Fast Recovery Area with 2 databases

Hey, I have to create 2 databases on my 2 nodes RAC ( 11gR2 on SLES11)
Currently I am configuring the second database.
Now I am confused regarding the FRA and DATA Diskgroup I created.
Do I have size those diskgroups now ?
eg
I set up the FRA size for my first db to use the whole diskspace for the diskgroup
My Data is 10G and FRA is 20G big.
Do I need to modify the settings for the first db now - to the half of the size
or ist Oracle intelligent and will calculate both values for the max sizes ?
Chris
A bit confusing
What is the answer ?
+FRA 20000 MB
Scenario 1:
db1 -
FRA Size = 20000 MB
db2 -
FRA Size = 20000 MB
or scenario 2
+FRA 20000 MB
Scenario 1:
db1 -
FRA Size = 10000 MB
db2 -
FRA Size = 10000 MB

We've had the experience that Oracle 'thinks' the FRA is full resulting in a database hang, while no redo-logs could be archived.
This while the disk was not full at all.
So I don't think Oracle is intelligent enough to start looking for the usage of the physical diskspace. It just used the size of the FRA in the spfile.
Therefore it is recommended to divide the available space in two.
The above experience we've has were with Oracle 10.2.0.3. Much may have improved with newer versions, so I can be mistaking here
Cheers,
FJFranken
My Blog: http://managingoracle.blogspot.com

Similar Messages

Fast recovery area 10g

Hi all,
Is there fast recovery area for 10g?
How is it different from flash recovery?
Thanks a lot,

asahide wrote:
Hi,
Is there fast recovery area for 10g?- The flash recovery area is the most powerful tool available from Oracle 10g, that plays a vital role in performing database backup & recovery operations. From Oracle 11g release2, flash recovery area is called as fast recovery area.
<<http://satya-dba.blogspot.jp/2009/02/flash-recovery-area.html>>
How is it different from flash recovery?NAME is different :)
Regards,Please do not feed the troll.

Move fast recovery area

Hi,
I need to move the fast recovery area between disk groups in ASM (11g RAC). Is it as simple as change the db_recovery_file_dest parameter?

user492400 wrote:
Hi,
I need to move the fast recovery area between disk groups in ASM (11g RAC). Is it as simple as change the db_recovery_file_dest parameter?I don't know what the "fast" recovery area is, but if you mean the "flash" recovery area .... ;-)
then yes, it is as simple as changing the value of db_recovery_file_dest

Fast Recovery Area space utilization

Hello!
I had an issue with my FRA getting filled consistently lately and it would give me all sorts of issues. I checked the alert.log and found this:
ORA-19815: WARNING: db_recovery_file_dest_size of 6815744000 bytes is 100.00% used, and has 0 remaining bytes available.
You have following choices to free up space from recovery area:
1. Consider changing RMAN RETENTION POLICY. If you are using Data Guard,
then consider changing RMAN ARCHIVELOG DELETION POLICY.
2. Back up files to tertiary device such as tape using RMAN
BACKUP RECOVERY AREA command.
3. Add disk space and increase db_recovery_file_dest_size parameter to
reflect the new space.
4. Delete unnecessary files using RMAN DELETE command. If an operating
system command was used to delete files, then use RMAN CROSSCHECK and
DELETE EXPIRED commands.
1) I checked the retention policy in RMAN it was : CONFIGURE RETENTION POLICY TO REDUNDANCY 1; # default
2) I dont have tape backup option
3) I kept increasing the db_recovery_file_dest_size and it just keep growing
SQL> show parameter db_recovery_file_dest_size
NAME                    TYPE     VALUE
db_recovery_file_dest_size     big integer 10G
4) I did fix it by manually deleting files through the OS ( yea I know rookie mistake) , crosschecking, then delete expired archivelog all;
SQL> SELECT Sum(percent_space_used)-Sum(percent_space_reclaimable) "PCT_SPACE_USED_NOT_FREEABLE" FROM v$flash_recovery_area_usage;
PCT_SPACE_USED_NOT_FREEABLE
          9.18
I dont know what to do to avoid this in the future, I changed my retention policy to a window of 7 days but I have no idea if that will help. I dont want to keep taking manual action , I want the files to get deleted automatically.
Help me out please! Any ideas?
RMAN> show all;
RMAN configuration parameters for database with db_unique_name DB11G are:
CONFIGURE RETENTION POLICY TO RECOVERY WINDOW OF 7 DAYS;
CONFIGURE BACKUP OPTIMIZATION ON;
CONFIGURE DEFAULT DEVICE TYPE TO DISK; # default
CONFIGURE CONTROLFILE AUTOBACKUP ON;
CONFIGURE CONTROLFILE AUTOBACKUP FORMAT FOR DEVICE TYPE DISK TO '%F'; # default
CONFIGURE DEVICE TYPE DISK PARALLELISM 1 BACKUP TYPE TO BACKUPSET; # default
CONFIGURE DATAFILE BACKUP COPIES FOR DEVICE TYPE DISK TO 1; # default
CONFIGURE ARCHIVELOG BACKUP COPIES FOR DEVICE TYPE DISK TO 1; # default
CONFIGURE MAXSETSIZE TO UNLIMITED; # default
CONFIGURE ENCRYPTION FOR DATABASE OFF; # default
CONFIGURE ENCRYPTION ALGORITHM 'AES128'; # default
CONFIGURE COMPRESSION ALGORITHM 'BASIC' AS OF RELEASE 'DEFAULT' OPTIMIZE FOR LOAD TRUE ; # default
CONFIGURE ARCHIVELOG DELETION POLICY TO BACKED UP 1 TIMES TO DISK;
CONFIGURE SNAPSHOT CONTROLFILE NAME TO '/u01/app/oracle/product/11.2.0/db_1/dbs/snapcf_DB11G.f'; # default

SQL> select * from v$flash_recovery_area_usage order by 1;
FILE_TYPE     PERCENT_SPACE_USED PERCENT_SPACE_RECLAIMABLE
NUMBER_OF_FILES
ARCHIVED LOG               4.59                0
     22
BACKUP PIECE               0                0
     0
CONTROL FILE               0                0
     0
FILE_TYPE     PERCENT_SPACE_USED PERCENT_SPACE_RECLAIMABLE
NUMBER_OF_FILES
FLASHBACK LOG               0                0
     0
FOREIGN ARCHIVED LOG          0                0
     0
IMAGE COPY               0                0
     0
FILE_TYPE     PERCENT_SPACE_USED PERCENT_SPACE_RECLAIMABLE
NUMBER_OF_FILES
REDO LOG               0                0
     0

Fast Recovery Area free space doubt

Hi All,
I have oracle 11.2.0.3.0 in Windows 2008 server 64 bit , i have Database with ASM and i have enabled Flashback in my database ...
I have assigned 8GB for db recovery area size , in that i have Rman backups , flashback logs, archive logs ..
My doubt is oracle automatically free space in case of no free space in FRA , i have checked for RMAN backups , automatically my obsolete backups are deleted from FRA but i have the doubt about flashback logs ...my flashback retention period is 1 day..
Is oracle automatically deletes the obsolete flashback logs in case of no free space in FRA aur manually i need to delete?
Thanks and regards,
Vikash Chauradia

Vikash Jain (DBA Trainee) wrote:
Hi All,
I have oracle 11.2.0.3.0 in Windows 2008 server 64 bit , i have Database with ASM and i have enabled Flashback in my database ...
I have assigned 8GB for db recovery area size , in that i have Rman backups , flashback logs, archive logs ..
My doubt is oracle automatically free space in case of no free space in FRA , i have checked for RMAN backups , automatically my obsolete backups are deleted from FRA but i have the doubt about flashback logs ...my flashback retention period is 1 day..
Is oracle automatically deletes the obsolete flashback logs in case of no free space in FRA aur manually i need to delete?Please read,
http://docs.oracle.com/cd/E11882_01/backup.112/e10642/rcmmaint.htm#BRADV89615
Aman....

EMERGENCY: List of files in Recovery Area not managed by the database

Hi,
this is an emergency request....
I have this data guard set up. And i accidently dropped online redo logs, standby logs and the archive log except the current online redolog.
i actually ran to cleanup the whole directory which included the controlfile and the datafiles (was on the wrong location on ASM)
but the controlfile and the datafiles did not get deleted. i can still see them on the ASM.
and now when i am trying to restore the archive logs, but i keep getting error as shown below...
WARNING: A file of type ARCHIVED LOG may exist in
db_recovery_file_dest that is not known to the database.
Use the RMAN command CATALOG RECOVERY AREA to re-catalog
any such files. This is most likely the result of a crash
during file creation.
so when i did the below i was getting error as:
RMAN> CATALOG RECOVERY AREA;
using target database control file instead of recovery catalog
searching for all files in the recovery area
no files found to be unknown to the database
List of files in Recovery Area not managed by the database
==========================================================
File Name: +DATA/afcpdg/controlfile/controlfile01.ctl
RMAN-07526: Reason: File is not an Oracle Managed File
File Name: +DATA/afcpdg/onlinelog/group_1.304.718194815
RMAN-07527: Reason: File was not created using DB_RECOVERY_FILE_DEST initialization parameter
number of files not managed by recovery area is 82, totaling 168.10GB
But i do see those data files that RMAN is complaining as shown above in the ASM.
I HOPE I DID NOT MESS UP THE CONTROLFILE and the DATFILES
CAN SOMEONE HELP ME FIX THIS MISTAKE THAT I DID.
Right now the data guard is running fine, the logs are getting shipped from primary and the getting applied at the DG.
And i am able to see all the redo logs got deleted.
But not sure how the deleted redo logs got created in the ASM (i did not create them manually)
Thanks.
Philip.
Edited by: user8898644 on May 5, 2010 8:52 AM
Edited by: user8898644 on May 5, 2010 9:16 AM
Edited by: user8898644 on May 5, 2010 11:43 AM

Yes this happened on the standby database.
I had to restore the archive logs because i deleted all the archive logs which got shipped from the primary.
I did re-create the standby logs. But when i checked back the redo logs that i deleted came back again.
I have no idea how they can back again. is it a feature of ASM where a mirrored copy of the files are stored on ASM and they re-create them when accidently dropped.
Can you help me understand how the redologs got created again automatically.
Right now the data guard is running fine but i do see the error
WARNING: A file of type ARCHIVED LOG may exist in
db_recovery_file_dest that is not known to the database.
Use the RMAN command CATALOG RECOVERY AREA to re-catalog
any such files. This is most likely the result of a crash
during file creation.
and when i do the following
RMAN> catalog recovery area;
i do see errors as below...
List of files in Recovery Area not managed by the database
==========================================================
File Name: +DATA/afcpdg/controlfile/controlfile01.ctl
RMAN-07526: Reason: File is not an Oracle Managed File
File Name: +DATA/afcpdg/onlinelog/group_1.304.718194815
RMAN-07527: Reason: File was not created using DB_RECOVERY_FILE_DEST initialization parameter
File Name: +DATA/afcpdg/onlinelog/group_2.372.718194833
RMAN-07527: Reason: File was not created using DB_RECOVERY_FILE_DEST initialization parameter
File Name: +DATA/afcpdg/datafile/system.596.715610261
RMAN-07527: Reason: File was not created using DB_RECOVERY_FILE_DEST initialization parameter
File Name: +DATA/afcpdg/datafile/undotbs1.571.715603525
number of files not managed by recovery area is 82, totaling 168.10GB
we do have the db_create_file_dest and db_recovery_file_dest pointing to the same location as +DATA
How do i fix the above error?
Thanks in advance.

Proper layout of datafiles, archivelogs, flash recovery area

Hi all,
Sorry a bit of a newb,
Have a question about how to layout my db files.
My understanding is that control and archivelog files should be multiplexed to a separate disk than the datafiles.
Is this also true for the flash_recovery_area?
Thanks for any help.

geeter wrote:
Hi all,
Sorry a bit of a newb,
Have a question about how to layout my db files.
My understanding is that control and archivelog files should be multiplexed to a separate disk than the datafiles.
Is this also true for the flash_recovery_area?
Thanks for any help.Lose your control file or your redo file, and you have a problem. Therefore, in a professional environment, the control file and redo logs need to be protected by multiplexing to different physical drives (not different partitions on same disk). One of those locations can be the same DASD as the database files.
Archivelogs should be bounced from online to nearline to offline storage. These days 'online' is inside flash recovery area. If you are paranoid, you can send archivelogs to multiple destinations (multiplex), although some form of disk protection (RAID 1 ... or 5 - ugh!) should be sufficient.
Flash Recovery Area can be inside ASM or on regular file system. If in ASM, consider using either RAID 1 under ASM and external failgroups, or normal failgroup mode to allow the system to do it's own RAID 1 equivalent. If on regular file system, use standard disk protection (RAID 1 or 5) to do the protection.
Oracle uses the SAME (Stripe and Mirror Everything) philosophy, so separating Fast Recovery Area (new name - used to be Flash Recovery Area in the ancient days of Oracle 10g) from database files is not needed.
Highly recommend you read the first chapter of the Database Concepts Manual and the first chapter of the Database Administrator Guide to get used to the docs, adn then use that as a pringboard (via table of contents) to the sections that answer your question. Start at http://tahiti.oracle.com to get to those docs - they are on the front page of the doc set for your favorite version.

How to disable flash recovery area feature?

Please guide how to disable flash recovery area with commands
Thanks,
Waheed

How about searching for it first,
http://www.oracle.com/pls/db92/db92.federated_search
Adith

Problem with recovery area

Hi everybody
I have a problem in a RAC environment
This is my environment:
Oracle RAC 10.2.0.4 with 3 nodes
RHEL 5.5
The node 3 stopped working and the alert log shows me the following:
Fri Mar 4 16:24:38 2011
ORA-19816: WARNING: Files may exist in db_recovery_file_dest that are not known to database.
ORA-27041: unable to open file
Linux Error: 13: Permission denied
Additional information: 2
Fri Mar 4 16:24:38 2011
WARNING: A file of type ARCHIVED LOG may exist in
db_recovery_file_dest that is not known to the database.
Use the RMAN command CATALOG RECOVERY AREA to re-catalog
any such files. This is most likely the result of a crash
during file creation.
I ran CATALOG RECOVERY AREA command in RMAN and the result indicates that some headers of archivelogs are corrupted.
The DB service was hung and I stoped the instance in node 3.
Now my DB is operative but i have fear that if start the instance in node 3 the service will be hung.
Any idea?
Thanks so much
pd: sorry for my english

SQL> show parameter db_recovery_file_dest
NAME                         TYPE      VALUE
db_recovery_file_dest               string      /u01/app/oracle/flash_recovery
                               _area
db_recovery_file_dest_size          big integer 4000M

Simplify recovery after opening the database with the RESETLOGS option

Hi All,
I just have doubt ,regarding the topic which is given in Oracle 11g Exam Syllabus for upgrading from 9i to 11g OCP.The opic is
*"Simplify recovery after opening the database with the RESETLOGS option"* in this topic what one need to study.
As i am little bit confuse related to "Recovery through RESETLOGS and reset of the incarnation of the database".
Kindly correct me if i am wrong
Thanks in advance

vk82 wrote:
Hi All,
I just have doubt ,regarding the topic which is given in Oracle 11g Exam Syllabus for upgrading from 9i to 11g OCP.The opic is
*"Simplify recovery after opening the database with the RESETLOGS option"* in this topic what one need to study.
As i am little bit confuse related to "Recovery through RESETLOGS and reset of the incarnation of the database".
Kindly correct me if i am wrong
Thanks in advanceyou are correct so you can mark this thread as answered
Handle:     vk82
Status Level:     Newbie (35)
Registered:     Nov 21, 2010
Total Posts:     733
Total Questions:     180 (115 unresolved)

How to use Flash recovery area only for flashback database feature

I would like to use flash recovery area only for flashback database feature, which means, flash recovery area will store only flashback logs.
I dont want to use that file system for any other files, like archive log files, database backups, control file copies, multiplex redo log, etc
is it possible? I want to use this feature only for flashback database option and hence i dont want to disturb any existing configuration in my database
including backup jobs. Is it required to store archive redo log files also in flash recovery area to use the flashback database feature?
I certainly can't afford to have a copy of entire database in this file system as size of my DB is more than the flash recovery area file system I have.
Thanks
Sarayu

user13312943 wrote:
Aman,
I think i was not clear in my question. Let me try again
My only requirement is to use flashback database feature (or be prepared to use if required). I dont want to create a large file system to use this FRA to store all these files. Out of all the files being stored here copy of data files seems to be consuming more space. I see these files are classified as transient and permanent files. I am okay with permanent files. I fear that this file system may consume more space if i store backup files on disk. I want database backups to go to tapes, not to the FRA disk. Is it possible?
Thanks for your reply
Okay, well in that case you can use RMAN and push the backups to the tape drives. FLB logs would be stored on the FRA.
Aman....

Flash recovery area issue. I can not open the database??

Hi all
I am using oracle 11.2.0 .1.0 on windows
when i tried to start the database I got the following error.
ORA-03113: end-of-file on communication channel
I clear out all information in Trace file. I cleared out all operating system level audit trails. then i deleted archivelog files using operating system level delete. still my database is not opening???
I tried to start this using RMAN but got the same error. now i have deleted the files from flash recovery area how can i let oracle system to know that those are deleted and there is enough space. Rman does not execute crosschek commands at mount state of the database?
Need solution.
Regards
MAlik

malikdba wrote:
the last option which ask to delete the files. i already have deleted the files at operating system level. now the issue is how to let the system that the space is free in flash recovery area? rman is not running crosscheck command at mount stage of the target database. What steps should i follow now?
Thanks
malik
Are you saying you are getting an error while trying to crosscheck at mount? Or you simply don't believe it will work?
oracle:orcl$ rman target /
Recovery Manager: Release 11.2.0.2.0 - Production on Thu Jul 11 10:52:48 2013
Copyright (c) 1982, 2009, Oracle and/or its affiliates. All rights reserved.
connected to target database (not started)
RMAN> startup mount;
Oracle instance started
database mounted
Total System Global Area     839282688 bytes
Fixed Size                     2231128 bytes
Variable Size                503317672 bytes
Database Buffers             331350016 bytes
Redo Buffers                   2383872 bytes
RMAN> crosscheck archivelog all;
using target database control file instead of recovery catalog
allocated channel: ORA_DISK_1
channel ORA_DISK_1: SID=18 device type=DISK
specification does not match any archived log in the repository
RMAN>
A perfectly normal crosscheck at the mount stage

Entity Bean Rollback fails: caches are out of synch with the database

We are encountering a critical issue where the rollback of a session transaction rolls back any database changes, but doesn't roll back bean changes in memory. When we next access the bean, it returns information that should have been rolled back. The memory cache is out of synch with the database.
We have a simple session bean S, using container managed transactions. We also have an entity bean E using bean managed persistence. E has a single member <name>.
We implement a method s() in S with <trans-attribute> Required as follows:
public void s() {
Context lookupContext = getLookupContext();
SHome shome = (SHome) lookupContext.lookup(SHome.NAME);
S s = shome.findByPrimaryKey(thePrimaryKey);
System.out.println("-- current name = " + s.getName());
s.setName("IGNORE");
EJBContext ejbContext = (EJBContext) getContext();
ejbContext.setRollbackOnly();
Since this is BMP, the setName call issues our SQL update against the S record. When the session transaction rolls back, the database changes are also rolled back. But the entity bean in memory shows the modified name. We can see this in the next call to s() since the getName() returns IGNORE rather than the original value. We have also gotten debugged into the underlying instance cache associated with the S bean descriptor. I.e.
AbstractEJBHome ejbHome = (AbstractEJBHome) lookupContext.lookup(SHome.NAME);
AlternateHashTable aht = ejbHome.getBeanDescriptor().getCache();
Looking at this map of bean instances also shows that the bean instance for S is not rolled back.
From previous conversations with the server technologies team, we understand that during the transaction, we should be working on a cloned copy of the S bean. Upon rollback, the copies should be discarded rather than merged back to the cached bean. They seem to be merged regardless of rollback state.
We are using OC4J 10.1.3.0.0 (build 060119.1546.05277).

Actually, what one or the developers in my group found out is two things:
1) The clone is a shallow copy.They copy the references but not objects themselves. This has been causing problems for us.
2) It seems that you don't work on a copy and you work on the original and they keep a copy. Now, this is fine IF you don't have multiple threads working on the same Entity bean and we moved our application from Weblogic where it throws an exception if that Entity bean is in use to OC4J which is suppose to copy the bean and doesn't so you have multiple threads possibly modifying the same object.

Disk array configurations with oracle redo logs and flash recovery area.

Dear Oracle users,
We are planning to buy the new server for oracle database 10g standard edition. We put oracle database file, redo logs, and flash recovery area on each of disk array. My question is what is the best disk array configuration for redo logs and flash recovery area? RAID 10 or RAID 1? Is that possible we can duplicate Flash recovery area to the other location (such as net work drive) at the same time? Since we only have single disk array controller to connect to the disk arrays, I am try to avoid the single failure that will lose archive logs and daily backup.
thanks,
Belinda

Thank you so much for the suggestion. Could you please let me know the answer for my question of FRA redundancy?
“Is that possible we can duplicate Flash recovery area to the other location (such as net work drive) at the same time? Since we only have single disk array controller to connect to the disk arrays, I am try to avoid the single failure that will lose archive logs and daily backup.”

I have taken pictures of items that are linked to a database that I am building. How can I bypass importing into iPhoto and simply downloading the images as "files" not "photos" so I can access them with my database? thanx.

I have taken pictures with both my Sony camera and my iPhone of items that are linked to a database that I am building, as well as my website. How can I bypass importing the images into iPhoto and simply downloading the images as "files" not "photos" so I can access them with my database? thanx.

If your Sony has a removable memory card you can use a card reader to copy the image files from to a folder on your Desktop and then move them anywhere you'd like.
Since the iPhone doesn't have a removable memory card you can try using Image Capture to see if you can manually upload the files to a folder on the Desktop.
If you have to import the photos, which are image files, into iPhoto you can then export them out of iPhoto to the Desktop and go from there to your database. Just because they are in iPhoto doesn't prevent you from using them elsewhere.
OT

Fast Recovery Area with 2 databases

Similar Messages

Maybe you are looking for