Refresh database through exp/imp

version 10203 on windows (old db on solaris 8170)
i Have to refresh database data, No of users/schemas are 400+. fastest way to do that would be to do full exp/imp.
but 1st drop current users cascade. (any command to drop all users in with one command?)
then validate all all tables/schemas are same & up-to-date( any suggestion to validate this efficiently?) i am thinking to check full exp logs on old & new db.(what that can take forever manually going through thousands of tables etc)

First, make sure that statistics are properly gathered in both the old and the new DB.
Next, use num_rows column in dba_tables. Since statistics are typically gathered by sampling, they would not necessarily match. They need to be quite close though.
After you are sure that all objects were transferred, you can issue a query to find all tables with “invalid” number of records.
The query can look something like this:
select s.owner, d.table_name , d.num_rows
from dba_tables d
where d.owner not in (‘SYS’,’SYSTEM’,…)
and not exists
           (select * from dba_tables@old_db s
          Where d.owner = s.owner
          And d.table_name = s.table_name
          And s.num_rows between 0.9*d.num_rows and 1.1*d.num_rows
          )You need to take care of some special cases, such as num_rows being NULL, partitioned tables, etc.
Iordan Iotzov
http://iiotzov.wordpress.com/

Similar Messages

  • Migrating from 9i to 10g through exp/imp

    Hi,
    We need to migrate a database on 9i in HP-UNIX to 10g in IBM AIX. Can you please tell me whether i can export data in 9i with 9i export utility and import it into 10g using 10g data pump utility for faster import ?
    We don't have the 10g software installed in the HP-UNIX platform.
    And what other alternatives we have to reduce the amount of time for the migration process as it's a critical Production database ?
    Thanks,
    Jayanta

    I believe that Justin is correct that if you use exp to unload the data you will need to use imp and not impdp to reload the data.
    To speed up the cross-platform exp/imp process I suggest you consider running multiple concurrent exp/imp jobs. You can generate a table=list into a spool file using sql. You can give each large table its own exp job and bunch the small tables.
    Depending on how your database objects are organized you might also be albe to export by owner or a combination of owner, tables= exports.
    You can use a full=y with the rows=n option to grab the public synonyms, non-owning users, packages, etc.... not brought in by the prior exp/imp.
    HTH -- Mark D Powell --

  • Refresh using exp/imp

    Hi All,
    I need to refresh one of my development database using exp/imp utility,i did refreshing using the copy datafiles but new to this ,i have gone thru this forum and did some search on google too but the part that am not getting is suppose my development db already created with same tablespaces name and size and containes every schema as that of source and has data in it which is refreshed earlier, so do i need to wipe out everything and do it right from scratch(create db,tablespace,users) or just wipe out tables and import with (ignore=Y) .Any feedback would be apperciated.
    Thanks

    <p class="MsoNormal">Hi,</p>
    <p class="MsoNormal"> </p>
    <p class="MsoNormal">>> do i need to wipe out everything and do it right from
    scratch(create db,tablespace,users)</p>
    <p class="MsoNormal"> </p>
    <p class="MsoNormal">Well you have choices, and you can do whatever you want,
    but there are some “best practices”.  I’m assuming that you are copying your
    PROD database into DEV (a great practice to have a full-sized production system
    the TEST and DEV):</p>
    <p class="MsoNormal"> </p>
    <p class="MsoNormal">- You can restore PROD into TEST using RMAN.</p>
    <p class="MsoNormal"> </p>
    <p class="MsoNormal">- You can use the
    <a style="color: blue; text-decoration: underline; text-underline: single" href="http://download-east.oracle.com/docs/html/B16227_02/oui7_cloning.htm">
    Oracle Universal Installer</a> to clone a database.</p>
    <p class="MsoNormal"> </p>
    <p class="MsoNormal">- You can “<a style="color: blue; text-decoration: underline; text-underline: single" href="oracle_tips_db_copy.htm">clone
    the whole database</a>”, moving the current datafiles and recreating the
    instance</p>
    <p class="MsoNormal"> </p>
    <p class="MsoNormal">- You can export from PROD, truncate the tables in DEV
    (after a backup) and import IGNORE=Y.  This can be very slow, especially for a
    large database.  However, there are some
    <a style="color: blue; text-decoration: underline; text-underline: single" href="http://asktom.oracle.com/pls/ask/f?p=4950:8:::::F4950_P8_DISPLAYID:1240595435323">
    tips to speed-up imports</a>.</p>
    <p class="MsoNormal"> </p>
    <p class="MsoNormal">Check the
    <a style="color: blue; text-decoration: underline; text-underline: single" href="http://search.oracle.com/">
    Oracle docs</a>.  Also, Dave Moore has a book on
    <a style="color: blue; text-decoration: underline; text-underline: single" href="http://www.amazon.com/Oracle-Utilities-Programs-Oradebug-Dbverify/dp/0972751351">
    Oracle Utilities</a> that you might like.</p>
    Hope this helps . . . .
    Don Burleson
    <p>
    www.dba-oracle.com

  • Database refresh of a 10gb database through export-import

    hi ,
    I am using oracle 10g R2 . I need to refresh the development server from the production server .
    The size of the database is 10GB . Can I do it with export-import ?
    Thank you .

    besides these mentioned methods, i.e. 1) exp/imp; 2) data pump; 3) transportable tablespaces; 4) rman duplicate. you can also use manual duplication of the production database which is faster and easier I would say. Below is link, which can be used for detailed steps.
    http://www.dba-oracle.com/oracle_tips_db_copy.htm
    Please note that, if your source database can not be shut down, so you can copy your datafiles by putting tablespaces or database into hot backup mode.

  • Full Database Exp & Imp

    Hi,
    I am trying to Exp & Imp a full database. I am working on Oracle 9i & 10g on Solaris 9. I am not using Data Pump. Can anyone please help me with the following:
    1. I am performing the full export using SYSTEM user.
    1.a Does a Full export include (or backups up) the DATA DICTIONARY of the database?
    1.b Does a Full export include the backup of SYS and SYSTEM objects?
    I am using the following command to export
    exp system/system@testdb file=$HOME/testdbfullexp.dmp full=y statistics=none
    I have tried importing the FULL export to another database and i did see that SYS and SYSTEM objects were also being imported ( got some errors regarding constraints and inconsistencies).
    I would like to ask like what are the ideal steps to follow to copy a database from DB1 to DB2 using EXP and IMP
    Any information will be of a great help
    Thanks,
    Harris.

    1 a) No, as the data dictionary will be automagicall recreated by implicit SQL.
    This means any non dictionary objects under SYS will be lost
    1 b) as above. SYSTEM however is a normal user.
    any %SYS user will NOT be exported (CTXSYS, MDSYS, etc)
    On import of SYSTEM there will be always errors, as SYSTEM is non-empty after initial database creation.
    Sybrand Bakker
    Senior Oracle DBA

  • Full database exp/imp  between RAC  and single database

    Hi Experts,
    we have a RAC database oracle 10GR2 with 4 node in linux. i try to duplicate rac database into single instance window database.
    there are same version both database. during importing, I need to create 4 undo tablespace to keep imp processing.
    How to keep one undo tablespace in single instance database?
    any experience of exp/imp RAC database into single instance database to share with me?
    Thanks
    Jim
    Edited by: user589812 on Nov 13, 2009 10:35 AM

    JIm,
    I also want to know can we add the exclude=tablespace on the impdp command for full database exp/imp?You can't use exclude=tablespace on exp/imp. It is for datapump expdp/impdp only.
    I am very insteresting in your recommadition.
    But for a full database impdp, how to exclude a table during full database imp? May I have a example for this case?
    I used a expdp for full database exp. but I got a exp error in expdp log as ORA-31679: Table data object "SALE"."TOAD_PLAN_TABLE" has long columns, and longs can not >be loaded/unloaded using a network linkHaving long columns in a table means that it can't be exported/imported over a network link. To exclude this, you can use the exclude expression:
    expdp user/password exclude=TABLE:"= 'SALES'" ...
    This will exclude all tables named sales. If you have that table in schema scott and then in schema blake, it will exclude both of them. The error that you are getting is not a fatal error, but that table will not be exported/imported.
    the final message as
    Master table "SYSTEM"."SYS_EXPORT_FULL_01" successfully loaded/unloaded
    Dump file set for SYSTEM.SYS_EXPORT_FULL_01 is:
    F:\ORACLEBACKUP\SALEFULL091113.DMP
    Job "SYSTEM"."SYS_EXPORT_FULL_01" completed with 1 error(s) at 16:50:26Yes, the fact that it did not export one table does not make the job fail, it will continue on exporting all other objects.
    . I drop database that gerenated a expdp dump file.
    and recreate blank database and then impdp again.
    But I got lots of error as
    ORA-39151: Table "SYSMAN"."MGMT_ARU_OUI_COMPONENTS" exists. All dependent metadata and data will be skipped due to table_exists_action of skip
    ORA-39151: Table "SYSMAN"."MGMT_BUG_ADVISORY" exists. All dependent metadata and data will be skipped due to table_exists_action of skip
    ......ORA-31684: Object type TYPE_BODY:"SYSMAN"."MGMT_THRESHOLD" already exists
    ORA-39111: Dependent object type TRIGGER:"SYSMAN"."SEV_ANNOTATION_INSERT_TR" skipped, base object type VIEW:"SYSMAN"."MGMT_SEVERITY_ANNOTATION" >already exists
    and last line as
    Job "SYSTEM"."SYS_IMPORT_FULL_01" completed with 2581 error(s) at 11:54:57Yes, even though you think you have an empty database, if you have installed any apps or anything, it may create tables that could exist in your dumpfile. If you know that you want the tables from the dumpfile and not the existing ones in the database, then you can use this on the impdp command:
    impdp user/password table_exists_action=replace ...
    If a table that is being imported exists, DataPump will detect this, drop the table, then create the table. Then all of the dependent objects will be created. If you don't then the table and all of it's dependent objects will be skipped, (which is the default).
    There are 4 options with table_exists_action
    replace - I described above
    skip - default, means skip the table and dependent objects like indexes, index statistics, table statistics, etc
    append - keep the existing table and append the data to it, but skip dependent objects
    truncate - truncate the existing table and add the data from the dumpfile, but skip dependent objects.
    Hope this helps.
    Dean

  • Using exp imp on 11i apps database

    Hi,
    Does anyone have experience in using exp and imp utility on 11i apps database? Is there any special method to achieve this.
    I want to do a full export and do an import.
    Do we get any performance benefits if we do this?
    Regards,
    SA

    Hi Sa;
    Does anyone have experience in using exp and imp utility on 11i apps database? Is there any special method to achieve this. You want to take exp imp for which issue?
    I want to do a full export and do an import.Follow:
    10g Export/Import Process for Oracle Applications Release 11i [ID 331221.1]
    Export/import process for R12 using 11gR1 [ID 741818.1]
    Do we get any performance benefits if we do this?It depends many segment but finaly u will make one process on server and it has cost,answer you can have some performance issue while you are taking exp-imp
    Regard
    Helios

  • Schema Refresh using exp/imp.

    Hello All,
    I want to perform Schema Refresh of SAMPLE user from producation to Testing envrionment using export/import.
    Cud u plz tell what is the command to perform it ?
    Also Cud anyone plz tell me whether same user(SAMPLE) in Test environment gets dropped before Import done.
    Can i Perform the exp/imp using sys/system or user SAMPLE?

    tvenkatesh07 wrote:
    Hello All,
    I want to perform Schema Refresh of SAMPLE user from producation to Testing envrionment using export/import.
    Cud u plz tell what is the command to perform it ?
    Also Cud anyone plz tell me whether same user(SAMPLE) in Test environment gets dropped before Import done.
    Can i Perform the exp/imp using sys/system or user SAMPLE?If you're runnnig 10g, then use Data Pump and read the documentation:
    http://download.oracle.com/docs/cd/B19306_01/server.102/b14215/dp_export.htm#i1007466
    http://download.oracle.com/docs/cd/B19306_01/server.102/b14215/dp_import.htm#i1007653
    My Oracle Video Tutorials - http://kamranagayev.wordpress.com/oracle-video-tutorials/

  • Exp/Import database schema vs Ch 10 Exp/Imp Content

    Hi I'm using Portal 10.1.2.02. I'm a DBA tasked with migrating a Dev portal into Test, then Production.
    There seems to be a number of ways to achieve this and I was hoping to clarify which is best for me.
    I've run through note: 330391.1(copying the Portal schema) which details running a perl script to export and import onto a newly installed server. This process worked will.
    Chapter 10 of the Portal guide details a fairly complex process of creating transports sets and migrating these over and importing.
    My question is: if I want the entire Portal copied is there any difference in these processes? ie. Do you end up with the same result?
    thanks in advance of any advice :-)

    The cloning model is not a rerunnable model, its not granular and conditional migraton is not possible.
    You can do the cloning only if u want to take a copy of the entire setup and rewire to a new midtier.
    Portal exp/imp model helps u to achieve granular, conditional and rerunnable method of moving portal objects.
    Apart from that, it comes with the readily available prechecks to intimate what's going wrong during the process.

  • Exp/imp for upgrading database

    Hi,   Is it a good idea to use exp / imp to upgrade a database from 9i to 11g.To overcome parallel I plan to set ll tables degree to 8 manually and then start the export.Which is better conventional upgrade or exp/imp

    HI,
    Please check Oracle support id : How to Perform a Full Database Export Import during Upgrade, Migrate, Copy, or Move of a Database (Doc ID 286775.1)
    Thank you

  • Using exp/imp to clone a database from Solaris to NT

    Hi All,
    We have an 8.1.6 Oracle database in production under Sun Solaris.
    We actually use this database to run Siebel CRM system.
    We have a Windows NT server which we used as a QA system
    for this application.
    We have a daily full export of our production database on the
    Sun machine. Now my manager wants to import this production
    dmp file into this Win NT database on a weekly basis.
    The goal here is to maintain the state of this NT database
    as much functional and close to the production one as possible.
    I would appreciate your help on this.
    I have some other general question:
    in which case can the Oracle imp utility be run in the
    full=y mode ?
    does this make any sense at all ?
    since the Oracle import utility does not delete
    any existing objects before it imports, these objects
    should be deleted manually and that's what I do
    with all Siebel users schemas; of course I do not
    do this with SYS, SYSTEM and other special Oracle
    users; question: are there any objects in these
    special Oracle schemas that have to be carried over from
    the source database to the target database to
    make the rest of the schemas functional ?
    Somehow I failed to find Oracle documentation on how
    to use exp/imp to create a complete clone of an Oracle
    database.
    Can direct path be used with full=y export/import mode ?
    Thank you in advance,
    Yuriy.

    As an alternative to using daily/weekly exports and imports, what would probably be more effective here would be replication. The databases could synch on a nightly basis and updated tables and rows will be transferred from your OLTP system to your read-only system. More information can be found on replication from "Oracle8i DBA Handbook" by Kevin Loney and Marlene Therialt.
    LM
    PS. IGNORE=Y

  • Step by Step EXP / IMP database from Oracle 9i to Oracle 11g

    Please help on the captioned where I did the below
    1. Use EXP to export the database from a Oracle9i database (O9_DB) OLD_Windows2003_SERVER
    2. Install the Oracle 11g (default service name is ORCL) to a new Windows 2003 server and use the IMP to import the O9_DB, however, I need to create the same physical path as the OLD_Windows2003_SERVER (e.g E:\Oralce_9i\Ordata\) in order to complete the IMP
    3. I can launch the SQL Developer and obtain the records count of evey tables being imported
    However, is there any best / proper way to do the EXP/IMP, the ideal outcome would be
    1. Create a new namespace say (OLDDB) which same as the Oracle9i database name so maybe I don't need to modify much on the application side to point to the new database ?
    2. IMP the OLDDB to the new namespace OLDDB
    Thanks in advance

    Hello,
    If you create the new Database on a new Server, you could keep the same name for the Database.
    To switch from the old Database to the new one, you'll have to change the hostname on the "tnsnames.ora" or in any configuration file which define the database connexion.
    About the best way to do the Export/Import, you may have some details on the link below:
    http://download.oracle.com/docs/cd/E11882_01/server.112/e10819/expimp.htm#i262317
    Else, why do you intend to stay on a Windows 2003 Server, why not moving to Windows 2008 ?
    NB: Oracle *11.1* is certified on Windows 2008 and Oracle *11.2* is certified on Windows 2008 R2.
    Hope this help.
    Best regards,
    Jean-Valentin

  • Database upgrade from 8i to 10g using exp/imp

    Dear Friends,
    Please provide the steps to upgrade from 8i to 10g using exp/imp.
    Thanks,
    Rathinavel

    Hi;
    Please also see cold backup option
    How to migrate from 8i to 10g to new server using cold backup [ID 742108.1]
    Also see:
    Upgrading from 8i to 10g with import utility
    http://searchoracle.techtarget.com/answer/Upgrading-from-8i-to-10g-with-import-utility
    Regard
    Helios

  • Exp/Imp 10.2.0.1 Database

    Can I export data from a database with 8k block size onto a target database with 16k block size?

    it doesn't matter what blocksize your target database has configured.
    imp generates simple insert statements for each row. you do not need an db_8k_cache_size

  • Exp/Imp alternatives for large amounts of data (30GB)

    Hi,
    I've come into a new role where various test database are to be 'refreshed' each night with cleansed copies of production data. They have been using the Imp/Exp utilities with 10g R2. The export process is ok, but what's killing us is the time it takes to transfer..unzip...and import 32GB .dmp files. I'm looking for suggestions on what we can do to reduce these times. Currently the import takes 4 to 5 hours.
    I haven't used datapump, but I've heard it doesn't offer much benefit when it comes to saving time over the old imp/exp utilities. Are 'Transportable Tablespaces' the next logical solution? I've been reading up on them and could start prototyping/testing the process next week. What else is in Oracle's toolbox I should be considering?
    Thanks
    brian

    Hi,
    I haven't used datapump, but I've heard it doesn't offer much benefit when it comes to saving time over the old imp/exp utilitiesDatapump will be faster for a couple of reasons. It uses direct path to unload the data. DataPump also supports parallel processes, so while one process is exporting metadata, the other processes can be exporting the data. In 11, you can also compress the dumpfiles as you are exporting. (Both data and metadata compression is available in 11, I think metadata compression is available in 10.2). This will remove your zip step.
    As far as transportable tablespace, yes, this is an option. There are some requirements, but if it works for you, all you will be exporting will be the metadata and no data. The data is copied from the source to the target by way of datafiles. One of the biggest requirements is that the tablespaces need to be read only while the export job is running. This is true for both exp/imp and expdp/impdp.

Maybe you are looking for

  • Invalid Node Structure on Time Capsule.

    Hi, I was using my time capsule (7 months old) as a temporary network drive (Until I get a new network drive) for my iTunes library. And, no, currently this is my only computer without a time machine backup. One day last week, I was listening to some

  • Scroll bar

    I borrowed some code to make a functioning scrollbar for a gallery thumbnail section. I want to tweak its function so that the content that is being moved by dragging the scrollbar slows to a stop rather than abruptly stops once the scrollbar draggin

  • Browser.places.migratePostDataAnnotations : False

    Dear Sir Since last week Oct 23, 2010 I am having problem using tabs on different web sites, for example. 1. After entering my Bank's web site I am not able to use the tab that opens the log in page (on opening Help/Troubleshooting I received the mas

  • ITunes Match Songs Keep Starting Over

    I've been using iTunes Match since it first released and absolutely love it! However, I've been experiencing a strange thing the last couple of months and it's happening more frequently. I haven't changed any settings or anything - it's an issue that

  • TRFC server

    Hi, I am writing a tRFC server using SAP NW RFC SDK. I am registering my server using RfcRegisterServer() and then i am trying to get the function description of INBOUND_IDOC_PROCESS whose handle i can send to RfcInstallServerFunction().. Below is th