GoldenGate for Dev/QA Database Refresh

Hi,
We have 600 GB Production database. We have to daily refresh one schema (600 GB) daily through export / import Data Pump (Oracle 10.2.0.5). Before when the data used to be around 400 - 500 GB, the refresh used to complete within 7 AM the other day, so it would be ready by the time development team comes to office. But, now, since the business is growing so thus the schema size.
I know that GoldenGate is used for replication. Can we use GoldenGate for this type is scenario ? If so, what are the steps to be followed ?
Database Version - Oracle 10g (10.2.0.5)
OS - SuSE Linux
Thanks in advance.

Sebastian Solbach (DBA Community) wrote:
Hi,
if this really is a whole refresh (import) of a schema you might want to give transportable tablespace a look. Maybe this solves your problem.
Regards
SebastianHi Sebastian,
Correct me if I am wrong. For TTS, we need the tablespace to put in readonly mode, before exporting ? Correct ??
The total size of my tablespace is ~ 600 GB.
Edited by: 893593 on Aug 8, 2012 9:07 PM

Similar Messages

Database refresh dependency for iBot

I have an iBot which uses a conditional request which checks if the database refresh is complete. All this does is that the report does not run. How do I make it such that the iBot keeps trying until the database refresh is over & once the refresh is over, it does not run again.
Thanks,
Dinesh.

I had a similar requirement, send an email once a day, as soon as something is ready, but then once it has run once, then dont send it again.
I created a REPOS variable and pulled in the date from the ibot job table as follows :
select max(convert(DATETIME,left(END_TS,11),102)) "DATE"
from
S_NQ_INSTANCE I,
S_NQ_JOB J
WHERE
J.JOB_ID = I.JOB_ID AND
J.NAME = 'ibot_last_run_date' and
I.EXIT_CODE > 0
Then build an answers request to return a row when this variable value is less than today. That should do it (works for me).
My Repos variable refreshes every 10 mins, the ibot is polling every 10 mins on an off-set schedule.
Only one ibot is run daily, and it runs within 10 mins of the data being ready.

Database Refresh: Save Job Components?

Our DBAs will be performing a database refresh of our dev/test environments based on a current Production model. To account for differences in our existing jobs/chains between Production and our other environments, it would be ideal to save off the existing objects in dev/test before the refresh. Once the refresh is complete, we could then restore the jobs/chains.
Has anyone experienced similar?
- If a schema refresh is done of the particular owner of the jobs, will the underlying jobs, chains, programs, steps, etc. be included in the schema refresh?
- If not, where, specifically, are the scheduler objects stored so they can be copied?
Any insight would be appreciated.

Thanks for the caution. After we performed the export I removed the jobs owned by that schema user and was able to rebuild using the DDL. I did notice that when the chain rules were created, the 'condition' syntax was converted. For example, instead of my first rule having a condition of 'TRUE', Oracle converted it to 1=1. Instead of 'STEP_1 Completed and STEP_2 Completed' it converted to
':STEP_1.STATE = 'COMPLETED' , etc. I don't think this should really be an issue.
However, the real kicker is that the chain rule 'action' came back null. This includes removing actions such as 'START "STEP_3" ', and even my 'END' action. Is that to what you are referring? Not sure if this is a bug and/or if there is a workaround?

Full database refresh

Hello Gurus,
I am need to take make full database refresh from one database to other, Can someone let me know step how to do it?
select * from V$VERSIONBANNER
Oracle Database 10g Enterprise Edition Release 10.2.0.5.0 - 64bi
PL/SQL Release 10.2.0.5.0 - Production
CORE     10.2.0.5.0     Production
TNS for HPUX: Version 10.2.0.5.0 - Production
NLSRTL Version 10.2.0.5.0 - Production

CKPT wrote:
jgarry wrote:
I'm with Ed and Brendan on this. 3 main reasons:
1. Legacy has some special requirements.
2. Definitely want different schema names, this has saved my bacon sooo many times. Even so, people still get it wrong (including me, and I'm totally paranoid about it to begin with).
3. There are many more copies of production than actual production, for various reasons (test, several dev, special projects, future app versions...) . We would rather have one db with many schemata than many db's, right?
Now if we could just rename user...Agreed some extent, I can say to refresh entire database the preferred method is RMAN restore, where it can manage very easily
how?
1) To import again we need to create a new database
2) Need to create all the tablespaces with same names or we can use REMAP option
3) need to monitor entire import process,
4) need to take care of other things like , temp utilization and all.
5) need to perform export, Need to copy those dumpfiles, We should have enough space on mount points of both source & destinations. ;-)
I would just caution there that you shouldn't think that because you are taking rman backups you have no need to also regularly export the db. A good backup policy includes both physical backup (rman) and logical backup of the data (export).
If you talk with couple of schemas the preferred method is EXPDP/IMPDP, totally agreed.
If in case of database size is huge, then better to go with RMAN to refresh database. Either DUPLICATE or manual way restore.
as ED said, still we could have chance to test RMAN backups too. Multiple advantages.
Let me know if I'm wrong. :)

GoldenGate for DB

Hi experts,
we have a replication requirement between oracle to DB2 database, please help on this, can we use data transformation like mapping and filtering in goldengate between oracle to DB2 databases? and the actual requirment is, there is 5 tables in source oracle database and some of the fields values from that 5 tables needs to replicat to single DB2 table , is it possible in goldengate? kindly help on this.
thanks in Advance.

902923 wrote:
Thanks a lot Mike, Can you please confirm the below points that would be great help to start,
1. Is it possible to do the bi directional replication between oracle to mainframe DB2 databases with filtering and mapping? is there any additional steps required for this?
2. For Bi directional replication between oracle to mainframe DB2 , need to create definition file for both oracle and DB2 databases and transfer to each side is it right?
3. Also , we can find profile check script for db2 database for linux, unix. and aix in MOS but can't execute the same script on mainframe environment , is there any separate script available for mainframe DB2 environment?
Thanks in advance.
Yes, bi-directional is possible in a heterogeneous replication scenario, with filtering, mapping, etc; see the topic in the docs on "bidirectional replication".
The main point to consider is avoiding "loops", so that replicated data isn't re-replicated. The means by which each DB handles this is different; in some cases, data replicated by GG is automatically detected (e.g., via a trace table) or by ignoring changes applied by a designated user that just does replication (i.e., "excludeUser"). See the GoldenGate guide section on "preventing data looping".
Data definition files would have to be created on each side, yes, and copied to the "other" side.
For scripts used for checking the database before using GG, you probably should check with Support for the latest; I can't say for sure. For DB2 on z/OS (which is usually what "DB2 on mainframe means), you'd use Unix System Services to run GoldenGate, which provides the same ksh/bash shell environment as other linux/unix (aix, et al) environments for executing shell scripts. (But do check with support for your options here.)

GoldenGate for Standard Edition One

Hi,
I'm considering implementing GoldenGate for bidirectional active-active replication between 2 Oracle Standard Edition databases.
I couldn't find any reference about the deference between GG on Enterprise Edition and Standard Edition.
So I would like to know what are the restrictions for SE1?
Can GG work as bidirectional active active replication include DDL on SE1?
Thanks, Ofir.

Hello,
I can not find documentation on licensing restrictions on Streams (SE/SE1) you speak.
Only this is documented in the manual license (https://docs.oracle.com/database/121/DBLIC/toc.htm) this restriction:
With SE / SE1 capture is async , that is you could not perform real-time capture.
But not indicate any restriction more.
In this case if you want to use real-time capture, GG would have to be used in classic mode.
Any way, A Golden Gate licence you need.
Arturo

Risks associated with database refresh.

Hi experts,
I need to clarify a doubt .I have a cloned SAP BW system (from another system i.e SAn to SAN copy)and then after few months after the first clone I need to refresh the database(Oracle 10 G).The reason being I will have some changes in data at the source system from where it was copied earlier.I shall be applying the application changes by tracking the transport requests(from the original system).
I need to clarify if we don't refresh the whole system again by SAN to SAN copy and then do a database refresh will there be any problem .Is there any risk associated with it.
Will the requests be available to me on the monitor with the new data or there will be some issues.
A quick response shall be very useful.
Shailja.

Hello,
Orders for Production , Maintenance, Deliveries, Transfer Order etc. are created according to the factory calendar .
You must be sure that all operations can be carried out in the seven days of the week (e.g. deliveries), because SAP can be scheduling aTransfer Order for a Sunday and somebody can be calling you for fixing the issue!
The normal case is that some operations can take place in the weekend (Production, Maintenance) but other Logistic Operations ( Deliveries) are done only from Mo to Fr. If this is your case, I would create an additional seven days calendar and assign it to the work centers in Production and maintenance, but the Plant would remain linked to the old five days calendar
Hope this can solve your issue.
Dario

Database Refresh From ASM Filesystem to Local Filesystem

Hi ALL,
I am performing a database refresh from production server to a demo server. Our Production database is 11.2.0.1 and it is using ASM filesystem to keep the data, redo and other files in ASM disks.
On the other hand demo server is not having ASM, all the database files are stored in a local filesytem.
I have taken a fresh backup of our production database, but I am not sure to perform the restore part as the demo server is not having ASM.
Can anyone suggest me how to perform this, I mean datafile restore from ASM to local filesystem.
Any usefull links will be helpful.
Regards,
Arijit

Hello,
You can restore the backup of your Production database which is using ASM to your demo server (using file system).
Make sure that the control_files parameter in the pfile/spfile is pointing to the file system location where you want to have contol files located on the demo server.
Next, before you use the restore command to restore the database, provide the location of the datafile where you need to restore using the "set newname" clause.
run
set newname for datafile 1 to '<file-system-location-on-demo-server>';
set newname for datafile 2 to '<file-system-location-on-demo-server>';
restore database;
switch datafile all;
recover database;
}

Database refresh in different version

Oracle 11.2.0.1 and 11.2.0.3.
Can rman refresh database from 11.2.0.1 to 11.2.0.3?
In another words, my source database is 11.2.0.1, my target database is 11.2.0.3
I want to refresh target database from source database. Is it ok since version is different?
Appreciate any help.
S.

Hello;
No is the short answer. To duplicate or restore the databases need to be the same version.
There is an upgrade method available, but I doubt you want to do that each time.
DUPLICATE considers 32-bit and 64-bit versions of the same operating system as belonging to the same platform.
See :
Master Note For Oracle Recovery Manager (RMAN) [ID 1116484.1]
RMAN Restore of Backups as Part of a Database Upgrade [ID 790559.1]
Upgrading Database Version Online Using Streams and RMAN [ID 782444.1]
Restoring a database to a higher patchset [ID 558408.1]
Different Upgrade Methods For Upgrading Your Database [ID 419550.1]
For additional information.
Best Regards
mseberg
Edited by: mseberg on Aug 24, 2012 12:31 PM

Database Refresh Status log (Info)

Dear All,
Is there thing that can help to find out completion status for database refresh.
Application log like hyplive.log located in the following location
Oracle\Middleware\user_projects\epmsystem1\diagnostics\logs\essbase\essbase_0\app\HypLive
the log file says this
"[Wed Apr 04 14:53:50 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1013091)
Received Command [Restructure] from user [admin@Native Directory]
[Wed Apr 04 14:53:50 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1019017)
Reading Parameters For Database [Drxxxxxx]
[Wed Apr 04 14:53:50 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1019012)
Reading Outline For Database [Drxxxxxx]
[Wed Apr 04 14:53:50 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1019024)
Reading Outline Transaction For Database [Drxxxxxx]
[Wed Apr 04 14:53:50 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1007043)
Declared Dimension Sizes = [1408 4 82 2 2 3 10 14 16 17 33 66 183 193 265 852 922 920 2 1 4 3 3 3 6 ]
[Wed Apr 04 14:53:50 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1007042)
Actual Dimension Sizes = [1395 4 81 1 1 3 10 12 16 8 26 64 145 163 245 848 898 914 0 0 0 0 0 0 5 ]
[Wed Apr 04 14:53:50 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1007125)
The number of Dynamic Calc Non-Store Members = [191 0 54 0 0 0 0 2 0 9 6 2 34 20 19 4 0 5 ]
[Wed Apr 04 14:53:50 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1007126)
The number of Dynamic Calc Store Members = [0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ]
[Wed Apr 04 14:53:50 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1007127)
The logical block size is [112995]
[Wed Apr 04 14:53:50 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1010008)
Maximum Declared Blocks is [26928283342271066000000000] with data block size of [115456]
[Wed Apr 04 14:53:50 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1010007)
Maximum Actual Possible Blocks is [5868290301232498000000000] with data block size of [32508]
[Wed Apr 04 14:53:50 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1200481)
Formula for member [2050103] will be executed in [CELL] mode
[Wed Apr 04 14:53:51 2012]Local/HypLive/Plan1/admin@Native Directory/Info(1007067)
Total Restructure Elapsed Time : [0.235] seconds
[Wed Apr 04 14:53:52 2012]Local/HypLive/Plan2/admin@Native Directory/Info(1007067)
Total Restructure Elapsed Time : [0.266] seconds
but i am looking for the status of database refresh failed or succeeded?
Any suggestions.
Regards
Ace

If it failed then you would of got an error generated in the essbase application log and the planning log.
Cheers
John
http://john-goodwin.blogspot.com/

Database refresh from standby database

All,
OS : Sun version : 11.1.0.7
I'm doing database refresh using stand by database datafiles in different server.
1)Disabled Data guard broker in standby database .
2)Copied all datafiles in mount stage to target server (different server)
3)created controlfile and started the database in mount stage in different server
4)Applied all archive log files .
5)But still the database not recovered .
SQL> recover database using backup controlfile until cancel;
ORA-00279: change 10401147296308 generated at 01/13/2012 09:03:43 needed for
thread 1
ORA-00289: suggestion :
/u01/app/oracle/admin/qecgdeva/arch/qecgdeva_727091849_1_57879.arc
ORA-00280: change 10401147296308 for thread 1 is in sequence #57879
Specify log: {<RET>=suggested | filename | AUTO | CANCEL}
/u01/app/oracle/admin/qecgdeva/arch/qecgdeva_727091849_1_57879.arc
Specify log: {<RET>=suggested | filename | AUTO | CANCEL}
/u01/app/oracle/admin/qecgdeva/arch/qecgdeva_727091849_1_57885.arc
ORA-00279: change 10401156453328 generated at 01/13/2012 10:48:01 needed for
thread 1
ORA-00289: suggestion :
/u01/app/oracle/admin/qecgdeva/arch/qecgdeva_727091849_1_57886.arc
ORA-00280: change 10401156453328 for thread 1 is in sequence #57886
ORA-00278: log file
'/u01/app/oracle/admin/qecgdeva/arch/qecgdeva_727091849_1_57885.arc' no longer
needed for this recovery
Specify log: {<RET>=suggested | filename | AUTO | CANCEL}
CANCEL
ORA-01547: warning: RECOVER succeeded but OPEN RESETLOGS would get error below
ORA-01196: file 1 is inconsistent due to a failed media recovery session
ORA-01110: data file 1:
'/mounts/qecgdeva_data/oradata/qecgdeva/dbfiles/qecgprod_system.dbf'
ORA-01112: media recovery not started
SQL> alter database open resetlogs;
alter database open resetlogs
ERROR at line 1:
ORA-01196: file 1 is inconsistent due to a failed media recovery session
ORA-01110: data file 1:
'/mounts/qecgdeva_data/oradata/qecgdeva/dbfiles/qecgprod_system.dbf'Am I missing any steps here ? Please advice me on this .
Thanks.

Hi,
When you copy files of standby server to another target server ? - Yesterday.
Is standby is SYNC with primary? or MRP is still running? - ---------yes, standby db is running fine .
Even you can open database with resetlogs after recreating controlfile, no need of recovery too -- >I was unable to open . So went to apply arch files.
Do you have archive of thread 1 is in sequence #57887, -- No , Because sequence 57887 yet generate. I have applied all recent available archive log files.
ORA-00289: suggestion :
/u01/app/oracle/admin/qecgdeva/arch/qecgdeva_727091849_1_57886.arc
ORA-00280: change 10401156453328 for thread 1 is in sequence #57886
ORA-00278: log file
'/u01/app/oracle/admin/qecgdeva/arch/qecgdeva_727091849_1_57885.arc' no longer
needed for this recovery
Specify log: {<RET>=suggested | filename | AUTO | CANCEL}
CANCEL
ORA-01547: warning: RECOVER succeeded but OPEN RESETLOGS would get error below
ORA-01196: file 1 is inconsistent due to a failed media recovery session
ORA-01110: data file 1:
'/mounts/qecgdeva_data/oradata/qecgdeva/dbfiles/qecgprod_system.dbf'
ORA-01112: media recovery not started
SQL> alter database open resetlogs;
alter database open resetlogs
ERROR at line 1:
ORA-01196: file 1 is inconsistent due to a failed media recovery session
ORA-01110: data file 1:
'/mounts/qecgdeva_data/oradata/qecgdeva/dbfiles/qecgprod_system.dbf'Edited by: 805877 on Jan 12, 2012 10:43 PM

Database refresh Steps

Hi,
Can anyone give the procedure for DB Refresh pls?
I couldnot find a clear explanation anywhere.
Bipin

Sory Werner if im wrong, Does the following forum explain about db refresh or they are just talking about cloning...
Database Refresh
I thought Refreshing the test db with production is termed as "db refresh"..
Is it so? Can u pls explain?
Bipin

Database Refresh Query

Hi,
A Very Happy New Year to all !!
We want to refresh our Test system database with Production data.
ECC 6.0 ABAP + JAVA system.
We will be using Backup-Restore method (BRTOOLS) for the same.
However, the problem is, our Production schema for ABAP and JAVA are SAPMI3 and SAPMI3DB resp. and that for Test are SAPMI2 and SAPMI2DB.
After the database refresh, the schema of Test server will be changed to that of Producion.
We can make the necessary changes for ABAP schema (ops$), however, would like to know, what changes (If any) are to be made for Java schema.
Request your suggetions.
Regards,
Kiran Pol

Hi Kiran
The JAVA schema can be set in the configtool -> secure store -> jdbc/pool/<SID>/User
However i am not fully sure, if you have to change it, i think the information is stored in the database already, so it will be already fine when you do a backup/restore.
Do the test and prod system have the same instance numbers? Otherwise you will have the additional struggle with the ports. For example the dispatchers http/https ports.
Don't forget to change the sidadm env variable dbs_ora_schema as well.
Best regards, Michael

Database refresh on quality system.

Dear Experts,
We need to refresh PRODUCTION system ERP 6.0 EHP5 to existing QUALITY system. We have gone through below thread https://forums.sdn.sap.com/thread.jspa?forumID=186&threadID=213898 but bit confused about procedure. Is it ok to go with this thread?
Remote client copy is not feasible in our case also it will take days to complete
Kindly let us know steps/ procedure for refreshing database.
Your help would be appreciated.
Thanks and Regards,
Pramod Jadhav

Dear Volker,
Thanks for replying.
As we have in worst situation. For database refresh we have deleted existing R3SIDDATA library. Is there any other option except SAPinst to refresh the database PRD to QAS please let us know.
Best regards,
Pramod Jadhav

SRM 7.0 Database Refresh

Hi everyone,
We are going to be performing a database refresh from our production system down to our development system. We would like to maintain the current org structure that is in development though. Does anyone know what tables we would need to export / import during the refresh to keep the org structure the same? Also, do we need to worry about the product category replications?
If anyone has had any experience with this your advice would be very appreciated.
Thanks,
Jeff

Hi Nikhil,
Ok thank you. Do you know anything about maintaining the same org structure that was in place already in our Dev system? It is different than our production system as we do not have all of the users setup in Dev. Also we have some test ID's in Dev that we would like to keep. Do you know how to maintain the same org structure?
Thanks,
Jeff

GoldenGate for Dev/QA Database Refresh

Similar Messages

Maybe you are looking for