What is redo log thread in oracle ?

On googling about the redo log thread, i have not found proper documentation  that clearly explains clearly what the redo log thread is . Here i am trying to cover the redo log threads in case of single instance and  RAC  taking reference from ASKTOM site .

Each instance has it's own personal set of redo and each redo thread is made up of at least two groups that have one or more members (files) .Two instances will never write to the same redo files - each instance has it's own set of redo logs to write to . Another instance may well READ some other instances redo logs - after that other instance fails for example - to perform recovery. Here is a scenario which helps us to understand the thread concepts .

Most V$ views work by selecting information from the corresponding GV$ view with a predicate "where instance_id = <that instance>". So V$SESSION in single Instance(i.e, 1) is actually 

SQL>select  *  from  gv$instance where inst_id= 1 ;

On a three node RAC database, if we select from v$session, we get sessions from that instance only. Selecting from GV$SESSION creates parallel query slaves on the other instances and gets the information back to our session. 

This works fine in almost all cases. There are few exceptions: in case of redo logs, the RAC instance must see all the redo logs of other instances as they become important for its recovery. Therefore, V$LOG actually shows all the redo logs, of all the instances, not just of its own. Contrast this with V$SESSION, which shows only sessions of that instance, not all. So, if there are 3 log file groups per instance (actually, per "thread") and there are 3 instances, V$LOG on any instance will show all 9 logfile groups, not 3. 

When we select form GV$LOG, remember, the session gets the information from other instances as well. Unfortunately, the PQ servers on those instances also get 9 records each, since they also see the same information seen by the first instance. On a three instance RAC, we will get 3X9 = 27 records in GV$LOG! 

To avoid this: 

1.) Always select from V$LOG, V$LOGFILE and V$THREAD in a RAC instance. GV$ views are misleading  or 

2.)  Add a predicate to match  THREAD#  with  INST_ID. (Beware: thread numbers are by default the same as the instance_id; but we may have defined a different thread number while creating the database) as 

SQL> select * from gv$log where inst_log=thread# ; 

Ref : http://asktom.oracle.com/pls/asktom/f?p=100:11:0::::P11_QUESTION_ID:18183400346178753



Enjoy    :-) 

User Managed Hot Backups in Oracle

A cold backup does have the somewhat bad side effect of wiping out our shared pool, our buffer cache and preventing our users from logging in to do work. Our database is  like a car, it runs better when it is warmed up.  If  we want to cold start it - be  prepared for rough running when we restart as we have to rebuild that shared pool, that  buffer cache and so on . I would never pick cold over hot given the chance.  No benefit, only downsides (Acc. to Tkye). The only kind of backup we do on our production systems here is hot .

There are two ways to perform Oracle backup and recovery : 

1.)  Recovery Manager (RMAN) : It is an Oracle utility that can backup, restore, and recover database files. It is a feature of the Oracle database server and does not require separate installation.

2.) User-Managed backup and recovery : We use operating system commands for backups and SQL*Plus for recovery. This method is  called user-managed backup and recovery and  is fully supported by Oracle, although use of RMAN is highly recommended because it is more robust and greatly simplifies administration.

There are basically two types of  backup .The backup are as 

1.) Consistent Backup :  This is also know as Cold Backup . A consistent backup is one in which the files being backed up contain all changes up to the same system change number (SCN). This means that the files in the backup contain all the data taken from a same point in time .

2.) Inconsistent Backup :  This is also known as Hot backup . An inconsistent backup is a backup in which the files being backed up do not contain all the changes made at all the SCNs . This can occur because the datafiles are being modified as backups are being taken. 

There are  some DBAs which prefer oracle user-managed backups.They put their database into backup mode prior to backing  up and take it out of backup mode after backup. If  we 're going to perform user-managed backups, we must back up all of the following file : 

• Datafiles
• Control files
• Online redo logs (if performing a cold backup)
• The parameter file (not mandatory )
• Archived redo logs
• Password file if used

The below diagram shows the Whole Database Backup Options :  

A hot backup requires quite a bit more work than cold backup.Below are steps required for Hot backup.

Step 1 :  Check the log mode of the database  Whenever we go for hot backup then the database must be in archivelog  mode . 
SQL> SELECT LOG_MODE FROM V$DATABASE ;
LOG_MODE
---------------
ARCHIVELOG

Step 2 :  Put the database into backup mode  If we are using the oracle 10gR2 or later , then we can put the entire database into backup mode and if we are using the oracle prior to 10gR2 ,then we have to put each tablespace in backup mode . In my case , I am having 11gR2 . 
SQL> alter database begin backup ; 
Database altered.
In case of oracle prior to 10gR2 use the below command as 
SQL> set echo off 
SQL> set heading off 
SQL>  set feedback off 
SQL> set termout  off 
SQL> spool backmode.sql 
SQL> select 'alter tablespace  '||name||'  begin backup ;'   "Tablespace in backup mode"  from v$tablespace;
SQL> spool off 
SQL>  @C:\backmode.sql 

Step 3 :  Backup all the datafiles  Copy all the datafile using the operating system command and Paste it on the desired backup location .Meanwhile,we can verify the status of the datafile by using the v$backup view  to check the status of the datafiles.

SQL> select  *  from  v$backup ; 
     FILE# STATUS                CHANGE# TIME
---------- ------------------ ---------- ---------
         1 ACTIVE                3967181 03-APR-12
         2 ACTIVE                3967187 03-APR-12
         3 ACTIVE                3967193 03-APR-12
         4 ACTIVE                3967199 03-APR-12
         5 ACTIVE                3967205 03-APR-12
         6 ACTIVE                3967211 03-APR-12
         7 ACTIVE                3967217 03-APR-12
         8 ACTIVE                3967223 03-APR-12
         9 ACTIVE                3967229 03-APR-12
The Column STATUS=ACTIVE  shows that the datafiles are in backup mode . 

Step  4  : Take out the database from backup mode  If we are using 10gR2 or above version of oracle , we use the below command to take out the database from backup mode as 

SQL> alter database end backup ; 
Database Altered 
If we are having version prior to 10gR2 , then we use the below command as above : 
SQL> set echo off 
SQL> set heading off 
SQL> set feedback off 
SQL> set termout  off 
SQL> spool end_mode.sql 
SQL> select  'alter tablespace  '||name||'  end backup ;'   "tablespace in backup mode"  from v$tablespace ; 
SQL> spool off 
SQL> @C:\endmode.sql 

Step 5 :  Switch the redolog file and backup archivelogs   After taking the database out of Hot Backup we must switch logfile (preferably more than once) and backup the archivelogs generated .We may backup archivelogs while the database is in backup mode but we must also backup the first archivelog(s) after the end backup. The best method to do both is to run the SQL command alter system archive log current. This switches the logfile but does not return the prompt until the previous redo log has been archived. We can run alter system switch logfile, but then we won't be sure that the latest redo log has been archived before we move on to the next step. 

SQL> alter system archive log current ; 
System altered.
SQL> / 
System altered.
Now backup the archivelogs to the backup location .


Step 6  : Back up the control file  Now , we can backup the controlfile as binary file and as human readable .We should use both methods to back up the control file; either one may come in handy at different times . The commands are as 
(Human readable)
SQL> alter database backup controlfile to trace ;  or 
Database altered.
SQL> alter database backup controlfile to trace as '<backup location>' ; 
Database altered.
(Binary format)
SQL> alter database backup controlfile to '<backup location>' ; 
Database altered.

Step 7 : Backup the passwordfile and spfile  We can backup the passwordfile and spfile though it is not mandatory.

Some Points Worth Remembering 

• We  need to backup all the archived log files, these files are very important to do recovery. 
• It is advisable to backup all of  tablespaces (except read-only tablespaces), else complete recovery is not possible.
• Backup of online redo log files are not required, as the online log file has the end of backup marker and would cause corruption if used in recovery.
• It is Preferable to start the hot backups at low activity time.
• When hot backups are in progress we  "cannot" shutdown the database in NORMAL or IMMEDIATE mode (and it is also not desirable to ABORT).

For More Click Here 


Enjoy    :-) 

Again A Time To Celebrate !!!!!!!

Today , my blog visitor's hits crossed 50,000 and offcourse a time to celebrate this moment . I am blogging from last year (i.e, march 2011)  and almost taken one year to acheive this . This month is very special to me as ever. There are lots of thing to celebrate in this month . Holi one of my favourite festival , My Birthday and this time my Blog Hits and others .

This blogs helps to find some  good friends whom i have never met and they are very kind and helpful to me. Whenever I am publishing any post on my blog, it boost up my confidence and motivate me to post  further . Every comment by you all is helping me to go more deep . Kindly free feel to comment and ask your doubt and i will try my best to give  solution .

I need  your support and wishes  to continue posting in future .  


Cheers  !!!!!!!