Cascading standby database on 9i

Hi,
PrimaryDB -> Archivelog Repository -> Physical Standby.
Archivelog repository is just holding redologs and pushing them to standby. Is this feature available in 9iR2?. I realize that it is available in 10G.
Thanks
Rajesh

Yes, check out the section "Cascaded Redo Log Destinations" in Oracle Data Guard Concepts and Administration in the 9iR2 docs.

Similar Messages

Cascaded standby database

Hi,
I've seen a lot of documentation about cascaded standby setup, but none have defined how to build the second standby. The link below only describes how to configure the second standby for cascaded setup. Can anyone please elavorate how to create the second standby? Should we just build it just like the way we build the first standby, and then configure it as cascaded? OR should we build it using the first standby?
http://emrebaransel.blogspot.com/2010/08/dataguard-cascade-standby-database.html
Primary -> First standbyu -> Second Standby
Edited by: user13489456 on May 28, 2013 4:07 AM

user13489456 wrote:
Hi,
I've seen a lot of documentation about cascaded standby setup, but none have defined how to build the second standby. The link below only describes how to configure the second standby for cascaded setup. Can anyone please elavorate how to create the second standby? Should we just build it just like the way we build the first standby, and then configure it as cascaded? OR should we build it using the first standby?
http://emrebaransel.blogspot.com/2010/08/dataguard-cascade-standby-database.html
Primary -> First standbyu -> Second Standby
Edited by: user13489456 on May 28, 2013 4:07 AMWhat is the version you are using?
Creating another standby is same as how you create first standby database i.e RMAN duplicate/RMAN restore/HOT backup/Cold backup method , Here only configuration changes to configure redo transport and other parameters.
Here you can use either backup of primary or standby, it doesnt make any difference.

Cascading standby databases

Hi All,
Can anyone please throw some light on cascading standby databases, I needed to implement it, but dont have an idea as to how to go ahead with it. All I could find was the following metalink Id: 409013.1. Any help from you guys would be really appreciated.
Thanks,
Ajinkya

Hi Santosh,
The previous thread was abt one of the issue which I faced and I thought of uodating all about the reason why it happened. This thread was created so that I can gather more info abt cascading standby db, thought both of them are absolutely diff topics to discuss on,hence 2 diff threadsThat's Ok, Ajinkya. My point was, you're facing problem while performing log switch, people were co-operating you and you suddenly concluded that thread without achieving any result. Your thread may be beneficial for others.
My protection mode is maximium performance. ***************************************************************************************
Maximum Performance
This protection mode provides the highest level of data protection that is possible without affecting the performance of a primary database. This is accomplished by allowing transactions to commit as soon as all redo data generated by those transactions has been written to the online log. Redo data is also written to one or more standby databases, but this is done asynchronously with respect to transaction commitment, so primary database performance is unaffected by delays in writing redo data to the standby database(s).
This protection mode offers slightly less data protection than maximum availability mode and has minimal impact on primary database performance.
Reference: http://download.oracle.com/docs/cd/B28359_01/server.111/b28294/protection.htm#CHDDAHJD
N.B.: the above statement is assuming that you're using standby redo logs. If not using SRL then local archiving will happen first and then that archive will be transferred to standby database. Please observe the functioning in the following diagram carefully.
http://download.oracle.com/docs/cd/B19306_01/server.102/b14239/log_transport.htm#i1265592
Now ,the database following the diagram while transferring the redo data, will slow down the whole system not only the database. Here is the problem, you need to investigate at this point. Are you using any anti-virus which performs blockwise check while write/read? This may result in slow performance. Can you afford disabling your anti-virus for some time and check? :) .
And the general idea of going for cascaded standby was that my primary faces performance issue due to the network latency between the production and DR servers. So instead of my primary hampers till the archives are sent to the remote db, y cant I send those archives to a standby db on LAN, and from there I can send it to the remote server. Example:
A -- My primary
B -- My standby on LAN
C -- My standby on remote destiantion.
So now instead of A sending archives to C, it would send archives first to B which in turn will send it to C. In this way my dependency of A on C reduces which in result doesnt hamper my performance of A. and then B in turn would send the archives to C.
Regards,
AjinkyaThat seems ok to me. the above link is enough for setting up a cascaded setup (It's nothing but cascading the archive destination).
Regards,
S.K.

More than One Physical Standby Database

Hi All!
Is it possible to have more than one physical standby database? Does the procedure on how to create a single physical standby is the same as creating more than one? How about the configuration of the Primary database? Im planning this because I want the other physical standby to be used for read only access for our testing environment.
Please enlighten me.
Regards,
cmadiam

Think about it.
Q. What's the crux of Dataguard setup?
A. Redo transfer from primary to standby
Q.Which is the parameter you use to direct the redo transfer
A. LOG_ARCHIVE_DEST_n
Q. How many LOG_ARCHIVE_DEST_n parameters can be set?
A. 10; LOG_ARCHIVE_DEST_1 to LOG_ARCHIVE_DEST_10
Of the 10 LOG_ARCHIVE_DEST_n, one of them should be LOCAL destination, which means you can have 9 other REMOTE destinations ~~ 9 standby database. Physical or logical or mix of both.
Cheers.
Note: Each of the 9 standby database can cascade redo transfer to 9 other cascaded standby databases. WOW! now you do the math.

Chained Standby database

Hi All,
I want to establish DRs in chain.
A --> B --> C
Where
A= Production database
B= Standby Database for A which is fetching archives from A.It's geographically near to A.
C= Standby Database for A which is fetching archives from B.It's geographically far from A.
So basically I want to create near DR and a far DR. Is it possible? How to accomplish this?
Thanks in advance

Yes, it is possible. The configuration is also known as Cascaded standby databases.
You didn't mention your database version.
If it is 10.2.x you can take a look at Oracle® Data Guard Concepts and Administration 10g Release 2 (10.2) - E Cascaded Destinations:
http://download.oracle.com/docs/cd/B19306_01/server.102/b14239/cascade_appx.htm#SBYDB01800
Also, you can find useful information in Metalink Note:409013.1 - Cascaded Standby Databases

Switchover scenario in a cascaded standby setup

Hi,
I have a setup of primary and standby databases in three locations A, B, C as below:
Database version: 11.1.7.0
In location A I have the primary database that ships redo log to a physical standby at location B alone.
In location B I have a physical standby that replicates to site C.
In site C I have the final physical standby that receives its redo logs directly from site B. In essence, this is a cascaded standby database.
My question is about the switchover:
Currently, both the instances in sites A and B have the LOG_ARCHIVE_DEST_3 set to point to site C and this is valid for standby role alone.
Which means in case the two sites - A & B reverse roles, site C will continue receiving redo logs from either A or B, depending on which one is the new standby.
However, the doubt comes in if site C were to become the new primary instance. I suspect that I would need to add two parameters LOG_ARCHIVE_DEST_2 and LOG_ARCHIVE_DEST_3 and point them to sites A and B respectively, and make them valid only for the primary role. In this case, if site C becomes the new primary, then both sites A and B can continue receiving redo logs from site C.
Anyone who has ever implemented this?
Regards,
dula

The scenario in the doc. may also help you:
http://download.oracle.com/docs/cd/B28359_01/server.111/b28294/cascade_appx.htm#i638620
Example E-1 Sample Use of Initialization Parameters in Cascaded Destinations
Boston Database (Primary Role)
DB_UNIQUE_NAME=boston
REMOTE_LOGIN_PASSWORDFILE=EXCLUSIVE
LOG_ARCHIVE_CONFIG='DG_CONFIG=(chicago,boston,denver)'
LOG_ARCHIVE_DEST_1='LOCATION=/arch1/boston/ VALID_FOR=(ALL_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=boston'
LOG_ARCHIVE_DEST_2= 'SERVICE=denver VALID_FOR=(STANDBY_LOGFILES,STANDBY_ROLE) DB_UNIQUE_NAME=denver'
LOG_ARCHIVE_DEST_3= 'SERVICE=chicago VALID_FOR= (ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=chicago'
Chicago Database (Standby Role)
DB_UNIQUE_NAME=chicago
LOG_ARCHIVE_CONFIG= 'DG_CONFIG=(chicago,boston,denver)'
REMOTE_LOGIN_PASSWORDFILE=EXCLUSIVE
LOG_ARCHIVE_DEST_1= 'LOCATION=/arch1/chicago/ VALID_FOR=(ALL_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=chicago'
LOG_ARCHIVE_DEST_2= 'SERVICE=denver VALID_FOR=(STANDBY_LOGFILES,STANDBY_ROLE) DB_UNIQUE_NAME=denver'
LOG_ARCHIVE_DEST_3= 'SERVICE=boston VALID_FOR= (ONLINE_LOGFILES,PRIMARY_ROLE) DB_UNIQUE_NAME=boston'
Denver Database (Standby Role)
DB_UNIQUE_NAME=denver
LOG_ARCHIVE_CONFIG= 'DG_CONFIG=(chicago,boston,denver)'
REMOTE_LOGIN_PASSWORDFILE=EXCLUSIVE
LOG_ARCHIVE_DEST_1= 'LOCATION=/arch1/denver/ VALID_FOR=(ONLINE_LOGFILES,ALL_ROLES) DB_UNIQUE_NAME=denver'
LOG_ARCHIVE_DEST_2= 'LOCATION=/arch2/denver/ VALID_FOR=(STANDBY_LOGFILES,STANDBY_ROLE) DB_UNIQUE_NAME=denver'
Both the Boston primary database and the Chicago physical standby database define the LOG_ARCHIVE_DEST_2 initialization parameter as SERVICE=denver VALID_FOR=(STANDBY_LOGFILES, STANDBY_ROLE). Hence, even if the Boston and Chicago databases switch roles, the redo data will continue to be forwarded to the Denver database. Remember, as part of the original setup of the physical standby database, you should define a local destination, VALID_FOR=(ALL_LOGFILES, PRIMARY_ROLE), that will be used for local archiving when the physical standby database transitions to the primary role.

Create Multilevel Standby database

HI All,
We have a 10.2.0.2 primary database with a physical standby database.
We would like to create 2 more physical standby databases for the same primary but, we dont want to have the logs shipping from primary database as it is prod.
Is there any way where we could ship the logs from 1st standby to 2nd to 3rd standby.
To me more precise we want to create multilevel standby databases primary--->stby1---->stby2---->stby3.
Thanks,
Anil.

What you describe is referred to as "cascading standby databases"
Let me point you to this thread where I already answered a similar question:
cascade dataguard
Bottom line: You don' need cascading standby - there is as good as no measurable performance impact on the primary with multiple standby databases compared to only one standby.
The only valid reason (it was not performance) for cascading standbys vanished with 11gR2
Kind regards
Uwe Hesse
http://uhesse.wordpress.com

How standby database works

Hi!!
I have created standby database (testcat) for primary database (test) using Duplicate Command of RMAN on same server (windows XP) on Oracle 11.2 . I know how to create standby database But i don't know how and when standby database works. I checked init.ora file of standby database it contains all parameters prefixed with *. as follow
*.db_create_file_dest='F:\app\Administrator\oradata'
*.db_domain=''
*.db_recovery_file_dest='F:\app\Administrator\flash_recovery_area'
*.db_recovery_file_dest_size=10737418240
*.db_unique_name='testcat'
*.DB_NAME='test'
*.control_files="D:\APP\ORACLE\ORADATA\TESTCAT\CONTROLFILE\CONTROL_STAND1.CTL‘'
While init.ora file of primary database contains entries like
db_create_file_dest=F:\app\Administrator\oradata
db_recovery_file_dest=F:\app\Administrator\flash_recovery_area
db_recovery_file_dest_size=4039114752
control_files="F:\APP\ADMINISTRATOR\ORADATA\TEST\CONTROLFILE\O1_MF_8TXW6QN0_.CTL"
So I want to know is
          1. IS there any difference between parameter that start with .* and start without .*
          2. Can i set db_create_file_dest parameter in init.ora of standby database to another location.
           3. How redo log will be applied to standby database when I run ALTER DATABASE RECOVER ..... command
          4. where archived log files of standby database will stored?

1. Fetch Archive Log (FAL) Client :- It requests for archived redo log files from the primary site.
That means FAL Client is required (should exist) on standby database server. Am i right here?
FAL Client is Standby database, FAL Client requre archived log from FAL Server (primary, Cascading standby database ) for Gap Resolution.
2. Fetch Archive Log (FAL) Server:- It requests for archive redo logs from FAL Client (Standby database).That means FAL Server is required (should exist) on primary database server.
If yes then, why it is required because FAL Server request for archived redo logs from standby database while on other side standby database can not have redo logs as it runs in READ ONLY Access mode (So no DML and DDL) ?.
Also primary database will be always one step ahead in terms of archive logs than standby database and hence there is no need to request for archived logs from stanby. Am i right here?
Yes you are right. Primary is newer require archived log from standby. You asked and answered.
Regards
Mahir M. Quluzade

Recover standby database after primary failed

Hi,
I'm having 11g setup with 2 standby databases. My scenario is i'm doing failover on one standby[new primary] and converting old primary as a standby. Question is what is the status of another standby? have to create new standby or can recover using flash option?
regards,
jp

'11g' is not a version, it is a marketing label. You need to post your version in the format <x>.<x>.<x>.<x>
Yes, I know this is asked much.
Also your question sadly lacks on details, as in this version of Oracle you can cascade standby databases (standby 1 can cascade to standby 2)
This begs the simple question:
Did you try?
If so, what happened?
If you didn't try, why didn't you try? What can happen?
Sybrand Bakker
Senior Oracle DBA

[Standby Database] Advantage and Disadvantage of using stanby redolog

Hi fellow DBA. I'd like to ask the disadvantage and advantage of using standby redo logs. I hope you could help me out. Thanks

Advantages
1. if your Standby needs to run on Maximum Availability or Protection modes you have to use the Standby redo logs
2. Redo data writen on the primary will be written to standby at the same time. therefore if you primary fails you have a fall back option without or minimum data lost.
(LGWR with Sync option on Maximum Protection will have Zero Data lost.)
3. It uses LGWR than ARCH process on the standby.
4. If you are using Cascaded Destinations (cascade Standby Databases - One standby sends the Redo to another Standby ) Then you have to use Standby redo logs.
5. Standby redo logs can create even after the standby database creation.
Disadvantages
1. Your Primary Has to Write data into two different Redo logs. Hence there will be a delay on the transaction response time.
(But with Maximum Availability and LGWR with aSync option will not wait the transactions to be commited on standby database) .
Refer the note 219344.1 Usage, Benefits and Limitations of Standby Redo Logs (SRL)
Hope you will have a fair idea about this

Cascading standby

I was following the note 409013.1 to set the Cascading standby. (11.2.0.3 on Linux)
Primary - Boston Standby - Chicago Cascading standby - Denver.
I see that Primary(Boston) is directly transferring the Archive log to Denver ( C-standby). (LOG_ARCHIVE_DEST_2).
Ideally, it should go from the Chicago to Denver. We have a value for (LOG_ARCHIVE_DEST_3) for this as well on Chicago.
I tried the same setup, by deferring the LOG_ARCHIVE_DEST_2, but chicago never casacdes the standby to Denver.
If i stop Denver and restart, it pulls and applies all the pending archives, but never applies the fresh ones.
When i ENABLE dest 3 on primary, it ships directly to denver.
There is no error in any of the alert log,V$MANAGED_STANDBY,v$dataguard_status etc.

Hi,
How long did you wait after deferring dest_2?
As per the note,
A cascaded destination will therefore always have a greater redo transport lag, with respect to the primary database, than the cascading standby database.
Check below.
Redo Transport Services
Thank you!!

Real-time apply cascaded logical standby database

Hi
I have a primary database orcl
Pysical standby database orcl_std
Cascaded logical standby database orcl_tri which receives archivelogs from orcl_std
Real time apply is enabled both in orcl_std (physical standby) and orcl_tri (logical standby)
When I create a table in primary orcl, I am unable to see it on orcl_tri (Although real time apply is enabled)
However, when I switch log in primary, I can see the new table on orcl_tri.
My question is, why realtime apply is not working in my scenerio ?
orcl_std : ALTER DATABASE RECOVER MANAGED STANDBY DATABASE DISCONNECT FROM SESSION USING CURRENT LOGFILE;
orcl_tri: ALTER DATABASE START LOGICAL STANDBY APPLY IMMEDIATE;
Oracle 11.2.0.3.0

Hi mseberg,
Thanks for your reply.
There is no load or network issue as I`ve just created these databases for the experiement.
I have the same output from standby and primary databases.
SQL> select bytes/1024/1024 from v$standby_log;
BYTES/1024/1024
             10
             10
             10I can see below output in standby alertlog
Fri Nov 16 08:39:51 2012
ALTER DATABASE START LOGICAL STANDBY APPLY IMMEDIATE
ALTER DATABASE START LOGICAL STANDBY APPLY (orcl)
with optional part
IMMEDIATE
Attempt to start background Logical Standby process
Fri Nov 16 08:39:51 2012
LSP0 started with pid=37, OS id=16141
Completed: ALTER DATABASE START LOGICAL STANDBY APPLY IMMEDIATE
LOGMINER: Parameters summary for session# = 1
LOGMINER: Number of processes = 3, Transaction Chunk Size = 201
LOGMINER: Memory Size = 30M, Checkpoint interval = 150M
LOGMINER: SpillScn 1953318, ResetLogScn 995548
LOGMINER: summary for session# = 1
LOGMINER: StartScn: 0 (0x0000.00000000)
LOGMINER: EndScn: 0 (0x0000.00000000)
LOGMINER: HighConsumedScn: 1955287 (0x0000.001dd5d7)
LOGMINER: session_flag: 0x1
LOGMINER: Read buffers: 16
Fri Nov 16 08:39:55 2012
LOGMINER: session#=1 (Logical_Standby$), reader MS00 pid=30 OS id=16145 sid=49 started
Fri Nov 16 08:39:55 2012
LOGMINER: session#=1 (Logical_Standby$), builder MS01 pid=39 OS id=16149 sid=44 started
Fri Nov 16 08:39:55 2012
LOGMINER: session#=1 (Logical_Standby$), preparer MS02 pid=40 OS id=16153 sid=50 started
LOGMINER: Turning ON Log Auto Delete
LOGMINER: Begin mining logfile during commit scan for session 1 thread 1 sequence 202, +DATA/orcl_std/archivelog/2012_11_15/thread_1_seq_202.349.799450179
LOGMINER: End mining logfiles during commit scan for session 1
LOGMINER: Turning ON Log Auto Delete
LOGMINER: Begin mining logfile for session 1 thread 1 sequence 202, +DATA/orcl_std/archivelog/2012_11_15/thread_1_seq_202.349.799450179
LOGMINER: End   mining logfile for session 1 thread 1 sequence 202, +DATA/orcl_std/archivelog/2012_11_15/thread_1_seq_202.349.799450179
Fri Nov 16 08:40:04 2012
LOGSTDBY Analyzer process AS00 started with server id=0 pid=41 OS id=16162
Fri Nov 16 08:40:05 2012
LOGSTDBY Apply process AS03 started with server id=3 pid=45 OS id=16175
Fri Nov 16 08:40:05 2012
LOGSTDBY Apply process AS04 started with server id=4 pid=46 OS id=16179
Fri Nov 16 08:40:05 2012
LOGSTDBY Apply process AS01 started with server id=1 pid=42 OS id=16167
Fri Nov 16 08:40:05 2012
LOGSTDBY Apply process AS05 started with server id=5 pid=47 OS id=16183
Fri Nov 16 08:40:05 2012
LOGSTDBY Apply process AS02 started with server id=2 pid=44 OS id=16171Do you think real-time apply wasnt setup properly ?

One primary and two physical standby database creation

hi
i want to create two standby (physical) and one primary dataguard.
so please tell me what is sequence for db_file_name_convert and db_file_name_convert.
i mean how i will give database name or path in this parameter.

Hello;
One option is Data Guard Cascading Standby.
Here are my complete test notes if it help :
http://www.visi.com/~mseberg/Data_Guard_Cascading_Standby_Setup_and_Test.pdf
Best Regards
mseberg

Creating new tables in Logical Standby database

Hi
I have a requirement to create new tables in logical standby database. These tables will not be present on primary database. Is it possible to do this ?
I have a new schema already created which has the privilege to CREATE new table.
I have stopped the logical standby apply.
When I am now trying to create a new table but it is failing with error : insufficient privileges.
When I am trying to run below statement on new schema it is also failing with error of insufficient privileges.
alter session disable dataguard;
Please help.

user8819121 wrote:
Thanks Mahir,
I was able to create the table after logging in as sysdba.
But I need my user on that table to execute DML statements. My user has privileges to insert,delete and update any table.
I tried the following statements to disable the guard but it is sill not working
ALTER DATABASE GUARD STANDBY.
Do I need to skip the tables created using dbms_logstdby package to not making it part of SQL Apply ? I guess not since the table is not in primary database.
Amit
You can skip only on primary, your created schema on Standby is not in primary.
Then you must change Status of data guard to NONE. NONE is means is not any security on your data.
In Guard status NONE can change all schema data.
Please check link: http://docs.oracle.com/cd/E11882_01/server.112/e10700/manage_ls.htm#CHDGFGHG
Following tests on user created before guard status is change from ALL to STANDBY.
C:\Users\Administrator>sqlplus / as sysdba
SQL> conn test/test
Connected.
SQL> select table_name from user_tables;
TABLE_NAME
T
SQL> insert into t values(22);
insert into t values(22)
ERROR at line 1:
ORA-16224: Database Guard is enabled
SQL> conn / as sysdba
Connected.
SQL> select guard_Status from v$database;
GUARD_S
ALL
SQL> alter database guard standby;
Database altered.
SQL> conn test/test
Connected.
SQL> insert into t values(1);
insert into t values(1)
ERROR at line 1:
ORA-16224: Database Guard is enabled
SQL> conn / as sysdba
Connected.
SQL> select guard_Status from v$database;
GUARD_S
STANDBY
SQL> alter database guard none;
Database altered.
SQL> select guard_Status from v$database;
GUARD_S
NONE
SQL> conn test/test
Connected.
SQL> insert into t values(1);
1 row created.
SQL> commit;
Commit complete.And Now I want share with you new tests :)
Now user creating when after guard status change
SQL> drop user test cascade;
User dropped.
SQL> select guard_status from v$database;
GUARD_S
STANDBY
SQL> create user test identified by test;
User created.
SQL> grant create session, resource, create table to test;
Grant succeeded.
SQL> conn test/test
Connected.
SQL> create table t (n number);
Table created.
SQL> insert into t values(1);
1 row created.
SQL> commit;
Commit complete.
SQL>It means when guard status is ALL then all of user created guarding.
When you changed status to STANDBY then Logical Standby guard only primary schema and created schema before change.
NONE is not guard any schema. it means you can delete standby schema data too.
Regards
Mahir M. Quluzade
Edited by: Mahir M. Quluzade on Apr 19, 2013 4:07 PM

Logical standby database keeps displaying ORA-16768

I have a logical standby database setup on a remote host to accept cascaded logs from a physical standby. Everything was working fine until yesterday when I get the above error and log apply services stopped. When I go into Grid Control to Data Guard I see the error and it displays the failed SQL statement which is a create index statement. I have the option of clicking the SKIP button. When I click SKIP it appears to have processed it as the error goes away but then a few minutes later the error comes back, log apply stops and the failed SQL statement is the same create index statement. I guess the Skip doesn't mean Skip, it means Retry... My question is how do I get past this? Logs have been piling up.

Hi,
Ok I applied :
SQL> ALTER DATABASE STOP LOGICAL STANDBY APPLY;
Database altered.
SQL> alter database start logical standby apply immediate;
Database altered.
SQL>
And now I received this :
ORA-16825: Fast-Start Failover and other errors or warnings detected for the database
Cause: The broker has detected multiple errors or warnings for the database. At least one of the detected errors or warnings may prevent a Fast-Start Failover from occurring.
Action: Check the StatusReport monitorable property of the database specified.
What does it mean to check the status report.....
I found this about monitorable status report !!
DGMGRL> show database 'M3RPT' 'StatusReport';
STATUS REPORT
INSTANCE_NAME SEVERITY ERROR_TEXT
* WARNING ORA-16821: logical standby database dictionary not yet loaded
DGMGRL>
What can I do ?
Thanks a lot
oldschool
Edited by: oldschool on Jun 4, 2009 2:37 AM

Cascading standby database on 9i

Similar Messages

Maybe you are looking for