Using primary data backup recovery

Hi all

Here we have an environment dataguard with db1 (db_unique_name) as primary and db2 (db_unique_name) Physics of the standby database. We have configured dataguard in environments of 10 g and we don't use a grid control. And now we want to configure rman for incremental backups. After you set up, we took full backup successfully in the newspapers of the previous day and archive. While the test case we dropped a primary data file and tried to get it back. It displays the following error. Also when I tried "overlapping backup" since the primary, is to show that all backups taken to sleep are "expiration". But the same backups show as 'available' RMAN connected standby mode. Can someone help with this problem?

Elementary school:

[oracle@db1 NIOS] $ sqlplus / as sysdba;

SQL * more: release 10.2.0.3.0 - Production on Sat Oct 6 02:58:11 2012

Copyright (c) 1982, 2006, Oracle. All rights reserved.


Connected to:
Oracle Database 10 g Enterprise Edition release 10.2.0.3.0 - Production
With partitioning, OLAP and Data Mining options

SQL > alter database datafile offline 4; (after the deletion of system-level data file)

Database altered.

[oracle@db1 NIOS] $ rman target sys/oracle@db1 catalog rman/oracle@appsdb

Recovery Manager: release 10.2.0.3.0 - Production on Sat Oct 6 02:58:58 2012

Copyright (c) 1982, 2005, Oracle. All rights reserved.

connected to target database: NIOS (DBID = 1590630518)
connected to the recovery catalog database

RMAN > restore datafile 4;

From catering to 12-SEP-06
allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid = 107 devtype = DISK

channel ORA_DISK_1: from datafile backupset restoration
channel ORA_DISK_1: specifying datafile (s) to restore from backup set
restoring datafile 00004 in /u01/app/oracle/oradata/NIOS/users01.dbf
channel ORA_DISK_1: backup /u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_nnndf_TAG20120905T174801_84gjqt0n_.bkp piece reading
ORA-19870: error reading backup total /u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_nnndf_TAG20120905T174801_84gjqt0n_.bkp
ORA-19505: impossible to identify the file ' / u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_nnndf_TAG20120905T174801_84gjqt0n_.bkp '.
ORA-27037: unable to get file status
Linux error: 2: no such file or directory
Additional information: 3
switch to the previous backup

channel ORA_DISK_1: from datafile backupset restoration
channel ORA_DISK_1: specifying datafile (s) to restore from backup set
restoring datafile 00004 in /u01/app/oracle/oradata/NIOS/users01.dbf
channel ORA_DISK_1: backup /u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_nnndf_TAG20120905T164212_84gdwdst_.bkp piece reading
ORA-19870: error reading backup total /u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_nnndf_TAG20120905T164212_84gdwdst_.bkp
ORA-19505: impossible to identify the file ' / u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_nnndf_TAG20120905T164212_84gdwdst_.bkp '.
ORA-27037: unable to get file status
Linux error: 2: no such file or directory
Additional information: 3
switch to the previous backup

channel ORA_DISK_1: restore datafile 00004
entry of data copy file recid = stamp 13 = filename=/tmp/users01.dbf 793208425
destination for the restoration of the data file 00004: /u01/app/oracle/oradata/NIOS/users01.dbf
RMAN-00571: ===========================================================
RMAN-00569: = ERROR MESSAGE STACK FOLLOWS =.
RMAN-00571: ===========================================================
RMAN-03002: failure of the restore command at 06/09/2012 02:59:10
ORA-19571: datafilecopy recid 13 stamp 793208425 not found in the control file
ORA-19600: input file is datafile 13-copy)
ORA-19601: output file is the data-copy 0 file (/ u01/app/oracle/oradata/NIOS/users01.dbf)

RMAN > crosscheck backup;

using channel ORA_DISK_1
verified backup item: found to be "EXPIRED".
backup handle=/u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_annnn_TAG20120905T164038_84gdsjlw_.bkp recid = piece stamp 18 = 793212040
verified backup item: found to be "EXPIRED".
backup handle=/u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_annnn_TAG20120905T164038_84gdtz6p_.bkp recid = piece stamp 19 = 793212087
verified backup item: found to be "EXPIRED".
backup handle=/u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_nnndf_TAG20120905T164212_84gdwdst_.bkp recid = piece stamp 20 = 793212132
verified backup item: found to be "EXPIRED".
backup handle=/u01/app/oracle/flash_recovery_area/DB2/autobackup/2012_09_05/o1_mf_s_793242038_84gdxw5j_.bkp recid = piece stamp 21 = 793212180
verified backup item: found to be "EXPIRED".
backup handle=/u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_annnn_TAG20120905T174759_84gjqrtn_.bkp recid = 22 = 793216080 stamp item
verified backup item: found to be "EXPIRED".
backup handle=/u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_nnndf_TAG20120905T174801_84gjqt0n_.bkp recid = 23 = 793216082 stamp item
verified backup item: found to be "EXPIRED".
backup handle=/u01/app/oracle/flash_recovery_area/DB2/autobackup/2012_09_05/o1_mf_s_793246288_84gjs9h4_.bkp recid = piece stamp 24 = 793216129
"" element of verified backup: turned out to be "AVAILABLE"
backup handle=/u01/app/oracle/product/10.2.0/db_1/dbs/c-1590630518-20120906-00 recid = piece stamp 5 = 793238431
Cut 8 objects

Stand-by:

[oracle@db2 NIOS] $ rman target sys/oracle@db2 catalog rman/oracle@appsdb

Recovery Manager: release 10.2.0.3.0 - Production Wed Sep 5 20:11:58 2012

Copyright (c) 1982, 2005, Oracle. All rights reserved.

connected to target database: NIOS (DBID = 1590630518, is not open)
connected to the recovery catalog database

RMAN > crosscheck backup;

allocated channel: ORA_DISK_1
channel ORA_DISK_1: sid = 91 devtype = DISK
"" element of verified backup: turned out to be "AVAILABLE"
backup handle=/u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_annnn_TAG20120905T164038_84gdsjlw_.bkp recid = piece stamp 18 = 793212040
"" element of verified backup: turned out to be "AVAILABLE"
backup handle=/u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_annnn_TAG20120905T164038_84gdtz6p_.bkp recid = piece stamp 19 = 793212087
"" element of verified backup: turned out to be "AVAILABLE"
backup handle=/u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_nnndf_TAG20120905T164212_84gdwdst_.bkp recid = piece stamp 20 = 793212132
"" element of verified backup: turned out to be "AVAILABLE"
backup handle=/u01/app/oracle/flash_recovery_area/DB2/autobackup/2012_09_05/o1_mf_s_793242038_84gdxw5j_.bkp recid = piece stamp 21 = 793212180
"" element of verified backup: turned out to be "AVAILABLE"
backup handle=/u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_annnn_TAG20120905T174759_84gjqrtn_.bkp recid = 22 = 793216080 stamp item
"" element of verified backup: turned out to be "AVAILABLE"
backup handle=/u01/app/oracle/flash_recovery_area/DB2/backupset/2012_09_05/o1_mf_nnndf_TAG20120905T174801_84gjqt0n_.bkp recid = 23 = 793216082 stamp item
"" element of verified backup: turned out to be "AVAILABLE"
backup handle=/u01/app/oracle/flash_recovery_area/DB2/autobackup/2012_09_05/o1_mf_s_793246288_84gjs9h4_.bkp recid = piece stamp 24 = 793216129
verified backup item: found to be "EXPIRED".
backup handle=/u01/app/oracle/product/10.2.0/db_1/dbs/c-1590630518-20120906-00 recid = piece stamp 5 = 793238431
Cut 8 objects

Thanks in advance.

Hello;

I would consider one of them step by step:

Use of primary database to retrieve backup RMAN backup files

http://oraganism.files.WordPress.com/2009/04/backup-from-physical-standby.PDF

How to use the backup of the previous day to restore primary db

http://alexzeng.WordPress.com/2008/09/13/how-to-use-standby-backup-to-restore-primary-DB/

Steps to recover the database to the primary database using a copy of the file to the standby database.

http://dbaregistry.blogspot.com/2010/01/steps-to-recover-primary-databases.html

RMAN in a data protection environment

http://Oracle.Su/docs/11g/backup.112/e10642/rcmarchi.htm#CHDIFGFH

Make a clone of database using a physical standby database Data Guard

http://gavinsoorma.com/2012/04/performing-a-database-clone-using-a-Data-Guard-physical-standby-database/

Best regards

mseberg

Tags: Database

Similar Questions

  • Simple question about data backup recovery

    Hi all

    It is the first time that I will propose Vmware Essentials Plus Kit as a solution to one of our customers.

    The solution will be X 2 DL180 server with 2 TB of storage each connected to an IBM iSCSI SAN DS3500 series.

    I want to be able to provide this scenario HA, and also, I want to use the data recovery backup to perform backups of virtual machines OF SAN DL180 server, so in the case of the SAN has very serious injury and can not be brought back online on the same day, in order to start the machines directly from the DL 180 servers.

    I would like to know if it is a supported scenario.

    Thank you

    Very welcoming. Unit vDR is certainly worth using because it allows long-term storage of backup and in a space images relatively small. Restoration of 10 mV because of a collapse of SAN he would not provide a simple and fast restoration.  The same 10 script ghettoVCB clones could be high in a few minutes.

  • A unit of data backup recovery

    I try on a VMware data recovery device and want to be able to backup the device itself.

    Is the best way to just the clone or the model of the unit?

    Hello

    Take a look at the following thread associated with the same room

    http://communities.VMware.com/thread/226152

    So this part is saved mentioned that you shouldn't really need to as it happens

  • OIM 11 g database using Active Data Guard recovery

    Hello

    Someone did he do IOM database failover configuration using the active data guard.

    Here are the details
    • Two application in primary servers and one in secondary sites
    • Two databases one in primary and the secondary sites.
    • Two application servers point to the primary database.
    • There is an Active Data Guard set up between primary and secondary data.
    • Secondary database is in stand-by mode.
    • Unique CNAME is configured and actively points to the primary database.
    • Failover of the database, the principal is down and the secondary database is marketed as primary and CNAME is changed to point to the secondary database.
    • All the data sources in Weblogic are configured to use CNAME.

    The problem here is the SID. In the primary, we have JSQOIM1P and in the secondary, the SID is JSQOIMDR.

    So when we switch the error we get is unknown TNS name like Weblogic data sources don't know the JSQOIMDR.

    How do configure us the failover in this scenario?

    Thank you
    Kunal jegou

    Hello

    I was able to achieve this by setting up a unique service name pointing to the primary db SID and the same name of unique service on the secondary db pointing to the db secondary sid.

    For the benefit of the users of the forum.
    CNAME - usoimdb pointing to primaryDBHost and can be changed to secondaryDBHost in the case of DR
    Service name - usoimdbsvc pointing to primaryDBSID and usoimdbsvc to secondaryDBSID on secondary db use

    Sources of data, directdb and Kingdom of security to help the chain of
    JDBC:Oracle:thin:@usoimdb:1523/usoimdbsvc

  • "BACKUP RECOVERY AREA DESTINATION" and several channels

    Hello

    I'm on 11 GR 2.

    I use the feature OF "< specific directory >" DESTINATION of the order of CONSERVATION RECOVERY AREA

    staging the backup, my CRF, a few hours earlier. I wish

    1 / to be able to compress this backup in my <>directory (my regular backup to the FRA is * not *)

    Tablet!) ;

    2 / and to parallelize this staging.

    Here http://docs.oracle.com/cd/E11882_01/backup.112/e10643/rcmsubcl018.htm#CIHBDIGE

    at some point, they say that "this paragraph, when it is used with the BACKUP RECOVERY AREA , allows.

    allowing you to designate a drive as the location channel': How do you understand this? Do you want that he woud allow you

    to specify things like

    Run {}

    ALLOCATE CHANNEL c1 DEVICE TYPE disk FORMAT ' / usr/oracle/FRAstaging ' MAXPIECESIZE 100 G;

    TYPE of DEVICE of c2 ALLOCATE CHANNEL FORMAT drive "/ usr/oracle/FRAstaging ' MAXPIECESIZE 100 G;

    recovery zone backup destination c1;

    recovery zone backup destination c2;

    }

    or should I rather do something like:

    Run {}

    TYPE of DEVICE of c1 ALLOCATE CHANNEL MAXPIECESIZE 100 G drive.

    TYPE of DEVICE of c2 ALLOCATE CHANNEL MAXPIECESIZE 100 G drive.

    recovery zone backup destination "/ usr/oracle/FRAstaging";

    }

    or simply? :

    Run {}

    TYPE of DEVICE of c1 ALLOCATE CHANNEL MAXPIECESIZE 100 G drive.

    TYPE of DEVICE of c2 ALLOCATE CHANNEL MAXPIECESIZE 100 G drive.

    backup recovery area;

    }

    Has anyone tried to stage his FRA on some directory of operating system or backup tape?

    Which of the above syntax is correct? Any idea?

    Thank you very much!

    Kind regards

    SEB

    Hello

    The correct syntax is the second. Take a look:

    RMAN> run {
    2>  ALLOCATE CHANNEL c1 DEVICE TYPE disk MAXPIECESIZE 1G;
    3>  ALLOCATE CHANNEL c2 DEVICE TYPE disk MAXPIECESIZE 1G;
    4>  backup recovery area to destination '/u04/temp';
    5>}
    
    allocated channel: c1
    channel c1: SID=453 device type=DISK
    
    allocated channel: c2
    channel c2: SID=36 device type=DISK
    
    Starting backup at 09-OCT-14
    specification does not match any datafile copy in the repository
    channel c1: starting compressed archived log backup set
    channel c1: specifying archived log(s) in backup set
    ......
    

    The first syntax is bad because you can't say "to destination c1. The destination must be a path.

    The third argument is false because ' RMAN-06603: DESTINATION option must be specified with the AREA of RECOVERY, RECOVERY of FILES or DB_RECOVERY_FILE_DEST on the disk device.

    Kind regards

    RJ

  • where are my data after the use of Dell backup and recovery

    Three check marks displayed on a Dell backup and recovery, which, I suppose, finishing meant that my backup was successful (using Windows on a Dell Inspiron 8.1).  Where my data is sent to after the use of Dell backup and recovery?

    Click on the 02:22 mark of the video.

    www.YouTube.com/watch

  • Can Equium A100 - I use E drive for data backup?

    My Equium A100-27 has a 80 GB hard drive, it is split in the C drive half and drive E.

    I understand that the E drive is used for system recovery purposes, but can I also use it to store data?

    If I use it for how much room must be left free for system data recovery?

    It's old laptop model, and there is no recovery of data stored on the second partition of HARD drive.
    If you can do what you want with this partition, and also use it for backup of data or backup data.

  • Using more of a VMWare backup recovery

    Hi all

    I'm sitting here und think next idea:

    We have 10 servers that must be saved. We use a VDR (VM Backup Recovery) for this. Thus, sometimes the catalog crashes and it takes neerly a day to fix it. While this time I can´t of backup or restore anything.

    These backups are on the same SAN as my ESX machines. So I have the big problem, if everything crashes (raid or more than 1 HDD) I have no chance to recover anything.

    Yes, I know that this is not 100% optimal. either way... we have no money to buy a second SAN!

    the idea was to develop an other VDR and connect it to a Terastation. Speed is absolutely secondary. This VDR would save computers same like the first VDR. (not at the same time, for example SQL Server is saved on VDR1 on VDR2 and 06:00 on 09:00)

    The biggest would be:

    -If VDR 1 catalog crashes, VDR2 works and I can fix the VDR1 without missing the backups or recoveryoptions

    -If my STI SAN crashing, I can use my Terastation to fix everything (the speed is secondary)

    What do you think of this idea?

    Good bye

    Andy

    You can define 2 different backup locations in a single device. So you can save on Monday for the first location and Tuesday for the second and so on. When an integrity check runs more than 24 hours, you have the second location-data store to restore on the same device operations. If you have enough backup-time, you can also save each VM twice in one day in two places at different times. In the event of a disaster you will be lucky and maybe lose only a location of data store and have all your backups without waitiing for the completion of the audit of integrity or something as a recataloguer. Up to now this works very well in my environment with 5 appliances and 180 VM. I use no more than 30 virtual machines for a single device, because integrity, check runs throughout.

  • guys, how can I stop backup during I use cellular data

    guys, how can I stop backup during I use cellular data

    iCloud backup requires a wi - fi connection. It does not backup when only cellular data are available. See: get help from backup of your device in iCloud - Apple Support

  • I can't use iCloud for backup of data from third-party applications

    I use an iPod touch 6th generation and I wanted to use iCloud for backup of Twitter, IG, Pinterest etc. I've tried searching steps. So far I've been finding the same set of instructions:

    Settings > iCloud > storage > select your iDevice > turn on apps you want to iCloud to save

    But after clicking on the storage, I don't see that 'manage storage' and below, the option 'change of Plan '. Why does my iPod not not the option choose iDevice? I can see the option to save the data of integrated applications, but not applications, third-party. My iCloud that unit is on but my Photos under icloud is disabled. I really need the extra space. What seems to be the problem?

    Click Manage store (settings > iCloud > storage > Storage manage and you see what moves backup iCloud storage.)

  • Using a USB flash drive for backup more driver and application backup recovery

    I have a 32 GB flash drive I used to create a recovery for my Aspire S7 backup using Windows 8.1. I understand that in creating this backup everything on the flash drive is deleted before creating backup. Can I now use this drive for other purposes, e.g., pilots and backup applications, pictures, documents, etc., or this flash drive now should he remain dedicated to the sole purpose of being a backup recovery?

    Hi zimmerbob

    I keep as a backup to dedicated recovery flash drive and use one for drivers & applications.

    It is always wise to keep the recovery discs, as well as the recovery disks. Flash memory is quite inexpensive for the moment!

  • Essbase data backup and recovery

    Hello

    We have a process in place for the App backup folder... I saw a process where data recovery is made by just the files ind and pag... and the restructuring of the DB... it is a good practice for the data backup and recovery...

    Assuming that the pag and index files are in the same location that the db files then, stop the application, restore the files, start up, up to you if you want to restructure as if you have restored all the files, it will be in the same State when the backup was performed.

    See you soon

    John
    http://John-Goodwin.blogspot.com/

  • NB500: How to use the DVD of recovery without a DVD player in the netbook?

    Background:

    I have a 500 NB of Netbook that came with a Win7 Starter. I've upgraded to Win8 but doesn't like the wsn can't resolution screen high enough to open applications. The system became slower and slower like time, I decided to reset the device back to the original settings by using the Win8 function.

    Progress in this process has been extremely slow and after seven hours (60% in fact) I had a scare that the content of my HARD disk can be copied somewhere on the web (cloud of NSA?  :)).

    I turned off the unit and then nothing would no longer work. I bought the recovery software with WIN7 Starter of Toshiba Support that I couldn't find an ISO for WIN7 starter file, but it happened on a DVD...

    I have:
    + NB 500
    + Toshiba Recovery DVD
    + Bootable USB stick (I created this)
    + Product key for Win7 Starter

    Question:
    What files are required to install Win7 from scratch (move files from DVD to the USB bootable)?
    This procedure deletes old data from the HARD drive or will it take space?

    Concerning
    LS1 - d

    It is not just that you think it is. As much as I know, you can't manipulate the data of recovery and work on your own boot media.

    A better and easier way for you is to get somewhere external USB ODD and use for installation of recovery image. I NB520 and I bought external BD ODD and can use it for various actions such as the installation of the software for example.

    I m wondering really you didn't create recovery USB drive. That was my first step after I bought this machine. Product key you have on the Microsoft sticker doesn't help. You can't use it for OS installation if you are using the version of the OS clean and you don t need it if you use original recovery image.
    After the installation of recovery image pre-installed Windows version is correctly activated.
    > Will this procedure remove old data from the HARD drive or will it take space?
    Yes. The whole HARD disc will be deleted so until you start with the installation of backup recovery image all your important data.

  • "backup recovery area" ignore all save them

    Hello everyone.

    I use 11.2.0.4 on OL 6.5 EE.

    I use suggested the Oracle backup strategy above I use "backup recovery area" at the end of the script and I spotted that he doesn't know to save them all. Please notify.

    Hello everyone and thanks for your help.

    I studied the issue and has managed to solve.

    First I mw show you the parameters persistent rman:

    Database with db_unique_name MYDB RMAN configuration parameters are:

    CONFIGURE RETENTION POLICY TO RECOVERY OF 7-DAY WINDOW;

    CONFIGURE BACKUP OPTIMIZATION

    SET UP DEFAULT DISK DEVICE TYPE;

    CONFIGURE CONTROLFILE AUTOBACKUP ON;

    CONFIGURE CONTROLFILE AUTOBACKUP FORMAT for DEVICE TYPE DISK to "+ fra/mydb/controlfile/%F";

    SET UP THE DEVICE TYPE DISK PARALLELISM 1 BACKUP BACKUPSET TYPE;

    CONFIGURE BACKUP OF DATA TO DISK FILE TYPE DEVICE TO 1;

    CONFIGURE BACKUP ARCHIVELOG FOR DEVICE TYPE DISK TO 1;

    CONFIGURE MAXSETSIZE TO UNLIMITED;

    CONFIGURE THE DATABASE ENCRYPTION

    CONFIGURE THE ENCRYPTION ALGORITHM "AES128";

    CONFIGURE COMPRESSION ALGORITHM 'BASIC' AND 'DEFAULT' LIBERATION OPTIMIZE FOR TRUE LOAD;

    CONFIGURE THE NONE ARCHIVELOG DELETION POLICY;

    CONFIGURE SNAPSHOT CONTROLFILE NAME TO ' / oracle/product/11.2.0/dbhome_1/dbs/snapcf_mydb.f';

    Now I use another script to take my backup and it works fine

    RMAN > run {}

    2 > crosscheck backup;

    3 > overlapping copy;

    4 > overlap archivelog all;

    5 > DELETE noprompt EXPIRED ARCHIVELOG ALL;

    6 > delete noprompt ARCHIVELOG until TIME ' SYSDATE-8';

    7 > DELETE noprompt EXPIRED BACKUP;

    8 > delete noprompt obsolete;

    9 > GET COPY OF DATABASE WITH TAG 'PAYMENT' until ' SYSDATE-8';

    10 > LEVEL 1 INCREMENTAL BACKUP TO RECOVER WITH THE TAG 'MYDB' COPY DATABASE;

    {11 >}

    12 > run {}

    13 > GAME CONTROLFILE AUTOBACKUP FORMAT for DEVICE TYPE DISK TO ' / storage/flash_recovery_area/MYDB/controlfile/%F';

    14 > save destination recovery area "/ storage/flash_recovery_area";

    15 > BACKUP VALIDATE CHECK DATABASE LOGIC ARCHIVELOG ALL;

    {16 >}

  • Backup/recovery of cold one Table

    I'm under 11r2 in a CCR environment. We have a primary and two standby (using Active Data Guard). Also, I'm a new s/n, bare with me. :)

    TableX in a given database is damaged. The corruption began a month earlier due to change in the code of the web developer. Changing the code was set a week ago.
    Summary: the data in TableX are corrupt until a week ago.

    Flashback does not extend that far in the past, so what I thought to do is, on our test server:

    (1) copy the relevant files of the production on the test server
    (2) backup the original TableX on production
    (3) to restore the system, sysaux, cancel and the tablespace that contains TableX on test server
    (4) drop all other areas of storage in condition to mount (or not, if this step is not necessary?)
    (5) to recover the database up to when the table was dropped or updated updated
    (6) take a TableX database export and import it in production, replacing the current TableX
    (7) update the imported production TableX with week last of 'good' data from the backup in step 2

    I have 3 questions.
    This seems like a good strategy?
    In addition, I have trouble to determine what the "record" is in step 1. Which would be copied on? Everything in the ArchiveLog file? Or in the db_recovery_file_dest folder? Or...?
    Finally, in step 3, is it possible to restore without overwriting pre-existing data on the test server?

    Thank you

    Hello;

    For future use since we both use Linux

    #!/bin/bash
    ####################################################################
    #
    # Data Pump Full script
    #
    ####################################################################
    
    if [ "$1" ]
    then DBNAME=$1
    else
    echo "basename $0 : Syntax error : use . expdp_full.sh  "
    exit 1
    fi
    
    export DMPDIR=/tmp
    
    #
    # Set the Environmental variable for the instance
    #
    . /u01/app/oracle/dba_tool/env/${DBNAME}.env
    #
    #
    
    cd $DMPDIR
    
    # check for existing logfile and
    # remove if found.
    if [ -s ${DMPDIR}/${LOGFILE} ] ; then
    /bin/rm ${DMPDIR}/${LOGFILE}
    fi
    
    # check for existing dump file and
    # remove if found.
    if [ -s ${DMPDIR}/${DUMPFILE} ] ; then
    /bin/rm ${DMPDIR}/${DUMPFILE}
    fi
    
    # change to the working directory
    cd ${DMPDIR}
    
    # call expdp and use the full path to it.
    $ORACLE_HOME/bin/expdp parfile=expdp_full.par DUMPFILE=${DUMPFILE} LOGFILE=${LOGFILE} DIRECTORY=DMPDIR
    
    # once the expdp is complete gzip it to save space.
    /bin/gzip -S .`date +%m%d`.Z ${DUMPFILE}
    
    # delete the existing dump file. Zip does not remove it.
    if [ -s ${DMPDIR}/${DUMPFILE} ] ; then
    /bin/rm ${DMPDIR}/${DUMPFILE}
    fi
    
    # rename the log file to preserve it.
    if [ -s ${DMPDIR}/${LOGFILE} ] ; then
    
    /bin/mv ${DMPDIR}/${LOGFILE} ${DMPDIR}/${LOGFILE}.`date +%m%d`
    fi
    
    # remove old logs and exports
    # 
    
    find ${DMPDIR}/ -name "*.Z" -mmin +1440 -exec \rm {} \;
    find ${DMPDIR}/ -name "*.log*" -mmin +1440 -exec \rm {} \;
    
    exit
    

    expdp_full.par

    USERID="/ as sysdba"
    FULL=Y
    

    Examples of Enviro (you will need more) (/ u01/app/oracle/dba_tool/env/PRIMARY.env)

    export LOGFILE=PRIMARY_expdp_full.log
    
    export DUMPFILE=PRIMARY_expdp_full.dmp
    
    export DMPDIR=/u01/oradata/PRIMARY_export
    

    I put the nominal file in the DMPDIR and run this daily. Then it goes to the band. Won't help but today...

    mseberg

Maybe you are looking for