Physical Standby Database Lags Far Behind the Primary Database scenario

Hi,
Have anyone successfully implemented described in the Oracle Documentation following
scenario?
For me it does not work at all.
After implementing all steps current_scn of standby database remains the same and when trying to push recovery using redo it asks for non-existing archived log which I can't restore on primary database as all backups containing that archived log already gone.
The only option is to recreate standby database from backup again.
Best Regards,
Alex

I already found the reason.
Just for your information:
Recovering a Standby database from a missing archivelog
Unfortunately documentation missed that important step of standby controlfile recreation.
To my luck I was copying the whole backup to the standby server when found this link.

Similar Messages

Checking Whether Physical Standby is in Sync with the Primary

I writing script to alert dba if primary and standby database are not in sync. I use article - How To Check Whether Physical Standby is in Sync with the Primary? ID 861595.1
It suggests comparing Last Sequence Generated on primary and Last Sequence Received on standby for each thread. And if they are not equal - NOT in sync.
My assumption is that if there are multiple threads we running RAC. Is it correct ?
And if we running RAC as far as I am aware it's sufficient to have one thread in sync.
Please comment on this
Thanks in advance
Edited by: user613441 on Jan 6, 2010 9:16 PM
Edited by: user613441 on Jan 6, 2010 9:17 PM

deleted

How To Check Whether Physical Standby is in Sync with the Primary

hi All,
I'm new in data guard. In our current production, my boss is asking me to write a shell scripts monitoring physical standby and primary archive is sync (archive gap)
I'm referring metalink [ID 861595.1], but when i ran the first query in Primary node, the screen hang.
On primary
========
SQL> SELECT THREAD# "Thread",SEQUENCE# "Last Sequence Generated"
FROM V$ARCHIVED_LOG
WHERE (THREAD#,FIRST_TIME ) IN (SELECT THREAD#,MAX(FIRST_TIME) FROM V$ARCHIVED_LOG GROUP BY THREAD#)
ORDER BY 1;
I tried to turn on 10046 SQL trace, the SQL consume a lot of CPU, and it full table scan in X$KCCAL table.
TKPROF result look like below:
call     count       cpu    elapsed       disk      query    current        rows
Parse        1      0.00       0.00          0          0          0           0
Execute      1      0.00       0.00          0          0          0           0
Fetch        1     88.29     115.59          0          0          0           0
total        3     88.30     115.60          0          0          0           0
Misses in library cache during parse: 1
Misses in library cache during execute: 1 Optimizer mode: ALL_ROWS Parsing user id: 80
Rows     Row Source Operation
      0 SORT ORDER BY (cr=0 pr=0 pw=0 time=21 us)
      0   FILTER (cr=0 pr=0 pw=0 time=19 us)
   1124    FIXED TABLE FULL X$KCCAL (cr=0 pr=0 pw=0 time=40537 us)
      0    FILTER (cr=0 pr=0 pw=0 time=115538625 us)
   1122     HASH GROUP BY (cr=0 pr=0 pw=0 time=115530193 us)
7465972      FIXED TABLE FULL X$KCCAL (cr=0 pr=0 pw=0 time=94648975 us)
Elapsed times include waiting on following events:
Event waited on                        Times   Max. Wait Total Waited
-----------------------------------   Waited ---------- ------------
SQL*Net message to client               1        0.00          0.00
control file sequential read        16841        0.05         30.88
SQL*Net break/reset to client           1        0.00          0.00Due to this is production environment, thus i had terminated the session, can anyone teach me or share with me any scripts to monitor physical standby is sync with primary or not? Or do you encounter the above issue when running the SQL?
My db version is Oracle 11.2.0.1.0
Thanks in advance.
Regards,
Klnghau

Hello;
Note 861595.1 has not been subject to an independent technical review. Not sure if that make it bad or not.
This is what I'm using: ( I spool this to a file and have it e-mailed to me daily)
PROMPT
PROMPT Checking last sequence in v$archived_log
PROMPT
clear screen
set linesize 100
column STANDBY format a20
column applied format a10
SELECT name as STANDBY, SEQUENCE#, applied, completion_time from v$archived_log WHERE DEST_ID = 2 AND NEXT_TIME > SYSDATE -1;
prompt
prompt----------------Last log on Primary--------------------------------------|
prompt
select max(sequence#) from v$archived_log where NEXT_TIME > sysdate -1;
Best Regards
mseberg

Logshipping has been broken after over writing the primary database

Hi All
One of DBA by mistakeover writing the primary database which is in logshpping ,Hence logshipping got failed.
Please some help us to re-configure the log shipping
Any doc or steps are very help for us..
Thanks in Advance

Could you please explain what exactly happened? What do you mean by overwriting?
You can always drop logshipping and recreate it.
Else you can take a full backup from Primary, Restore it in secondary with no recovery, remove all backups from the copy and restore shares and start the log shipping backup job, it should take the log backup, move it and restore it.
Regards, Ashwin Menon My Blog - http:\\sqllearnings.com

Shared services database not found in the specified database

Hi Guys,
In Hyperion EPM System confirurator,i gave all the fields and click next button it show error message "Shared services database not found in the specified database"

Hi ,
1. Pls confirm me one thing. Did you re -install version 11.
2. Re intstallation of version 11 has lot of issues ( atleast 11.1.1.1).
Sandeep Reddy Enti
HCC
http://hyperionconsutlancy.com/

Logical Standby Database Not Getting Sync With Primary Database

Hi All,
I am using a Primary DB and Logical Standby DB configuration in Oracle 10g:-
Version Name:-
SQL> select * from v$version;
BANNER
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 Solaris: Version 10.2.0.5.0 - Production
NLSRTL Version 10.2.0.5.0 - Production
We have build the logical standby last week and till date the Logical DB is not sync. I have checked the init parameters and I wont see any problems with it. Also archive log destinations are also fine enough.
We have a important table named "HPD_HELPDESK" where record count is growing gradually whereas in logical standby it's not growing. There are some 19K record difference in the both the tables.
I have checked the alert log but it is also not giving any error message. Please find the last few lines of the alert log in logical Database:-
RFS LogMiner: Registered logfile [oradata_san1/oradata/remedy/arch/ars1_1703_790996778.arc] to LogMiner session id [1]
Tue Aug 28 14:56:52 GMT 2012
RFS[2853]: Successfully opened standby log 5: '/oracle_data/oradata/remedy/stbyredo01.log'
Tue Aug 28 14:56:58 GMT 2012
RFS LogMiner: Client enabled and ready for notification
Tue Aug 28 14:57:00 GMT 2012
RFS LogMiner: Registered logfile [oradata_san1/oradata/remedy/arch/ars1_1704_790996778.arc] to LogMiner session id [1]
Tue Aug 28 15:06:40 GMT 2012
RFS[2854]: Successfully opened standby log 5: '/oracle_data/oradata/remedy/stbyredo01.log'
Tue Aug 28 15:06:47 GMT 2012
RFS LogMiner: Client enabled and ready for notification
Tue Aug 28 15:06:49 GMT 2012
RFS LogMiner: Registered logfile [oradata_san1/oradata/remedy/arch/ars1_1705_790996778.arc] to LogMiner session id [1]
I am not able to trace the issue that why the records are not growing in logical DB. Please provide your inputs.
Regards,
Arijit

How do you know that there's such a gap between the tables?
If your standby db is a physical standby, then it is not open and you can't query your table without cancelling the recovery of the managed standby database.
What does it say if you execute this sql?
SELECT PROCESS, STATUS, THREAD#, SEQUENCE#, BLOCK#, BLOCKS FROM V$MANAGED_STANDBY;The ARCH processes should be connected and MRP waiting for a file.
If you query for the archive_gaps, do you get any hits?
select * from gv$archive_gapIf you're not working in a RAC environment you need to query v$archive_gap, instead!
Did you check whether the archives generated from the primary instance are transferred and present in the file system of your standby database?
I believe your standby is not in recovery_mode anymore or has an archive_gap, which is the reason why it doesn't catch up anymore.
Hope it helps a little,
Regards,
Sebastian
PS: I'm working on 11g, so unfortunately I'm not quite sure if the views are exist in 10gR2. It's worth a try though!
Edited by: skahlert on 31.08.2012 13:46

Can we create standby database on same server as primary database

I have Oracle 10g Rel2, OS Windows server 2003. Can I create standby database on same server on which we have our primary database?

Bharat Kaushik wrote:
I have VMware server s/w, do i need to create 2 virtual machines inside the one installation of VMware server? Or do i need to have 2 different installation of VMware server?You could think of VMWare Server in similar terms are Oracle Database Server.
The HOME: The VMWare Server s/w is put into a "VMWare Home" (I use VMWare Workstation, it goes into C:\Program Files\VMware\VMware Workstation)
The INSTANCE: WHen I create a VM Image, I specify the name (like a SID), configuration (like PFILE), and location (like ORACLE_BASE/oradata/ ...)
And just like Oracle with which I can have 2 or more instances running on the host (if there is sufficient memory and other resource), I can have 2 or more VM Images running on my host.
And just like Oracle, if I have insufficient resources, things get slow like molasses in the middle of winter.
Hope your desktop has at least 4GB of OS-addressible RAM and at least dual-core CPU. ALternately ... I hope you have lots of coin for the coffee machine.

Standby database,Possible network disconnect with primary database

Hello
I have a database on linux with standby configured on different server with same os and database version.When i was about clone primary database through OEM,it stuck in the middle and popup for recovery. I restored and recover whole database. Then, the alert log on the standby database shows this error.
Possible network disconnect with primary database.
I checked standby databsase,listener is up & running. But archive files are getting into standby database.
What should i do now? Please anybody can put some light on this.
Thanks
Ariz

Is there anything in the primary alert log? Can you tnsping the standby instance from the primary server, and the primary instance from the standby server?

What is the logic behind the oracle database connections....

Hi,
We have crontab alerts are enabled for the oracle database client connection.
For Production databases the alerts are coming continuously until it gets connected.
For QA databases it will throw the message that the oracle client connection fails and again it will prompt whenever it gets connected.
Please let me know the logic behind these scnerions.
Pavan..
Edited by: dm_ptldba on Feb 14, 2012 6:45 AM

Hi,
Thanks for the update, Sorry the question was not clear. let me put it in a clear way.
Following is our crontab which alerts us if any database/listener goes down. But it alerts us only once for one successful/unsuccessful connection. For Eg: If the crontab not able to connect to the database for once then it throws an alert only once and waits for a successful connection.
I would like to change the logic in this crontab in such a way that, it should keep on alerting us for all the unsuccessful connections and once or twice for successful connection.
. /home/oracle/.bash_profile
. /opt/oracle/cron/cron_email
sidfile=/home/oracle/scripts/db-list.txt
dboutfile=/home/oracle/scripts/dboutfile.tmp
echo $ORACLE_HOME
TNS_DIR=$ORACLE_HOME/network/admin
ORA_BIN=$ORACLE_HOME/bin
cat $sidfile | while read SIDNAME
do
$ORA_BIN/sqlplus -s system/******@$SIDNAME 2> /dev/null >> $dboutfile <<EOF
@/home/oracle/scripts/db_up.sql
EOF
if [ $? -eq 0 ]
then
STATUS=1
if [ -f /home/oracle/scripts/${SIDNAME}-down.txt ]
then
/bin/mail -s "Alert :Oracle database instance \"${SIDNAME}\" is up & connected..." [email protected] </dev/null
rm -rf /home/oracle/scripts/${SIDNAME}-down.txt
fi
else
if [ -f /home/oracle/scripts/${SIDNAME}-down.txt ]; then
echo "";
else
touch /home/oracle/scripts/${SIDNAME}-down.txt
echo ${SIDNAME} "not Connected ..."
/bin/mail -s "Alert :Oracle database instance \"${SIDNAME}\" is down......" [email protected] </dev/null
fi
fi
done
Thank you.
PTLDBA

Creating a new database vs. Using the existing database

Hi,
I have to build a new desktop application over an existing oracle database -
Current database is around 300 gb
Has 1500+ tables
Tables are in denormalized form and has ungrouped table columns
High level of nesting is done in the triggers.
Now before I take decision to use existing database I want to know
1. what factors to consider while choosing between the existing db or a new db?
Thanks in advance.

Some of the things I would consider before including the two desktop applications together:
Would you mind impacting one or the other application (downtime) if you needed to bounce the database for any reason, or if there were issues with the database. Keeping them separate would prevent having to bring down both applications if the database went down for any reason (planned or otherwise).
Would the type of activity in both databases be so similar that all global optimizer and other parameter settings be perfect for both cases?
Would the SLA be the same for both databases? Would your customers agree to impacts of one application data if you had to roll back the database for the other application?
By having the two applications in the same database, you are also stuck with having to wait for both application vendors to certify their product with the next patch level or version for upgrades. Keeping them separate eliminates this problem.
Is there any compelling reason to keep them together?
Personally, I would keep them separate for all of the above reasons, and many more which I did not bother to write.

How to Create a database copy and specify the source database.

Hello,
In my dag environment I would like to create a new database copy and specify the source. My company has database copies in 2 locations. Instead of transferring the data from one location to the other, I would like to specify the source be the copy that is
in the same location.
Thank you,
~Mark

Hi,
We can create a mailbox database, then run EMC and EMS to create a database copy.
More details about Add a Mailbox Database Copy, for your reference:
https://technet.microsoft.com/en-us/library/dd298080(v=exchg.141).aspx
Note: please pay attention to Prerequisites section, for example:
1. The specified Mailbox server must not already host a copy of the specified mailbox database.
2. The server hosting the specified database and the server that will host the database copy must both be in the same database availability group (DAG). The DAG must also have quorum and be healthy.
Thanks
Please remember to mark the replies as answers if they help, and unmark the answers if they provide no help. If you have feedback for TechNet Support, contact [email protected]
Allen Wang
TechNet Community Support

Clone of auxillary database using rman with the source database unavailable

Hi,
Is it possible to clone a database with just the source database's backup pieces. The source database is currently unavailable. The source database backup was taken through rman - nocatalog mode. We have been provided the backup pieces. Any clues on how to go about it?

NAME TYPE VALUE
db_file_name_convert string /u34/oracle/data/FRD9, +DG_OAP
PST1/OAPPST1/DATAFILE, /u34a/o
racle/data/FRD9, +DG_OAPPST1/O
APPST1/DATAFILE, /u35/oracle/d
ata/FRD9, +DG_OAPPST1/OAPPST1/
DATAFILE, /u36/oracle/data/FRD
9, +DG_OAPPST1/OAPPST1/DATAFIL
E, /u37/oracle/data/FRD9, +DG_
OAPPST1/OAPPST1/DATAFILE, /u38
/oracle/data/FRD9, +DG_OAPPST1
/OAPPST1/DATAFILE, /u39/oracle
NAME TYPE VALUE
/data/FRD9, +DG_OAPPST1/OAPPST
1/DATAFILE, /u40/oracle/data/F
RD9, +DG_OAPPST1/OAPPST1/DATAF
ILE
log_file_name_convert string /u32/oracle/data/FRD9, +DG_
OAPPST1_AUX1/OAPPST1/ONLINELOG
SQL>

Standby database on same machine as primary database

hi,
I am testing the standby database and read all document on metalink also I am able to start the standby database in nomount mode nut when trying to mount the database I am getting following error
SQL> startup nomount
ORACLE instance started.
Total System Global Area 47679696 bytes
Fixed Size 731344 bytes
Variable Size 29360128 bytes
Database Buffers 16777216 bytes
Redo Buffers 811008 bytes
SQL> alter database mount standby database;
alter database mount standby database
ERROR at line 1:
ORA-01103: database name 'PRIMDB' in controlfile is not 'PRIMDBS'
I think the database name must be different on same machine. My os is Sun Solaris
Regards
Bhushan

When you need to have the standby database on the same server as the master, use lock_name_space in the init.ora.
This parameter permits this to occur without having to worry about changing the db_name.
http://www.oug-ottawa.org/pls/htmldb/f?p=103:21:729607614317189728::::P21_TIP_NUM:821

Since 10.6 install typing lags far behind keyboard

Whenever I type now, after just a few words the keystrokes showing up on the screen are wayyyy behind what I'm typing on they keyboard. Any ideas?
Same problem with all blue tooth keyboards (used three different ones to test), and happens in all programs.
Nice update. Every time I use my imac something newwwwwwwwwwwwwwwwwwwwww is wrong. Plus, did you see all those wwwwwwww's? That's because if I don't turn key repeat OFF, it just does it randomly.

So no one has any idea why I can't use ANY bluetooth keyboard in OS X 10.6?
I am now using a wired keyboard because not a single bluetooth keyboard works with 10.6. None. Not one.

Why do you clowns not optimize this for windows 7? Are you that far behind the times?

The question is above. I am going to delete Firefox is this is not fixed. I have a Samsung Galaxy S II
Fix the damn thing

How can Mozilla fix anything if you are not really saying what your issues are?
Also the Samsung Galaxy S II is compatible with the mobile Firefox for Android.

Physical Standby Database Lags Far Behind the Primary Database scenario

Similar Messages

Maybe you are looking for