instance recovery, control the check redo blocks

   fast_start_mttr_target

   log_checkpoint_timeout

   log_checkpoint_interval

   fast_start_io_target

   v$instance_recovery

   v$fast_start_servers

   v$fast_start_transactions

6.tunning the phases of the roll forward and roll back

   1.recovery_parallelism

   2.fast_start_parallel_rollback (false,lo,hi)

   3.v$fast_start_servers; v$fast_start_transactions

 

  

6.MTTR:mean time to recovery

   MTBF:mean time between failures

 

7.