Why is backup RMAN 2 TB DB still 2 TB when expdp dump was only 600 MB?

RDBMS version: 11.2.0.3/Oracle Linux 6.2

We took a RMAN of below mentioned DB Backup.


Since DBA_SEGMENTS showed 2 TB, I thought that RMAN backup size will be half of this IE. 1 TB. But the size of the RMAN backups was 1.95 TB!

expdp full DB was only 600 MB!

SQL > select sum (bytes/power (1024,3)) of dba_segments;

SUM (BYTES/POWER (1024,3))
------------------------
2056.27728

SQL > select sum (bytes/power (1024,3)) sumbytesGB, sum (maxbytes/power (1024,3)) sumMaxBytes of dba_data_files;

SUMBYTESGB SUMMAXBYTES
---------- -----------
5451 427.998978


SQL > col file_name format a75
SQL > set 300 pages
SQL > select file_name, bytes/power (1024,3) bytesGB, maxbytes/power (1024,3) from dba_data_files;

FILE_NAME BYTESGB MAXBYTES/POWER (1024,3)
--------------------------------------------------------------------------- ---------- ----------------------
+CMS_DATA/bcmsprd/datafile/cms_s_asset_xa_dat_01.dbf 20 0
+CMS_DATA/bcmsprd/datafile/cms_s_asset_xa_dat_02.dbf 20 0
+CMS_DATA/bcmsprd/datafile/cms_s_asset_xa_dat_03.dbf 20 0
+CMS_DATA/bcmsprd/datafile/cms_s_asset_xa_dat_04.dbf 20 0
+CMS_DATA/bcmsprd/datafile/cms_s_asset_xa_dat_05.dbf 20 0
+CMS_DATA/bcmsprd/datafile/cms_s_asset_xa_dat_06.dbf 20 0
+CMS_DATA/bcmsprd/datafile/cms_s_asset_xa_dat_07.dbf 20 0
+CMS_DATA/bcmsprd/datafile/cms_s_asset_xa_dat_08.dbf 20 0
+CMS_DATA/bcmsprd/datafile/users.271.803502221 6 31.9999847
20 20 +CMS_DATA/bcmsprd/datafile/undotbs1.265.803502221
+CMS_DATA/bcmsprd/datafile/SYSAUX.272.803502221 10 31.9999847
+CMS_DATA/bcmsprd/datafile/system.260.803502221 10 31.9999847
20 20 +CMS_DATA/bcmsprd/datafile/undotbs2.261.803502407
20 20 +CMS_DATA/bcmsprd/datafile/undotbs3.267.803502407
+CMS_DATA/bcmsprd/datafile/cms_s_asset_xa_dat_09.dbf 20 0
+CMS_DATA/bcmsprd/datafile/cms_s_asset_xa_dat_10.dbf 20 0

< snipped >

RMAN backs up all blocks of formatting - even if the lines have been removed and the blocks are "logically empty."  expdp, obviously, do not save deleted rows.  Similarly, if you have created a table, and then later he truncated, the blocks are formatted and be candidates for RMAN backups.  Truncated tables would be "empty" and would result in 0 rows exported.

So, you will have this problem if you have deleted a large number of lines and/or issued truncates.

Hemant K Collette

Tags: Database

Similar Questions

  • Why the help and support of software still pop - up when I click on the buttons in Windows Vista SP1?

    I have this laptop to my cousin, because he does not use it more because of some problems. I tried several methods, but I still have this problem. When I open any software, automatically help and support appears.

    I hope you can help tnx I...

    Hi Jalke116,

    You can try these steps and see if it helps.

    Step 1:

    You can try to perform a clean boot and check if the problem persists.

    A clean boot to check if startup item or services to third-party application is causing this issue.

    You can read the following article to put the computer in a clean boot:

    How to troubleshoot a problem by performing a clean boot in Windows Vista or in Windows 7

    After you resolve the problem, see the section on boot link to how to restore your computer to a Normal startup mode.

    Step 2:

    You can also read the following article and try to download the Microsoft Safety Scanner and perform a full scan on your computer.

    Microsoft safety scanner

    Note: When you perform the analysis, there are chances of losing the data that you can take a backup of important data before performing analysis.

    Hope this information is useful.

  • Help remove backups rman (delete the repository)

    Hi all.

    I will soon be administering a few databases Oracle (11g Rel2). Currently these DBs area in log mode archive with a backup rman in place policy.


    I don't like this policy and you want to use my own, incremental backups from updadateable, so this is the plan I have in mind.


    (1) disable mode archive.


    (2) a clean shutdown and take a backup to cold.


    (3) since after you disable archiving, backups RMAN ALL will be unusable, I want to remove them (including newspapers archive and backup sets).


    (4) after the correct deletion, set up some persistent RMAN settings, such as retention policy, controlfile autobackup and optimization.


    (5) turn on archiving


    (6) to set up the new policy based backups backup incremental updateable with duplex location for archiving logs.


    Help is requested on the best way to clean all the backups (3).


    I can manually delete the backups and do overlap, but according to the Oracle Documentation "Delete backup" will be delete all backups for data recorded in the RMAN Repository.


    So which is the best way easier and faster? Will delete backup include the archive logs?


    Thanks in advance. Luis!


    myluism wrote:

    Can I still use them?

    I would say it is "the wrong answer to evil. When you disable log mode archive, you can only back up the database when it is stopped and only to restore and recover the database until the last stop.

    Looking at your plan, I wonder what you are trying to accomplish first log mode archive disabling and then enabling archive mode log again. Why do you think that this is necessary and what you're trying to accomplish.

    You need not disable and enable log mode archive in order to eliminate the obsolete achivelogs.

    If you are not satisfied with the retention policy, simply change it. All backups that are outside your retention policy will be considered as obsolete and you can remove them using RMAN delete obsolete, which will remove all files as well as the appropriate data backup from the RMAN Repository.

    For example you can archivelog delete all the entries backup and then delete the backupset or remove them at the OS level, do overlap and then use delete noprompt expired archivelog all.

    The retention policy applies to backup and archivelogs. However, you can have a separate archivelog deletion policy that applies to archivelogs.

  • Behavior of abnormal (Tag Date and place of automatic backup) RMAN script

    Hi all

    I have 11 GR 1 material running on Sun SOLARIS 10.

    The parameter default RMAN for backup of control file is on DISK (format: u02/oracle/backup/rman/control/%F.ctl)


    I have the RMAN script following (L0.rman):
    Run {}
    allocate channels ch1 type SBT_TAPE parms 'ENV=(TDPO_OPTFILE=/opt/tivoli/tsm/client/oracle/bin64/tdpo.opt) ';
    tag backup "${CURRENT_DATETIME}" additional 0 filesperset 3 format 'DB_%T_%U.dbf' database more archivelog;
    output channel ch1.
    }
    host ' cp /oracle/app/oracle/product/11.1.0/db_1/network/admin/tnsnames.ora/u02/oracle/backup/rman/$(echo $CURRENT_DATETIME) /';


    This is called by the following (L0.sh) script:
    #! / bin/bash
    CURRENT_DATETIME = "$(date + 20 %y %m %d %H %M)»;) export CURRENT_DATETIME
    export ORACLE_SID = EDWH
    export ORACLE_HOME=/oracle/app/oracle/product/11.1.0/db_1
    mkdir/u02/oracle/backup/rman/$(echo $CURRENT_DATETIME)
    RMAN nocatalog target / cmdfile = L0.rman log = L0_EDWH_$ CURRENT_DATETIME.log


    The anomalies are as follows:

    1. the backupsets are created with a literal tag "${CURRENT_DATETIME}" instead of "201010051915" - why?
    FYI, the command 'host' at the bottom of L0.rman successfully copies the file in the right directory (/ u02/oracle/backup/rman/201010051915 /).

    2. after the backup finishes, the automatic backup for the control and spfile is taken on SBT_TAPE instead of DISK - why?

    Any help will be appreciated.

    1. well, this $CURRENT_DATETIME is variable shell, RMAN does not interpret it (she can't and doesn't have to.) HOST calls the external commands by spawning the shell in non-interactive mode and when placing the order for her, and since you have exported the variable is available to all processes having engineered and develop properly, then files are copied in the expected location. If you remove the '; CURRENT_DATETIME export"from your shell script, you will see files copied in/u02/oracle/backup rman / / because the variable will not be in the environment of the layed shell (it will be local to the script) and will expand to the empty string. You must pass $CURRENT_DATETIME to the RMAN script as a variable substitution so that the shell expands it before calling RMAN, like this:

    L0. RMAN: {code}
    Run {}
    allocate channels ch1 type SBT_TAPE parms 'ENV=(TDPO_OPTFILE=/opt/tivoli/tsm/client/oracle/bin64/tdpo.opt) ';
    backup format tag & 1 extra strength 0 filesperset 3 'DB_%T_%U.dbf' database archivelog;
    output channel ch1.
    }
    host ' cp /oracle/app/oracle/product/11.1.0/db_1/network/admin/tnsnames.ora/u02/oracle/backup/rman / $ CURRENT_DATETIME /';
    {code}
    L0.sh: {code}
    #! / bin/bash
    CURRENT_DATETIME = "$(date + 20 %y %m %d %H %M)»;) export CURRENT_DATETIME
    export ORACLE_SID = EDWH
    export ORACLE_HOME=/oracle/app/oracle/product/11.1.0/db_1
    mkdir/u02/oracle/backup/rman / $ CURRENT_DATETIME
    RMAN nocatalog target / cmdfile = L0.rman using $CURRENT_DATETIME log = L0_EDWH_$ CURRENT_DATETIME.log
    {code}

    Note the * & 1 * variable substitution in the RMAN script and * using * keyword in the RMAN command line.

    2. control file record is taken to the same media, that the last backup, this is normal. You can't "CONFIGURE CONTROLFILE AUTOBACKUP ON the DISC" or something like that, you can only enable or disable it and specify format for various types of devices. Default device type does not play here, you are precisely the configured default device type substitution tape backup.

  • UNDO Tablespace backup-RMAN

    Hi/team champions,
    I'm confused in full DB Backup RMAN why CANCEL tablespce is included even we include also archive logs.
    in a Word?
    Why UNDO tablespace backup is necessary while taking backup RMAN?

    Kind regards
    Shitesh Shukla

    If I do not want to include the UNDO tablespce in my backup

    What happens if I took only backup files except CANCELLATION n TEMP

    NOT an option. You must include UNDO. 11g has optimization in order to safeguard an Undo tablespace.
    10 gr 2 what you pourriez do is

    1. create a new Undo Tablespace
    2. position yourself in the new Undo Tablespace
    3. check that the old Tablespace Undo is no longer in service - confirm that there is no transaction active in the inside undo segments
    4. remove the old Undo tablespace
    5. start your RMAN backup

    You run the risk of
    a. wait long to step 3
    b. ORA - 1555 s for users once you drop the Undo tablespace to step 4.

    Hemant K Collette
    http://hemantoracledba.blogspot.com

  • backup rman design database with phsical ensures

    Hello

    12 c database, after physical expectation is created, I need backup rman configuration,

    The goal is: a) backup of database b) remove archivelogs that have been backed up, free up space FRA.

    Design as below:

    1:

    in RMAN (primary and standby):
    RMAN >
    Configure controlfile autobackup on;
    set the type of backup device type disk compressed backupset;
    CONFIGURE ARCHIVELOG DELETION POLICY TO APPLIED ON ALL STANDBY;


    2: primary school:
    RMAN > CONFIGURE RETENTION POLICY to the 5 DAY RECOVERY WINDOW;

    3: in priamry, turn on track changes to block

    4: primary, implemented since rman level 0 (full backup) on Sunday.

    5: in the primary, the installation program rman level 1 backup from Monday to Saturday.

    6: primary, set up since rman "save archive backup delete them all obsolete" every hour;

    questions here:

    1. as in step 5, for the protection of the o and delete archivelog, the script using:

    Run {}
    allocate channel d1 type disk;
    SQL 'alter system archive log current';
    overlap archivelog all;
    backup AS COMPRESSED BACKUPSET archivelog all delete them all input;
    output channel d1;
    }

    I wonder if ' archivelog all delete all entries "will delete archivelog which was not taken on the eve of the database and broken standby database

    (hanging and out-of-sync with primary)... it is a concern, is that possible?

    2. If the primary database crashed, failover to standby mode, then restore/recovery primary, switch back to the primary, this design is just? If you have any good suggestions, please let me know.

    (to put it simply, I don't want to full/incremental backup to install standby, primary server CPU is good enough)

    Thank you very much!

    A few brief notes worth a visit:

    Based on Oracle 11 g Data Guard for the backup database and restore

    http://www.databasejournal.com/features/Oracle/article.php/3830766/leveraging-Oracle-11g-Data-Guard-for-database-backup-and-recovery.htm

    Standby database using RMAN

    http://Arup.blogspot.com/2007/07/standby-database-using-RMAN.html

    Best regards

    mseberg

  • Can store backups RMAN in a diskgroup in ASM?

    RDBMS version: 11.2.0.4

    OS: Oracle Linux 6.6

    RAC environment

    Can store backups RMAN in a diskgroup in ASM? If so, is it recommended?

    Yes, you can:

    CONFIGURE CHANNEL DEVICE TYPE DISK FORMAT '+diskgroupname';
    

    It depends on your requirements and backup strategy.

  • The user managed based backup Rman backup cold cold vs

    Hi all

    I just want to know the advantages and disadvantages for backup managed user cold and cold backup Rman.

    What is the best cold backup?

    Some some can help me?

    Thank you

    Cedric Joshi

    Hello

    Both are good only... If you took the clod backup (managed by the user), you can also do with RMAN catalog.

    I feel clod RMAN backup is good.

    Reason:

    ======

    ++ with RMAN you can avoid used used block, you can only backup used blocks

    ++ RMAN has its own compression algorithm

    ++ We also have the ability to assign multiple channels to parallelize the backup

  • Do not lock - automate the deletion of backups RMAN old day after a success

    Hi all

    Admin: Do not lock this thread. I moved here the database instance based on the suggestion of another Member, and the question to this forum. If you want to lock a thread, lock one in the forum of the database, not this one.

    I am looking for a way to automate the deletion of backups RMAN old days only after the occurrence of a success. At the present time, I have daily backups of L0 via crontab, and I need to create a script that can tell if a successful backup is done, then delete the old L0 if it has. Any suggestions on how to start? I'm stuck in code to see if a successful backup has been performed.

    Oracle 11.2.0.3

    Exadata (4 nodes)

    Linux 5.10

    * a from database forum

    A backup is completed when the backup command is complete.

    You have two 'blocks execution' in your rman script.

    Put the DELETE OBSOLETE between the end of the last execution block and EXIT:

    OUTPUT CHANNEL fs1b;
    OUTPUT CHANNEL fs2b;
    OUTPUT CHANNEL fs3b;
    CHANNEL fs4b;
    CHANNEL fs5b;
    CHANNEL fs6b;
    CHANNEL fs7b;
    CHANNEL fs8b;

    }

    remove the backup noprompt obsolete;
    "exit";

    You also want to do some other household:

    cross-checking of backup;

    delete expired backup noprompt.

    overlap archivelog all;

    delete noprompt expired archivelog all;

  • backup rman Exadata with zfs

    Our company has recently decided to start backup rman ZFS instead of regular tape backup.

    I would like to know what is the right measures to check if ITS configured right device of ZFS, the implementation of rman backups, etc.

    I'd appreciate someone here can point me to the right document.  I read ZFS Q and A and backup using ZFS best practices. All very unclear and seems none of them provide a digital step step.

    I wonder if Oracle provides an instruction step by step for the DBA.

    9233598-

    Q documents & you examined include the "Oracle ZFS storage pool: FAQ: Exadata RMAN backup with Oracle ZFS Storage Appliance (Doc ID 1354980.1)" Note? This note has a lot of information and is updated regularly and includes information about configuring ZFS and RMAN backups to the device of ZFS.

    Some useful books of whites, an I have used in the past and has been updated again last year: Oracle Exadata with the Sun ZFS Storage Appliance protection: best practices in Configuration and backup and recovery Performance and best practices using Sun ZFS Storage Appliance with Oracle Exadata Database Machine

    These documents will in many details - including providing examples of scripts.

    HTH.

    Thank you

    Kasey

  • Why the effects do not work on still images.

    Why some effects do not work on still images?

    As the active camera or earthquake?

    It seems to me that to have an earthquake or the busy camera, a series of images will have to be moved by the software.  In other words, you need a video clip with separate images.  A still is an image.

    You might be able to get around that.  Place your image in a dedicated project, a video share/export.  There will be many executives you need.  Report a project like the video rather than as a still.  Then you should be able to apply your chosen effects.

    I did not had time to try it, but I think it should work.

    Bill

  • How to use backups RMAN to restore a RAC database to single instance on another host?

    How to use backups Rman to restore a RAC database to single instance on another host?

    I tried to copy these inline for you:

    ------------




    Backups RMAN disk HowTo restore database RAC to single Instance on another node (Doc ID 415579.1)

    Down

  • Why can I not change my menu still in Photoshop?

    Why can I not change my menu still in Photoshop? When I click on the [edit in Photoshop] button, it happens in Photo Viewer Windows saying cela "Windows Photo Viewer cannot open this photo because Photo Viewer does not support this file format, or you do not have the latest updates for photo viewer."  I don't even want to open it in Photo Viewer, I tried to Photoshop! Is someone can you please tell me what is the problem?

    You must re-associate with Photoshop psd files. You're on a PC?

    See this:

    "Edit Menu in Photoshop" does not?

  • Restoring a backup RMAN to a new instance of the database

    Hi guys,.

    I have a problem with restoring a backup RMAN I have with me a new instance of the database. I have with me are:

    1. backup of the SPFile. (O1_MF_NCNNF_CTLFILE_BU_92DVGYO2_. BKP)

    2. backup of the control file (O1_MF_NNSNF_SPFILE_BU_92DVGW9H_. BKP)

    3. the full backup (O1_MF_NNSNF_FULLPROD_BU_92DVGW9H_. BKP)

    I received these files to a client environment and I need to restore these to a new instance of the database on a local computer. I gave an attempt as follows (in Oracle 10 g):

    1, creates a new instance of the (Test75) database using the Oracle Database Configuration Assistant

    2 SET ORACLE_SID = Test75

    3 connection to this instance with RMAN

    4 tried to restore the SPFILE from the BKP file received.


    I make mistakes and cannot continue. I don't know if I will carry out the correct steps. Please help me telling me how I should achieve this goal.


    Thank you.

    1. backup of the SPFile. (O1_MF_NCNNF_CTLFILE_BU_92DVGYO2_. BKP)

    2 backup of control file (O1_MF_NNSNF_SPFILE_BU_92DVGW9H_. BKP)

    The first backuppiece resembles a ControlFile and the other looks like a SPFILE'

    > I get error and cannot continue

    What is the error?  (I assume that you have tried to restore the O1_MF_NNSNF_SPFILE_BU_92DVGW9H_ SPFILE. BKP

    Note that you must also SET the DBID before restoring the controlfile

    Hemant K Collette

  • case of disaster with backup rman and archivelogs

    Salvation;

    When the disk failure comes and drop everything on disks, if we have a cold backup and logs archiving ONLY after a cold backup up to disk failure, we can restore files (datafiles, controlfile and everything) and apply to these cold backup archive logs and make the database updated up to the failure of disks.

    I had this case in real life, and it works.


    but

    What happens if you don't have backup rman online that was taken before the failure of disk instead of cold backup and archive logs that was created after the backup online up to the failure of the drive.


    We can restore the online of rman backup files, but can we apply archive logs to these restored files to update the database to the disk failure?



    Best regards

    Use the SET clause UNTIL - for example the VALUE up to TIME - before running the command RECOVER the DATABASE.

    See http://docs.oracle.com/cd/E11882_01/backup.112/e10643/rcmsubcl019.htm#i1000331

    and example 4-38 on the same page.

    Hemant K Collette

Maybe you are looking for