Data Guard - How?

Hi!
What I need to install Data Guard? I read, I need OEM Grid Control too. On download page there are 3 zips! This is more than database software! Am I right? Generally I'm looking for nice "step by step" tutorial for Data Guard.
And License question: Now, I'm using downloable software from Oracle, but later I will use CDs from my customer: 10gR2 EE. Is this version enough?
Thanx,
Jacek

And License question: Now, I'm using downloable software from Oracle, but later I will
use CDs from my customer: 10gR2 EE. Is this version enough?There is no (none, nada, zero, zilch) diffference between the downloaded version and the CD version if the version number is the same.
Long ago, Oracle separated the software from the license. Some consideration:
a) Software
- [Most] Oracle software is free;
- All Oracle software is used under specific license;
b) License
- The license defines the 'right to use' conditions of using the software;
- Some forms of license require associated payment;
Patches
- In this discussion (ie. in my definition), patches are not considered software;
- Patches require paid support;
- Paid support can only be attached to specific 'right to use' licenses (usually for-fee licenses);
And finally, Oracle reserves the right to audit you (and your customers) to make sure you have the correct license for your use.
By the way - Data Guard is part of the database and does not require anything from Grid Control. Go to this page to get help: http://www.oracle.com/technology/deploy/availability/htdocs/DataGuardOverview.html

Similar Messages

Configuring data guard: how many standby redolog do I need?

Hi all, I want to add a standby databse to my RAC test environment but I don't understanad how many redolog I need.
The main db is a 2 node RAC (10.2.0.4.0); every node has 3 log groups; every log group has 2 logfile (every log group is multiplexed on DG_DATA1/ASR/ and FLASH_RECOVERY_AREA/ASR/ )
The standby db will have all its datafile on the same filesystem, so I already configured the trasnslations parameters:
DB_FILE_NAME_CONVERT='/opt/oracle/app/oracle/oradata1/ASRSB','+DG_DATA1/ASR','/opt/oracle/app/oracle/oradata2/ASRSB','+FLASH_RECOVERY_AREA/ASR'
LOG_FILE_NAME_CONVERT='/opt/oracle/app/oracle/oradata1/ASRSB','+DG_DATA1/ASR','/opt/oracle/app/oracle/oradata2/ASRSB','+FLASH_RECOVERY_AREA/ASR'
Replication via rman works well, but I have to manually add the standby redolog on the standby db and which formula to use to calculate the number of the standby redolog needed. Which is the correct formula between :
1)
# of standby log = ( # of redolog files + 1) * # of thread = (6 +1)*2 = 14 standby redolog files
2)
# of standby log = ( # of redolog groups + 1) * # of thread = (3 +1)*2 = 8 standby redolog files
thanks,
andrea

Andrea,
I think 8 should be the answer on your question. Because you have 3 log groups and 2 threads therefore (3+1)*2 = 8.
According to the documentation :
http://download.oracle.com/docs/cd/B19306_01/server.102/b14239/create_ps.htm#SBYDB00426
Minimally, the configuration should have one more standby redo log file group than the number of online redo log file groups on the primary database. However, the recommended number of standby redo log file groups is dependent on the number of threads on the primary database. Use the following equation to determine an appropriate number of standby redo log file groups:
(maximum number of logfiles for each thread + 1) * maximum number of threads
I think you may get confused with maximum number of logfiles. But sometimes logfile is used as a synonym to log group. The actual log group files are usually called log members. (for instance MAXLOGFILES parameter specifies the maximum number of log groups and MAXLOGMEMBERS specifies the maximum number of members for each group).
Each additional member in a redo log group is just a multiplexed copy. You can also consider multiplexing the standby redo log files, the same why you do for redo log members.
In order to learn more about the usage, benefits and limitations of Standby Redo Logs (SRL) you can refer to Metalink Note 2193444.1.
I hope this helps.
--Mihajlo

Active Data Guard-How to configure Auxiliary Data Base's Listener.

HI,
I am stuck in creating listener/tnsnames for auxiliary database as the problem happens because of the auxiliary instance has just started. It is now in NOMOUNT stage. The instance registration with the listener is performed by PMON process and to start the PMON process database need to be in mount stage. So, before the instance registration by PMON with the listener there is actually noting to register and hence the instance is BLOCKED.
http://www.oracle.com/technology/deploy/availability/pdf/oracle-openworld-2009/adg_hol_2009.pdf
I am following the above document.
LSNRCTL> status
Connecting to (DESCRIPTION=(ADDRESS=(PROTOCOL=IPC)(KEY=LISTENER)))
STATUS of the LISTENER
Alias LISTENER
Version TNSLSNR for Linux: Version 11.2.0.1.0 - Production
Start Date 05-JUL-2010 11:06:23
Uptime 0 days 0 hr. 11 min. 44 sec
Trace Level off
Security ON: Local OS Authentication
SNMP OFF
Listener Parameter File /u01/app/11.2.0/grid/network/admin/listener.ora
Listener Log File /u02/app/oracle/diag/tnslsnr/oracle/listener/alert/log.xml
Listening Endpoints Summary...
(DESCRIPTION=(ADDRESS=(PROTOCOL=ipc)(KEY=LISTENER)))
(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=192.168.127.133)(PORT=1521)))
(DESCRIPTION=(ADDRESS=(PROTOCOL=tcp)(HOST=192.168.127.140)(PORT=1521)))
Services Summary...
Service "+ASM" has 1 instance(s).
Instance "+ASM1", status READY, has 1 handler(s) for this service...
Service "eagledb" has 1 instance(s).
Instance "eagledb1", status READY, has 1 handler(s) for this service...
Service "eagledbXDB" has 1 instance(s).
Instance "eagledb1", status READY, has 1 handler(s) for this service...
Service "eagledb_DGB" has 1 instance(s).
Instance "eagledb1", status READY, has 1 handler(s) for this service...
Service "nyc" has 1 instance(s).
Instance "nyc", status BLOCKED, has 1 handler(s) for this service...
The command completed successfully
I found some simillar solution on below link but it is aplicable for 10g or below.
http://arjudba.blogspot.com/2008/05/connection-to-auxilary-instance-failed.html
I am using enterprise linux with clusterware 11g rel 2 with data base 11f rel 2 on single machine.
Thanks
Anis
99826412
Muscat
Edited by: user12979506 on Jul 5, 2010 10:13 PM

The "auxiliary" is, I presume, actually a physical standby.
If so why is it in NOMOUNT? Why don't you start it?
alter database recover managed standby database cancel;
alter database open read only;

How do i find dataloss in Data Guard?

We are using redo log, in async mode, following is our setting.
SERVICE=xxx_sb max_failure=100 reopen=600 LGWR ASYNC VALID_FOR=(ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=xxx_sb
When i query V$managed_Standby for delay_mins, its always zero. Meaning there is no delay in copying a log. I have 2 questions..
1. How can i communicate to business that in worst case we will lose x Mintues of data? Its an OLTP, where the transactions are less then 2mins. Also during the night there are some batch jobs where the transactions are 60mins longs.
2. Most of the time during peak hours there is a log switch happening every 10-15mins but during non-peak it may not happen for a long period of time, is it advisable to set ARCHIVE_LAG_TIME to 10 mins? as im not using archiver , we are using log writer for standby.
any explanation or point to documentation would be appreciated,
Thanks,

Production databases who are running with fully fined configured Data Guard, do'nt have any dataloss because failover operation ensures zero data loss if dataguard is configured with maximum protection mode or maximum availability mode at failover time.
http://www.dbazone.com/docs/oracle_10gDataGuard_overview.pdf
The above pdf is oracle white paper which too confirmed it.
LGWR SYNC AFFIRM in Oracle Data Guard is used for zero data loss. How does one ensure zero data loss? Well, the redo block generated at the primary has to reach the standby across the network (that's where the SYNC part comes in - i.e. it is a synchronous network call), and then the block has to be written on disk on the standby (that's where the AFFIRM part comes in) - typically on a standby redo log.
Can you have LGWR SYNC NOAFFIRM? Yes sure. Then you will have synchronous network transport, but the only thing you are guaranteed is that the block has reached the remote standby's memory. It has not been written on to disk yet. So not really a zero data loss solution (e.g. what if the standby instance crashes before the disk I/O).
To sum up -> LGWR SYNC AFFIRM means primary transaction commits are waiting for ntk I/O + disk I/O acks. LGWR SYNC NOAFFIRM means primary transaction commits are waiting for ntk I/O only.
Source:http://www.dbasupport.com/forums/showthread.php?t=54467
HTH
Girish Sharma

How to remove data guard broker configuration when ORA-16625?

I setup data guard broker for a standby database. However, we recreate the database and re-setup the standby database. But find the earlier broker configuration still exists. But I cannot remove or disable the configuration or any database within the configuration. When I try to do so I got the error, although all network settings are correct:
Error: ORA-16625: cannot reach the database
How to remove the configuration at this stage?
Thanks for help,

Hi, I saw same issue when doing switchover testing in my lab environment.prerequisite is primary role and standby role switched and og can be applied without data guard broker.
Here is the step I resolved the issue
1)on both primary and standby database
SQL> alter system set dg_borker_start=false;
on primary DB:
SQL>alter system set dg_broker_config_file1='?/dbs/dr1afterswichoverpry.dat';
SQL>alter system set dg_broker_config_file1='?/dbs/dr2afterswichoverpry.dat';
on standby DB:
SQL>alter system set dg_broker_config_file1='?/dbs/dr1afterswichoverstby.dat';
SQL>alter system set dg_broker_config_file1='?/dbs/dr2afterswichoverstby.dat';
2) enable dg_borker_start on both primay and sandbby db
SQL> alter system set dg_borker_start=true;
3)on primary database to create configuration
Hope this can help you!
email: [email protected]

How to access Data Guard option in Oracle 10g OEM

Hi All,
I have Oracle 10g ENTERPRISE EDITION on Unix envoirement. I want to setup logical standby Database through Data Guard technology. But when I brows through OEM I do not find Link for DATA GUARD????
1) How to access Data Guard option from oracle 10g OEM.
2) Does It come with Enterprise Edition OR I have to install it separately.
Regards,
Darshan

I am managing almost everything through EM Grid Control. It is easy to setup and configure.
Anyways, if you have read the documentation on
Oracle® Data Guard Concepts and Administration and setup your environment accordingly then you should not have any problems. Atleast I can assure you that your production database will not be affected by it.
Just make sure that you follow each step and read thoroughly. If you have spare machines then you can test and record your configuration before performing it on your production db. This link will provide you with the information on DataGuard.
http://download-uk.oracle.com/docs/cd/B19306_01/server.102/b14239/concepts.htm
If you think that way then you will never be able to do any new stuff since you will always fear that something might go wrong. Go ahead pal, try it on test machines and you will know that it is not that difficult.
There are real experts who help people like us in this forum and I know that they are doing a pretty good job.
All the best.
Rgds
Adnan

How to configure Enterprise Manager in Active Data Guard?

Hi,
I have installed oracle 11g r2 rac with asm in two nodes. OS is hp-Unix.
I have installed oracle11g Active Data Guard using Single alone Database.
Both RAc and Active Data Guard works fine.
I have not configure Data Guard Broker.
Question: How to configure Enterprise Manager in Active Data Guard?
Thanks
Solaiman

http://www.gokhanatil.com/2012/04/manage-your-existing-data-guards-with.html

How to setup Data guard for SAP database?

Can someone please tell me how to setup data guard for SAP databases?
Thanks,
Abhi

Hi Abhi,
have a look at OSS 105047 - Support for Oracle functions in the SAP environment you find this under
14
Oracle Data Guard
You can use "Physical Standby".
You cannot use "Logical Standby".
You are allowed to use Fast Start Failover (FSFO) but SAP Support is not provided.
You can use Data Guard Broker.
You can use Maximum Performance Mode, Maximum Availability Mode and Maximum Protection Mode.
In the case of Maximum Availability and Maximum Protection, you must pay particular attention to a fast network connection in order to avoid performance problems.
Maximum Protection causes the primary database to terminate if problems occur in the standby database.
And here you find on Oracle white paper from 2010 http://www.oracle.com/us/solutions/sap/wp-ora4sap-dataguard11g-303811.pdf
Perhaps some SAP user have answers for you http://scn.sap.com/community/oracle/content?query=guard
regards
Kay

How to implement Data guard

hi all
I am new to data guard.I want help to implement dataguard.so can anybody tell me link for good document,video to implement dataguard.

I am new to data guard.I want help to implement dataguard.so can anybody tell me link for good document,video to implement dataguard.What is the version is it 10g or 11g?
is it Windows/Unix?
You can follow emre baransel mentioned documents and also some more here.
if 11g you can implement dataguard without any backup. However mentioning both the links. Go through.
*Creating a Standby Database on a new host [ID 374069.1]*
*How to create physical standby database with 11g RMAN DUPLICATE FROM ACTIVE DATABASE [ID 747250.1]*
*Step By Step Guide To Create Physical Standby Database Using RMAN [ID 469493.1]*

How many physical standbys can be created in one data guard environment?

Hi Experts,
How many physical standbys can be created from primary in one data guard environment? Why LOG_ARCHIVE_CONFIG limits up to 9 unique database names? Thanks.

Victor Jin-Oracle wrote:
Hi Experts,
How many physical standbys can be created from primary in one data guard environment? Why LOG_ARCHIVE_CONFIG limits up to 9 unique database names? Thanks.
Post the source of your information abouit the limit of 9 names.
Even the Oracle doc shows 1..30 names in the same provided although that doesn't necessarily indicate a limit.
https://docs.oracle.com/database/121/REFRN/refrn10237.htm

How to add two protection mode in oracle data guard

Hi all,
We have a discussion on oracle physical data guard yesterday.....
As we all know that data guard has 3 protection mode
1. Maximum Performance
2. Maximum Availability (Default)
3. Maximum Protection
Now we were discussing about is it possible to have two protection mode on same data guard simultaneously... for e.g. Maximum performance and maximum availability.
We have searched a lot but don't find any help on google.
Can any one help on this with any doc, pdf or links..
Thanks & Regards

Hi,
you don't find anything on this because it's not possible and would make no sense even if you could do it.
Maximum Availability uses SYNC commit between primary and standby db. If you combine Max. Availabilty with Max. Performance then your db clients have to wait on commit until primary and the max. availability standby have commited. That they would not have to wait on another Max. Performance DB doesn't change anything to the client. You would only combine the disadvantages of both approaches, the wait on sync commit for max. availability and beeing not in sync with primary for the max. performance db.
The max. protection mode requires by definition that the primary db shuts down in case no standby can process the commit. If the primary is not up and running it is pointless to have another (max. availability) standby db at hand, if this max. availability db would take over the service this would violate the defintion of max. protection. No Dataloss is valued higher than db availability, having the data commited only on one db site is not allowed.
Regards
Thomas

How to switch from data guard really time apply (redo) to archive log

Hi
Standby database configured with broker and applying the redo in really time; however, I want to change this to archive log apply mode without losing the broker configuration. Is it possible? If it is not possible to use broker to do archive log apply, can I remove the broker and use data guard to set up the standby to use archive log apply?
Regards

user3076922 wrote:
Hi
Standby database configured with broker and applying the redo in really time; however, I want to change this to archive log apply mode without losing the broker configuration. Is it possible? If it is not possible to use broker to do archive log apply, can I remove the broker and use data guard to set up the standby to use archive log apply?
RegardsHi
I think mseberg is answered correct, you can use enable/disable apply log with change of state on standby database with DGMGRL, as writen mseberg.
or you can disable recover standby database with following script from SQL*Plus.
SQL> alter database recover managed standby database cancel;Regards
Mahir M. Quluzade
www.mahir-quluzade.com

11.2 DB & Data Guard : ORA-16014 how to archive a sequence?

Hi,
I've installed 11.2 Oracle Database in my laptop with Oracle Enterprise Linux 5.3 and I have created two standby databases, orcl (primary) and orclstby (physical standby). I performed a switchover to orclstby and consequently, orclstby was the new primary and orcl the physical standby, I checked those values with SQLPlus executing the select database_role from v$database statement, so, there were no problems during the switchover.I also shutteddown and started both primary and stanby databases to check all was fine.
Today, I've tried to start the environment again and I've encountered the following problem during the startup of the primary database:
[oracle@mredon-es ~]$ sqlplus / as sysdba
SQL*Plus: Release 11.2.0.1.0 Production on Wed Oct 21 17:34:21 2009
Copyright (c) 1982, 2009, Oracle. All rights reserved.
Connected to an idle instance.
SQL> startup
ORACLE instance started.
Total System Global Area 839282688 bytes
Fixed Size 2217992 bytes
Variable Size 507512824 bytes
Database Buffers 322961408 bytes
Redo Buffers 6590464 bytes
Database mounted.
ORA-03113: end-of-file on communication channel
Process ID: 5562
Session ID: 9 Serial number: 3
I've checked the newest log created and this is its content:
Dump file /u01/app/oracle/product/11.2.0/dbhome_1/rdbms/log/orclstby_ora_5382.trc
*** 2009-10-21 17:34:30.467
Oracle Database 11g Enterprise Edition Release 11.2.0.1.0 - 64bit Production
With the Partitioning, OLAP, Data Mining and Real Application Testing options
ORACLE_HOME = /u01/app/oracle/product/11.2.0/dbhome_1
System name: Linux
Node name: mredon-es.us.oracle.com
Release: 2.6.18-128.el5
Version: #1 SMP Wed Jan 21 08:45:05 EST 2009
Machine: x86_64
Instance name: orclstby
Redo thread mounted by this instance: 0 <none>
Oracle process number: 0
Unix process pid: 5382, image: [email protected]
*** 2009-10-21 17:34:30.526
2009-10-21 17:34:30.457: [ default]utgdv:2:ocr loc file /etc/oracle/olr.loc cann
ot be opened. errno 2
2009-10-21 17:34:30.527: [ default]utgdv:2:ocr loc file /etc/oracle/ocr.loc cann
ot be opened. errno 2
I've searched in Metalink information about ocr.loc and olr.loc and they seem to be part of an Oracle RAC installation, I don't know why the database needs these files to start if I am using single instance...
Any idea would be kindly appreciated because I'm a bit confused and I don't really know what steps I am suppposed to take in order to solve this problem.
Thanks in advance.
Edited by: mredon on Oct 26, 2009 5:22 AM
Reason: title of the post changed

sorry for the delay, here is the alert log:
<msg time='2009-10-21T23:03:09.576+02:00' org_id='oracle' comp_id='rdbms'
client_id='' type='UNKNOWN' level='16'
host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130' module='[email protected] (TNS V1-V3)'
pid='6582'>
<txt>Data Guard: version check completed
</txt>
</msg>
<msg time='2009-10-21T23:03:09.642+02:00' org_id='oracle' comp_id='rdbms'
client_id='' type='UNKNOWN' level='16'
host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130' module=''
pid='6486'>
<txt>LGWR: STARTING ARCH PROCESSES
</txt>
</msg>
<msg time='2009-10-21T23:03:09.709+02:00' org_id='oracle' comp_id='rdbms'
msg_id='ksbrdp:3833:3697353022' type='NOTIFICATION' group='process start'
level='16' host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130'
pid='6588'>
<txt>RSM0 started with pid=23, OS id=6588
</txt>
</msg>
<msg time='2009-10-21T23:03:09.831+02:00' org_id='oracle' comp_id='rdbms'
msg_id='ksbrdp:3833:3697353022' type='NOTIFICATION' group='process start'
level='16' host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130'
pid='6590'>
<txt>ARC0 started with pid=24, OS id=6590
</txt>
</msg>
<msg time='2009-10-21T23:03:10.832+02:00' org_id='oracle' comp_id='rdbms'
client_id='' type='UNKNOWN' level='16'
host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130' module=''
pid='6486'>
<txt>ARC0: Archival started
</txt>
</msg>
<msg time='2009-10-21T23:03:10.832+02:00' org_id='oracle' comp_id='rdbms'
client_id='' type='UNKNOWN' level='16'
host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130' module=''
pid='6486'>
<txt>LGWR: STARTING ARCH PROCESSES COMPLETE
</txt>
</msg>
<msg time='2009-10-21T23:03:10.833+02:00' org_id='oracle' comp_id='rdbms'
client_id='' type='UNKNOWN' level='16'
host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130' module=''
pid='6590'>
<txt>ARC0: STARTING ARCH PROCESSES
</txt>
</msg>
<msg time='2009-10-21T23:03:11.180+02:00' org_id='oracle' comp_id='rdbms'
client_id='' type='UNKNOWN' level='16'
host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130' module='[email protected] (TNS V1-V3)'
pid='6582'>
<txt>ARCH: LGWR is scheduled to archive destination LOG_ARCHIVE_DEST_2 after log switch
</txt>
</msg>
<msg time='2009-10-21T23:03:11.183+02:00' org_id='oracle' comp_id='rdbms'
client_id='' type='UNKNOWN' level='16'
host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130' module='[email protected] (TNS V1-V3)'
pid='6582'>
<txt>ARCH: LGWR is scheduled to archive destination LOG_ARCHIVE_DEST_1 after log switch
</txt>
</msg>
<msg time='2009-10-21T23:03:11.258+02:00' org_id='oracle' comp_id='rdbms'
client_id='' type='UNKNOWN' level='16'
host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130' module='[email protected] (TNS V1-V3)'
pid='6582'>
<txt>Errors in file /u01/app/oracle/product/diag/rdbms/orclstby/orclstby/trace/orclstby_ora_6582.trc:
ORA-16014: log 1 sequence# 27 not archived, no available destinationsORA-00312: online log 1 thread 1: '/u01/app/oracle/product/oradata/orclstby/redo01.log'
</txt>
</msg>
<msg time='2009-10-21T23:03:11.263+02:00' org_id='oracle' comp_id='rdbms'
client_id='' type='UNKNOWN' level='16'
host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130' module='[email protected] (TNS V1-V3)'
pid='6582'>
<txt>USER (ospid: 6582): terminating the instance due to error 16014 </txt>
</msg>
<msg time='2009-10-21T23:03:11.429+02:00' org_id='oracle' comp_id='rdbms'
msg_id='ksbrdp:3833:3697353022' type='NOTIFICATION' group='process start'
level='16' host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130'
pid='6592'>
<txt>ARC1 started with pid=25, OS id=6592
</txt>
</msg>
<msg time='2009-10-21T23:03:12.728+02:00' org_id='oracle' comp_id='rdbms'
client_id='' type='UNKNOWN' level='16'
host_id='mredon-es.us.oracle.com' host_addr='192.168.237.130' module='[email protected] (TNS V1-V3)'
pid='6582'>
<txt>Instance terminated by USER, pid = 6582
</txt>
</msg>
I've marked in bold the errors which may be causing the problem, ¿what do you think?

How to setup a remote recovery process - Not using Data Guard?

Hello -
Has anyone (or can send me some links) set up recovery at a remote site - and not used Data Guard? For example, shipped off logs via FTP, applied the logs, and kep the DB in recovery mode?
Thanks!

Mlike,
This manual procedure is described in the Dataguard Concepts manual for 10gR2, hopefully is this your unmentioned version (BTW what happens if you do type your version? Does your boss beats you up for doing that?). I don't know why someone wants to replace something robust by something unreliable by default, using FTP. Not only unreliable, also unsecure!
Sybrand Bakker
Senior Oracle DBA

Data Guard adding new data files to a tablespace.

In the past, if you were manually updating an Oracle physical standby database there were issues with adding a data file to a tablespace. It was suggested that the data file should be created small and then the small physical file copied to the standby database. Once the small data file was in place it would be resized on the primary database then the repication would change the size on the standby.
My question is, does Data Guard take care of this automaticlly for a physical standby? I can't find any specific reference on how it handles a new datafile.

Never mind, I found the answer.
STANDBY_FILE_MANAGEMENT=auto
Set on the standby database will create the datafiles.

Data Guard - How?

Similar Messages

Maybe you are looking for