Oracle database Architecture

Architectureoracle

I need to know to correct order between these following steps, it's hard to find information on the internet on what happens first, i'd be really grateful if someone could give me the answer

A. Oracle searches for a init.ora to determine the configuration
parameters

B. The proces SMON sometimes executes an instance-recovery

C. memory for the SGA gets allocated and the background processes that are necessairy get started

D. Oracle searches the database files and redo log files

E. oracle starts an instance

F. oracle searches for the control files with the parameter CONTROL_FILES

Best Answer

A. Oracle searches for a init.ora to determine the configuration parameters

without a parameterfile, you will get an LRM-00109, ORA-01078 error, and nothing gets started

E. oracle starts an instance

we could argue what exactly starting an instance means, I have just simply put it here because this is the "exact" message that appears first when starting an instance

C. memory for the SGA gets allocated and the background processes that are necessairy get started

background processes like LGWR, DBWR, CKPT, SMON, PMON start in NOMOUNT, way before doing anything with control files...

F. oracle searches for the control files with the parameter CONTROL_FILES

thats the next one when going to MOUNT

D. Oracle searches the database files and redo log files

now tell me, how could the database perform an instance recovery without knowing where the datafiles and redo logs are

B. The proces SMON sometimes executes an instance-recovery

finally an instance recovery is performed, and database gets opened

Related Solutions

The best architecture to run Oracle sucessfully run on VMWare

You can run Oracle under VMware just fine. The big question is what is the current hardware spec and what VMware version and edition will be hosting the database?

Determine redo log size, number of redo log files

The recommendation is to have log switches occur a few times an hour. If you have a 100 log switches an hour with 50MB redo logs, you need e.g 1GB redo logs for 5 log switches/hour. 100*50 MB redo/hour for a 6.5GB database is unusually much though.

You can add the new logs, e.g:

ALTER DATABASE ADD LOGFILE THREAD 1 SIZE 1G;
ALTER DATABASE ADD LOGFILE THREAD 1 SIZE 1G;
ALTER DATABASE ADD LOGFILE THREAD 2 SIZE 1G;
ALTER DATABASE ADD LOGFILE THREAD 2 SIZE 1G;

Then force log switches until the new logs are used (CURRENT status) and old logs become ACTIVE/INACTIVE by:

ALTER SYSTEM SWITCH LOGFILE;

Then drop the old logs:

ALTER DATABASE DROP LOGFILE GROUP 1;
ALTER DATABASE DROP LOGFILE GROUP 2;
ALTER DATABASE DROP LOGFILE GROUP 3;
ALTER DATABASE DROP LOGFILE GROUP 4;

If you can't drop a logfile group because it is ACTIVE, do

ALTER SYSTEM CHECKPOINT;

so it becomes INACTIVE and try again.

Best Answer

Related Solutions

The best architecture to run Oracle sucessfully run on VMWare

Determine redo log size, number of redo log files

Related Question