Creating a Data Guard Database with RMAN in 10&11g

I found this notes for 9i, looking for same for 10g & 11g
183570.1 Creating a Data Guard Database with RMAN (Recovery Manager) using Duplicate Command

These notes just show duplicating dbs without datagurd, i am looking for Creating a Data Guard DB with RMAN in 10 &11g.

Similar Messages

EM Job scheduling with Data Guard Database

Hi there,
I have an "Issue" on efficient scheduling Enterprise Manager (12.1.0.3) Jobs again databases which are protected with Oracle Data Guard.
What I want:
A Job should be run against the database, which has the "Primary Database Role" from Data Guard point of view.
When I'm scheduling the job with the Database System as job target, where the Primary and Standby Databases are exisitng in, the Job is executed against all databases.
I can put check-step as first step in the job to check the role by sql, but when I let 2 of 3 job executions ( 2 standby 1, primary ) fail, I will have o mange these failed jobs all time.
Creating an dynamic group also isn't helping, as the property of database role can't be used for defining the group membership.
I have taken a look at emcli but also there I dont find a possibility tho check for database role.
Any one a idea on this topic?
br
Jörg

These notes just show duplicating dbs without datagurd, i am looking for Creating a Data Guard DB with RMAN in 10 &11g.

Trying to restore a database with RMAN in new server

Hello,
Database: Oracle 10gR2
O/S: Windows Server 2003
My RMAN backuppieces are stored in a network drive:
\\old_location\FULL
and
\\old_location\INC
I'm trying to restore a whole database with RMAN in another server from my FULL backup.
For that, I copy all backuppieces files
from: \\old_location\FULL
to new server: D:\new_location\
Also I created the relevant folder hierarchy under <ORACLE_BASE> directory, as follows:
a. Create BDUMP, CDUMP, and UDUMP directories under '<ORACLE_ BASE>\admin' directory
b. Create <DATABASE_NAME> directory under '<ORACLE_BASE>\oradata' directory
c. pfile.ora and controlfile (CTL_MTDPROD_2308_20120401_1.CTL) are backup from original database and they are in D:\new_location\
d. I copied pfile.ora and controlfile to folder: C:\oracle\product\10.2.0\db_1\database\
I started creating an instance in new server from scratch:
C:\>oradim -new -sid MTDPROD -intpwd mtdprod
C:\> SET ORACLE_SID=MTDPROD
C:\> rman
RMAN> connect target sys
RMAN> SET DBID=21692876873
RMAN> STARTUP force nomount PFILE=C:\oracle\product\10.2.0\db_1\database\init_mtdprod.ora
RMAN> restore controlfile from 'C:\oracle\product\10.2.0\db_1\database\CTL_MTDPROD_2308_20120401_1.CTL';
RMAN> alter database mount;
#Since I want to read my backuppieces from D:\new_location and not from \\old_location\FULL then I catalog and restore like:
RMAN> catalog start with 'D:\new_location\';
RMAN> restore database;
Starting recover at 25/04/12
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid=541 devtype=DISK
channel ORA_DISK_1: starting incremental datafile backupset restore
channel ORA_DISK_1: specifying datafile copies to recover
recovering datafile copy fno=00003 name=D:\BD_MTDPROD\CONTROLFILES\SYSAUX01.DBF
recovering datafile copy fno=00004 name=D:\BD_MTDPROD\CONTROLFILES\USERS01.DBF
recovering datafile copy fno=00007 name=D:\BD_MTDPROD\DATAFILES_DATA\MSE.ORA
recovering datafile copy fno=00010 name=D:\BD_MTDPROD\DATAFILES_DATA\EXPEDIENTES2.ORA
channel ORA_DISK_1: reading from backup piece \\old_location\FULL\FULL_MTDPROD_20120401_2304_1.BAK
channel ORA_DISK_1: restored backup piece 1
channel ORA_DISK_1: failover to piece handle=D:\new_location\FULL\FULL_MTDPROD_20120401_2304_1.BAK tag=TAG20120401T190004
.....(continues)
Questions:
1. RMAN still reads from old_location. The only way to avoid that would be deleting all files from \\old_location\FULL ?
2. The message "failover to piece handle=D:\new_location\...." indicates that that file is not usefull? I just copy them from old_location to new_location... how can they get useless?
I suspect there is something in controlfile that redirects to \\old_location\FULL\ and doesn't let to read from D:\new_location\
Please, advice.

user521219 wrote:
Questions:
1. RMAN still reads from old_location. The only way to avoid that would be deleting all files from \\old_location\FULL ?
2. The message "failover to piece handle=D:\new_location\...." indicates that that file is not usefull? I just copy them from old_location to new_location... how can they get useless?
I suspect there is something in controlfile that redirects to \\old_location\FULL\ and doesn't let to read from D:\new_location\Is this a cold backup? Did you change the location of control file in your PFILE or is it still pointing to old location?
You may want to check the output of RESTORE DATABASE PREVIEW command, which will just report which backup files it uses without actually restoring thhe DB.

What are the differences of creating a physical standby database with ASM?

I want to create a physical standby database for my 10g(10.2.0.1) database. But it is using ASM. Is it possible to create a standby database for a database using ASM?
What are the differences of creating a physical standby database with ASM?

tell us what your research shows and we'll try to
validate it. I am simulating Oracle ASM on a PC and I want to try
DataGuard on it now. So I am trying to create DataGuard
on an ASM system. The purpose is only learning and test.That's telling us how you are researching, not what you have observed.
Your research could have included looking at the documetation, and you might have found http://download.oracle.com/docs/cd/B19306_01/server.102/b14239/scenarios.htm#sthref1642
which discusses exactly what you are looking for.
Message was edited by:
Hans Forbrich

Will Oracle ever create a Data Guard certification?

Will Oracle ever create a Data Guard Certification?
If so would it be more of a "HA" - high availability,
or would it be only about Active Data Guard.
Thanks

We have a Data Guard Admin certification coming soon. Unfortunately, I don't have a closer date than that.
https://blogs.oracle.com/certification/entry/0856_31
Regards,
Brandye Barrington
Forum Moderator

Problem creating physical Standby database with RMAN

Hi All
I am trying to learn oracle dataguard and as part of the process learning creating standby database.
Platform : Sun-Fire-V250 Sparc, Solaris 10
Database Version - Oracle 11R2
I am creating standby database on same server, so directory structure is different.
Following the instructions on Oracle site I managed to create a functional physical standby database. But I am not able to create standby database using RMAN. These are the steps that I followed-
1.Set up all necessary parameters on primary database as done while creating physical standby database manually, eg setting force logging, creating standby logs etc.
2.Edited parameter file on primary database as done while creating manual pysical standby database creation. Some of the changes done are-
On Primary Database:
*.FAL_CLIENT='orcl11020' #Primary database unique name
*.FAL_SERVER='stdby_11' #Standby database unique name
db_file_name_convert='/<dir>/oradata/stdby_11','/<dir>/oradata/orcl11020'
log_file_name_convert='/<dir>/oradata/stdby_11','/<dir>/oradata/orcl11020','/<dir>/oradata/stdby_11/redo_mem','/<dir>/oradata/orcl11020/redo_mem'
standby_file_management=auto
*.log_archive_config='DG_CONFIG=(orcl11020,stdby_11)'
*.log_archive_dest_1='LOCATION=/<dir>/flash_recovery_area/ORCL11020/archivelog
VALID_FOR=(ALL_LOGFILES,ALL_ROLES) db_unique_name=orcl11020'
*.log_archive_dest_2='SERVICE=stdby_11 LGWR ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) db_unique_name=stdby_11'
*.LOG_ARCHIVE_DEST_STATE_1='ENABLE'
*.LOG_ARCHIVE_DEST_STATE_2='ENABLE'
*.LOG_ARCHIVE_FORMAT='%t_%s_%r.arc'
*.LOG_ARCHIVE_MAX_PROCESSES=30
Copied same pfile for standby database and modified following-
*.control_files='/<dir>/oradata/stdby_11/stdby_11.ctl','/<dir>/fra_stdby/stdby_11/stdby_11.ctl'
*.db_name='orcl1102'
*.db_unique_name='stdby_11'
*.FAL_CLIENT='stdby_11'
*.FAL_SERVER='orcl11020'
db_file_name_convert='/<dir>/oradata/orcl11020','/<dir>/oradata/stdby_11'
log_file_name_convert='/<dir>/oradata/orcl11020','/<dir>/oradata/stdby_11','/<dir>/oradata/orcl11020/redo_mem','/<dir>/oradata/stdby_11/redo_mem'
standby_file_management=auto
*.log_archive_dest_1='LOCATION=/<dir>/fra_stdby/STDBY_11/archivelog
VALID_FOR=(ALL_LOGFILES,ALL_ROLES) db_unique_name=stdby_11'
*.log_archive_dest_2='SERVICE=orcl11020 LGWR ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE)
db_unique_name=orcl11020'
3. Add relevant information in tnsnames.ora and listener.ora files and then restart listener.
3. Created password file with same credential as primary database.
4.Up-to-date RMAN backup of primary database available.
5.Create standby controlfile with rman
While primary database s open (I tried with primary database in mount mode as well)-
$>rman catalog rman/paswd@rman target /
RMAN> BACKUP CURRENT CONTROLFILE FOR STANDBY;
6. Open a new terminal and startup standby database in nomount mode using parameter file created -
$>ORACLE_SID=stdby_11
$>export ORACLE_SID
$>sqlplus / as sysdba
SQL>STARTUP NOMOUNT pfile='<location/initfilename.ora'
SQL>quit
$> rman AUXILIARY / target sys/passwd@orcl11020 catalog rman/passwd@rman
RMAN>DUPLICATE TARGET DATABASE FOR STANDBY DORECOVER;
RMAN finishes without error but archive logs are not being tranported. Looking at the log, following caught my eye-
Error 1017 received logging on to the standby
Check that the primary and standby are using a password file
and remote_login_passwordfile is set to SHARED or EXCLUSIVE,
and that the SYS password is same in the password files.
returning error ORA-16191
FAL[client, ARC2]: Error 16191 connecting to orcl11020 for fetching gap sequence
Errors in file /<>dir>/diag/rdbms/stdby_11/stdby_11/trace/stdby_11_arc2_24321.trc:
ORA-16191: Primary log shipping client not logged on standby
Errors in file /<dir>/diag/rdbms/stdby_11/stdby_11/trace/stdby_11_arc2_24321.trc:
ORA-16191: Primary log shipping client not logged on standby
So on both primary and standby I confirmed
SQL> show parameter remote_login_passwordfile
NAME TYPE VALUE
remote_login_passwordfile string EXCLUSIVE
To make double sure that password files are same, I shutdown both databases, delete password files and recreated with same credentials.
Password files are called - orapworcl11020 and orapwstdby_11
Can someone guide me where thisngs are going wrong here please.

Not sure if I understood it clearly.
SELECT * FROM V$ARCHIVE_GAP;
returns no rows so there is no gap.
But could you please explain me the result of the previous query. To catch up again, on standby when I check
SELECT SEQUENCE#,APPLIED FROM V$ARCHIVED_LOG
SEQUENCE# APPLIED
75 NO
74 NO
76 NO
77 NO
I understand that though archive files have been copied across but they are not applied yet.
On primary when I give your query -
SELECT name as STANDBY,SEQUENCE#,applied, completion_time
2 FROM v$archived_log
3 where dest_id=2
4 and sequence# BETWEEN 74 and 80;
I get -
STANDBY SEQUENCE# APPLIED COMPLETIO
stdby_11 74 YES 28-JUN-11
stdby_11 75 YES 28-JUN-11
stdby_11 76 YES 29-JUN-11
stdby_11 77 YES 29-JUN-11
stdby_11 78 YES 29-JUN-11
stdby_11 79 YES 29-JUN-11
stdby_11 80 YES 29-JUN-11
stdby_11 75 NO 07-JUL-11
stdby_11 74 NO 07-JUL-11
stdby_11 76 NO 07-JUL-11
stdby_11 77 NO 07-JUL-11
stdby_11 78 NO 07-JUL-11
I have intentionally given
sequence# BETWEEN 74 and 80
because I know in the current incarnaion of the database, max sequence is 78.
So my understanding is, the rows between 28-29 June are from previous incarnation, correct me if I am wrong
Archive files of the current incarnation, since I successfully created standby database are shipped but yet to be applied - am I right?
Then my final question is, when will these archives be applied to standby database?
I am sorry to ask too many questions but I am just trying to understand how it all works.
Thanks for your help again

How to create new database with rman backup

hi,
i took a rman backup with nocatelog. can i make a new database with that rman backup.
if yes then how ? i m using oracle 9i.
thxs

'NEW' doesn't seem to be the right word here, but you can DUPLICATE (clone) an existing database or restore it to a new host:
http://download.oracle.com/docs/cd/B10501_01/server.920/a96566/toc.htm
Werner

Cloud 12.1.0.4: Creating a Data Guard Configuration asymetric

Hello,
I'd like to créate a Data Guard from Cloud Console.
The primary database is a RAC of two nodes, but the standby node is one node and I dont't like to use ASM.
The wizard , ask the target node , the Standby Instance Name, but in the step 3 show this error:
No ASM instance is being managed as an EM target on the selected standby host.
To create a standby database on the selected host, an ASM instance must be added as an EM target.
Is not posible to use from Cloud this mix of storage configuration?
Thanks
Arturo

Hi,
It is recommended to use ASM on both sides. Normally OEM Expects the Current DB structures on STBY OEM.
Please follow the bleow procedure will help you.
Data Guard – ASM primary to filesystem physical standby – using RMAN duplicate
https://pythianpang.wordpress.com/2009/07/07/data-guard-asm-primary-to-filesystem-physical-standby-using-rman-duplicate/
Regards
Krishnan

Active data Guard but with delay on applying redo logs

Hello,
A question: Active Data Guard gives the possibility of having a physical stand-by database open read-only, for instance for making 'up-to-date' reports.
Data Guard also has the possibility to apply logs with delay, for instance with a delay of some hours. Every commit on primary database is applied to standby after the delay period.
Can this delay be combined with Active Data Guard? This means that standby database is open read-only but with the delay... In case the delay is 2 hours, every report created on standby is ("up-to-date" minus 2 hours)...
Thanks for your answer,
Jan.

Hello All,
I answered my own question: indeed it is possible to activate 'real time apply' but with a delay in applying the logs. In EM-11 check the box "enable real time query" and apply. Nect go to thje tab "standby role properties" and change the value after "Apply delay" to for instance 30 minutes. Apply also.
Log on on the primary, create a table, insert some rows and do some logswitches. Log in on the standby and check for the new table. Notice that after 30 minutes the switched logs are applied, even when you change the delay period within the 30 minutes. The information about the delay seems to be in the log, since it is possible to change the delay immediately after the logswitches to 3 minutes, drop the table and do some logswitches again. The logs with the 'create table' statement in it are NOT applied after 3 minutes, but still after 30 minutes. The drop of the table however is done after 3 minutes on the standby. So the 'delay' information is in the log sent over to the Stand-by!

Problem in getting data into database with standard direct input program

HI All,
I am having problem which is not updating the records in MM01 or MM02 with standard direct input program. i have data in internal table. from that table i am trying to upload into database by using background job MRP_MATERIAL_MASTER_DATA_LOAD.
when i execute my program it is showing message job is started. then i go into sm37 and seethe job status by executing. there also i am seeing job completed succesfully.
but if i go to mm03 and find the materials are updated or created there i couldn't find the material numbers which are from internal table.
So if ny one help me it wil be great.
Thanks in Advance
Venkat N

Hi Anil,
Thanks for your answer, but i am facing problem is i have material no and denominator and Actual UOM and nominator field values in the flat file.....
by using RMDATIND direct input program with MRP_MATERIAL_UPLOAD as job name for background job while uploading data into database.
here i am not getting data in to database, but when i execute the job in sm37 it is showing that message job processing successfully completed...this is my status..
if u can help me in this it will be gr8ful..
Thanks,
Venkat N

Standby Database with RMAN on ASM

Hi,
I am using redhat Linux AS 4 and my database is 10.2.0.1. This database used ASM for storage of data files, control files, online redo logs. Flash recovery area is also configured in the ASM.
I want to create a physical standby database on a different machine by using RMAN. Can someone kindly point me to some easy to follow document? I have scoured the oracle docs, but am more confused than ever.
regards

Step By Step Guide To Create Physical Standby Database Using RMAN [ID 469493.1]
Note: the examples in this document use regular filesystems but the process is the same regardless, ie, in db_file_name_convert use your diskgroup rather than a filesystem (db_file_name_convert=+DG_SOURCE,+DG_TARGET)
another note to complement the first one:
Step By Step Guide On Creating Physical Standby Using RMAN Duplicate In ASM Filesystem For ASM Primary [ID 837102.1]
Edited by: rui_catcuddler on Sep 2, 2010 6:51 AM

Data Guard setup with Clustered primary ?

Hi,
I am sure I am not in this unique requirement of setting up a Data Guard configuration where my primary database is clustered (for local server failure) and I am setting up the Data Guard (for the site level failure).
I could do the Data Guard setup without any problem where my primary and standby are on single individual servers (not clustered) but when I do the set up with primary database as clustered and do a local failover (not using Data Guard but using the veritas cluster) the DB Broker just stops recognizing the configuration.
I did come across a metalink note (BUG 4404952 ) which says DG Broker does not support Cold Failover Cluster.
Has anyone found any work around to it?
Has anyone tried using the above configuration in 10.2.0.4? Does it work in 10.2.0.4?
The Bug is dated 2007 so I am sure I am not the first one to encounter this issue.
Would really appreciate if someone can tell me what the workaround would be or guide me to any good link which would tell me the same.
Thanks a lot in anticipation.
Manoj Datar

I read somewhere that cold failover support in 11g is also ONLY for the oracle clusterware configuration.
Its really so hard to believe that there is NO ONE who wants to have a clustered primary (for local server failures) and a Data Guard (for the Site level failure or DR) as a combination in their HA environment.
I am still in the testing phase, What else could I be looking for in the solution if I need to achieve both these objectives (ie Local Server failures and Site level failures or DR)? I already have the Local server failure taken care by using the Veritas cluster so what do I need to include in the mix to achieve the site level failure?

Creating A New MySQL Database With Coldfusion Code

Hi,
I was wondering if anyone knew how to create a whole entire
new database simply by executing a Coldfusion template (via a form
for example).
I'm pretty confident at creating a new table in an already
existing database but not sure how to go about creating an entirely
new database from pure ColdFusion code using MySQL.
I basically want to create some portal for myself to do this
to save on arduous time. It'll need to be a form because I'll have
to give it a name, username, password etc. I need the seperate
databases for my separate clients you see.
One point I should make is that it is on a shared server so I
don't have direct access to the CF Administrator.
Any advice on achieving this or highlighting the pitfalls
would be greatly appreciated.
Many thanks in advance!
Michael.

quote:
Originally posted by:
Sabaidee
if it is on a shared server, then you very likely do not have
enough permissions to create new batabases in MySQL either...
is there an option for you in your hosting control panel (or
whatever it is called) to create a new MySQL database for your
account without any involvement of server admins? if so then,
technically you may be able to create a new db programmatically.
but i doubt you can...
I know it sounds cheeky but any chance you could supply
example code for doing this with ColdFusion and MySQL? This way I
will then be able to try it rather than waiting for my host to get
back to me.
I would have tried doing it already because I use SQL quite
often with queries etc, but I just don't have the technical know
how for this specific task at hand.
Many thanks for your help!

Confidential site - Create Due Date Notification Email with Nintex

I hope you can help
I'm trying to create a workflow that will send multiple notifications dependant on a column called 'Remediation deadline'(This is a date field) The notifications are planned for 3 months
before the remediation date as an 'Action', 2 months before the remediation date as 'High' and 1 month for 'Critical'.
I have created a calculated field to put the date into the correct format using =TEXT([Remediation deadline],'dd/mm/yy') in an effort to run the workflow from this date if possible as the formats
are correct(Not sure if the format even makes a difference?)
I'm new to SharePoint which means my knowledge is very limited, also the site I'm currently working on is a confidential site, so I have no access to designer and if I choose to add a WebPart,
I have to send it over to another team to action which is highly frustrating as they are not as eager as me to reach my goal.
The problem I face is where I go from here with such limited access. The only notification process that works is based within Nintex, but from the documentation I’ve read, you need to
be able to run the 'CallWebPart' function. I also have 1200 records that can't be removed from the task list, I tried a loop process but it caused the server to fall over as our main server is in a different country so each loop had to bounce back and
forth through the process for each record!!
We have the availability of Nintex but are unable to use the 'CallWebAccess' function as this would require our server ID and password (and that won't be happening anytime soon).
I have created a column called 'Notify Date3' (And so on for the additional date notifications)which contains the following formula
=TEXT([Remediation deadline]-92,"dd/mm/yyyy") which I use to create the date around 3 months before the first remediation date so I can create a notification.
Any advise on the best way to process will be greatly received as i'm slowly pulling my hair out
Thanks in Advance
Nick

Hi,
According to your description, you might want to implement a functionality which can send notification in different date based on the value of a column.
If you can use SharePoint Designer 2010 workflow, there are workflow actions like “Pause for Duration” and “Pause until Date” which can pause the workflow for a period. Plus
with using some conditions in the logic of the workflow process, it will be able to send notification in different date.
Workflow actions
in SharePoint Designer 2010: A quick reference guide
Here is a demo about the similar requirement provided by Laura for your reference:
http://www.wonderlaura.com/Lists/Posts/Post.aspx?ID=64
If you want to use Nintex workflow, it would be more appropriate to post to its forum to get quick and confirmed answers:
http://connect.nintex.com/forums/default.aspx
Best regards
Patrick Liang
TechNet Community Support

Using OEM to monitor Data Guard database

Can someone please send me the link or docs of how to use OEM monitoring data guard? In specific, I would like to use OEM to monitor and be sure logs are applying to stand by site.
Any ideas?

Hello ,
I will extend the document on the Fast Failover feature one of these days.
What you need to do is:
For using Fast Failover, you need to configure the Data Guard observer process.
The configuration of this process can be done by selecting the Fast-Start Failover Disabled link on the Data Guard page of the database (Primary or Standby).
You will be automatically redirected to the Fast-Start Failover: Configure Page.
From this page you can do the configuration of the DG observer process, that then will enable you to activate the Fast Failover feature.
Regards
Rob
http://oemgc.wordpress.com

Creating a Data Guard Database with RMAN in 10&11g

Similar Messages

Maybe you are looking for