Unable to export specific tables in the schema.

Similar Questions

• avoid triggers on all tables in the schema

  I need to put a CREATION TIMESTAMP to the timestamp of the database server (not the timestamp of session) in all tables in a schema to create and update. Is to create a trigger on all the tables in the schema a less time consuming way to do?
  
  Similarly, I need to set up columns such as CREATE_USER, LAST_UPDATE_USER.
  
  Thank you in advance.

  You can easily generate the DDL for adding new columns.

  The extent of the filling of the columns, your choice is either to use insertion befire table and triggers to update to fill the columns or the application to provide the necessary information.

  Basic trigger logic would be pretty much the same for all tables, then write a little SQL or PL/SQL to generate the code of the trigger should be simple enough.

  Depending on your application, such as web based with only one Oracle user, you may need to get the real user through dbms_application_info by the logic of the server-based application.

  HTH - Mark D Powell.

  Edited by: Mark D Powell 5 May 2010 07:48

• Data Pump: export/import tables in different schemas

  Hi all

  I use Oracle 11.2 and I would like to use the data pump to export / import data into the tables between the different schemas. The table already exists in the source and target schemas. Here are the commands that I use:

  
  Script working export:

  expdp scott/tiger@db12 schema = source include = TABLE:------"IN (\'TB_TEST1 ', \'TB_ABC')\ 'directory = datapump_dir dumpfile = test.dump logfile = test_exp.log

  Script to import all the desks:

  Impdp scott/tiger@db12 remap_schemas = rΘpertoire source: target = datapump_dir dumpfile = test.dump logfile = test_imp.log content = data_only table_exists_action = truncate

  Script error for some tables to import:

  Impdp scott/tiger@db12 remap_schemas = source: target include = TABLE:------"IN (\'TB_TEST1')\' directory = datapump_dir dumpfile = test.dump logfile = test_imp.log content = data_only

  Export is good, I got the folling error when I try to import the table TB_TEST1 only: "ORA-31655: no metadata data_or objectsz selected for employment. The user scott is a DBA role, and it works fine when I try to import all tables being export without the include clause.

  It is possble to import some tables but not all tables in the export file?

  Thanks for the help!

  942572 wrote:

  It's good to import all tables exported through Scott, who do NOT have the clause 'include '.

  I have the error only when I try to import some tables with clause "include".

  Can I import only some table since the export dump file? Thank you!

  you use INCLUDE incorrectly!

  do below yourself

  Impdp help = yes

  INCLUDE

  Include the specific object types.

  For example, INCLUDE TABLE_DATA =.

• Update of the specific data of the schema in the remote databases.

  I have same application running in places different, each with its own database (gr 10, 2). I want to find solutions to how can I change each database with the data in the other place so that my users of the application can check the status & reports based on data from other sites. There is no physical connectivity between databases in the form of network LAN or WAN.
  
  The only option, in my view, is to take backup at one location and updated at another location. But the problem is that the names of the diagrams and tables are the same on all sites, that is, the database is designed for Central connectivity to all locations. But after the design of the application, we are forced to use databases distributed without any connectivity between sites.
  
  We cannot use direct import/export or impdb/expdb because the schema name and the tables are the same in all locations.
  
  Is it possible to record/track DML statements on my specific sites with timestamp scheme so that these DML can be used as a source of data import. There are a lot of tables in my diagram so I need a solution more widespread rather than write triggers on each table.
  
  I also checked Oracle Audit and I want some sort of solution so that I can manage without the support of DBA in each location.

  If I reframe your question will be it seems like below:

  User: UserA
  Database: Db1
  Location: Loc1

  User: UserA
  Database: Db1
  Location: Loc2

  You want to synchronize/keep the latest information in databases without:

  1. connectivity.
  2.Import/Export
  3 oracle DBA

  Right?

  If Yes, then
  DML trigger will fill a new table in the two locations, that will all the new values that a user made on one site. Now in the other place, this table will be sent to run a pl/sql block using this table and fill in the data.

  On the other
  What is your question, we need more details.
  End if

  Concerning
  Girish Sharma

• DataPump - expdp.open creates tables in the schema

  Hello

  I use datapump in Oracle 10 g to archive old main schema to another schema partitions.

  I noticed that when dbms_datapump.open is called and a new table is created by dbms_datadpump for internal purposes. This is verified in the oracle documentation

  http://docs.Oracle.com/CD/B12037_01/AppDev.101/b10802/d_datpmp.htm#997383

  Notes on use

  • When the task is created, a master table is created for work under the scheme of the appellant in the default tablespace of the appellant. A handful referring to employment is returned that attaches to the current session to use. Once attached, the handle remains valid until that either explicit or implicit detach occurs. The handle is only valid in the session of the appellant. Other handles can be attached to the same task from another session using the ATTACH procedure.

  Does anyone know if this table can be removed by a call to dbms_datapump Oracle 'clean', or if it has to be cleaned manually.

  can confirm that this is what you do

  v_job_handle: = DBMS_DATAPUMP. OPEN('EXPORT', 'TABLE', NULL, v_job_name);

  -The parallelism 1 value and add the file

  DBMS_DATAPUMP. SET_PARALLEL (v_job_handle, 1);

  DBMS_DATAPUMP. ADD_FILE(v_job_handle, v_job_name |) '_' || v_partition. PARTITION_NAME | ".dmp", "PARTITION_DUMPS");

  -Apply filters to process only a partition in the table

  DBMS_DATAPUMP. METADATA_FILTER(v_job_handle, 'SCHEMA_EXPR', 'IN ("SIS_MAIN")');

  DBMS_DATAPUMP. METADATA_FILTER(v_job_handle, 'NAME_EXPR', ' AS "' | t_archive_list (i) |) '''');

  DBMS_DATAPUMP. DATA_FILTER(v_job_handle, 'PARTITION_EXPR', ' IN ("' | v_partition.partition_name |)) ')', t_archive_list (i), "SIS_MAIN");

  -Use statistics (rather than blocks) to estimate the time.

  DBMS_DATAPUMP. SET_PARAMETER(v_job_handle, 'ESTIMATE', 'STATISTICS');

  -Start the work. An exception is returned if something is not installed correctly.

  DBMS_DATAPUMP. START_JOB(v_job_handle);

  -The export job should work now. We loop until its finished

  v_percent: = 0;

  v_job_state: = "UNDEFINED";

  WHILE (v_job_state! = 'DONE') and (v_job_state! = "STOPPED") LOOP

  DBMS_DATAPUMP.get_Status (v_job_handle,DBMS_DATAPUMP.ku$_status_job_error + DBMS_DATAPUMP.ku$_status_job_status + DBMS_DATAPUMP.ku$_status_wip,-1, v_job_state, m);

  js: = sts.job_status;

  -As percentage-complete in this loop changes, displays the new value.

  IF js.percent_done! = v_percent THEN

  v_percent: = js.percent_done;

  END IF;

  END LOOP;

  -Once the work is finished, view the status before you loose work.

  PRC_LOG (f1, t_archive_list (i) |) ': Export complete with status: '. v_job_state);

  -DBMS_DATAPUMP. Detach (v_job_handle);

  -Use STOP_JOB instead of DETACHING otherwise 'master' table that is created when the OPEN is called is not removed.

  DBMS_DATAPUMP. STOP_JOB(v_job_handle, 0, 0);

• Creating table in the schema has with the user B and granting permissions to the user C

  Hello, I have a problem fun - we have a large table which requires a batch to make millions of updates, and he needs to finish more quickly. It takes several hours as an update, but a select create table did the same thing in about a minute. Yay! Then we just delete the old table and rename the new table to the former, rebuild all the index and grant select/insert/update/delete permissions to another user who needs access... and there is the problem, Oracle said insufficient privs.
  
  We have A figure, which is the owner of the table, userid B who done all the stuff of application batch and userid C which performs inserts for a process of middleware. None of them are actual users, of course. We want that user B to do the ETG, rename, rebuild and re-grant of the authorisation of the index. But user B apparently cannot grant permissions on objects in the schema A to user C, even if B has created the table and has the role DBA (not ideal, I know)!
  
  What's really crazy is that there is no sense unique user b can grant permissions to user C, which is to:
  
  grant select any table USER c;
  grant insert any table USER c;
  grant update any table USER c;
  Grant delete any table USER c;
  
  It seems really perverse can create tables in diagram A and even grant C privs "whole table" that user B, but may not grant privs specifically on this object in the schema A, even with DBA privs. I must be missing something... right?

  Yes. By default, the stored procedures are stored procedures from DEFINER rights. If A is the owner of a stored procedure, this stored procedure can do whatever A enjoys the privileges to directly. If A grants B privileges RUN this stored procedure, when B calls the stored procedure, the procedure runs with the privileges.

  I agree point stew, however, re-creating objects in Oracle is generally a bad idea and a TRUNCATE with one insert direct-path, possibly combined with the deactivation and the rebuilding of the index would be more appropriate than a DEC. If you're going to stick with this Assignment, however, that really do in a stored procedure owned by A while B has no need to CREATE ANY TABLE.

  Justin

• Select all the tables in the schema

  Hi, I'm looking for x, y of the tables in a schema, get the error "missing expression," working with oracle 11g.
  
  
  declare
  
  v_sql varchar2 (4000);
  number of v_x;
  number of v_y;
  number of V_n;
  
  Start
  
  for rec in loop (select table_name as table_name from all_tables where table_name like '% oumou %' ORDER BY 1)
  
  
  v_sql: = ' select a.idnumber, t.x, t.y, table (sdo_util.getvertices (a.geometry)) t OF ' | Rec.table_name | « a » ;
  V_sql EXECUTE IMMEDIATE v_n, v_x, v_y;
  dbms_output.put_line(V_n|| v_x || v_y);
  
  
  end loop;
  
  
  end;

  ...
    --only for fetching implizit CURSOR cur1
    CURSOR t2 IS SELECT CAST('' AS VARCHAR2(20))idnumber, CAST(0 AS NUMBER) x, CAST(0 AS NUMBER) y FROM dual;
    rec2   t2%ROWTYPE;
  ...
  

  To the implicit CURSOR cur1, you can set a variable-FETCH with CUR1% ROWTYPE.
  So, I defined an explicit CURSOR t2, bringing me the types for the columns of results:
  IDNUMBER (VARCHAR2 (20)), X (NUMBER) and Y (NUMBER)
  This T2 of cursor, I can set a % ROWTYPE VARIABLE t2 to fetch the results.

  Please give points for the solution.

• How to delete all TABLEs in the schema SYS that are created from 09:15?

  Unfortunately a script created a lot of tables in the diagram (= SYS) and bad Tablespace (= SYSTEM).
  
  How can I delete all TABLES that are created inTablespace = SYSTEM and SCHEMA = SYS (in a DDL command)
  respectively for the past 3 hours from 09:15 to 25 Sep 2011?
  
  Alternative: How can I move these tables in another schema (for example ATEST) and Tablespace (USERS)?
  Is this possible with Oracle XE or Oracle Enterprise?
  
  Peter

  You can use ALTER TABLE... MOVE. It offers a TABLESPACE clause, which allows you to choose a different tablespace as target. Refer to the documentation for your version of the database.

• querying table inside the schema of the IOM

  Hello
  
  I created a table looks in the scheme of the IOM. I want to insert data into this table custom through my scheduled task. As I try to insert data into a table that is located in the schema of the IOM itself, I hope I don't have to make a JDBC connection. Could you get it someone please let me know how can I go about it?
  
  Thank you
  PETREA

  You can use the statement prepared for her and use tcDataProvider (IOM) to make connections.

• get information from inside a trigger of the other tables in the schema even

  Addendum:
  
  ... Sorry I forgot: Oracle Database 10g Enterprise Edition Release 10.2.0.4.0 - 64 bit
  
  *******************
  
  Hello
  
  I'm trying for a few hours to get information from inside a trigger to another table in the same pattern.
  
  My trigger is PSE_BKB. NUM_PHANTOM_BP
  
  I first tried a simple solution...
  
  
  CREATE OR REPLACE TRIGGER PSE_BKB. NUM_TR_PHANTOM_BP
  BEFORE DELETING, INSERTION OR UPDATE
  ON PSE_BKB. NUM_PHANTOM_BP
  REFERRING AGAIN AS NINE OLD AND OLD
  FOR EACH LINE
  DECLARE
  
  v_recht NUMBER (1): = 0;
  v_recht2 NUMBER (1): = 0;
  v_change_mitarbeiter NUMBER (1): = 0;
  v_recht_mitarbeiter NUMBER (1): = 0;
  v_typ varchar2 (20);
  v_obj_key varchar2 (11): = "gugus";
  v_ph_key varchar2 (11);
  -ph_key varchar2 (11);
  v_obj_keys varchar2 (4000): = ', ';
  
  anz_num NUMBER: = 0;
  anz_num_tmp NUMBER: = 0;
  anz_st_si NUMBER: = 0;
  
  
  BEGIN
  
  ...
  Select count (*)
  in anz_num
  of pse_bkb.num_objekt o
  where o.obj_key in
  (select distinct (ov.obj_key)
  of pse_bkb.num_objekt_verb ov
  where ov.phantom_key =: old.phantom_key
  )
  and o.typ = "NUM";
  
  ...
  
  Explanation: my trigger table is PSE_BKB. NUM_PHANTOM_BP.
  
  On the pse_bkb.num_objekt_verb of the table, I have a relationship 1: n to one or several object-key (field named obj_key!) on my referenced key: old.phantom_key. With this obj_key, I want to watch the weather, I have one or more lines in the table type num_objekt "NUM". So the internal selection gives me all the num_objekt_verb obj_key with my: old.phantom_key and with whom I count the number of lines I have on num_objekt with type "NUM".
  
  The problem is that the trigger does not see the table.
  
  I tried many variations.
  
  The latest version I tried was with a cursor inside the trigger definition as in the block of code below.
  
  For debugging purposes, I inserted a RAISE_APPLICATION_ERROR in the inner loop - see below. The v_obj_key variable is never defined, as in all the other variations, I tried-, I still see the predefined "gugus" in the declare section.
  
  It seems that oracle can not read other tables at this point. La: old.phantom_key is defined (in this simple example there would come a return obj_key).
  
  Thanks in advance for your help - and excuse my English.
  
  Trigger code:
  
  CREATE OR REPLACE TRIGGER PSE_BKB. NUM_TR_PHANTOM_BP
  BEFORE DELETING, INSERTION OR UPDATE
  ON PSE_BKB. NUM_PHANTOM_BP
  REFERRING AGAIN AS NINE OLD AND OLD
  FOR EACH LINE
  DECLARE
  
  v_recht NUMBER (1): = 0;
  v_recht2 NUMBER (1): = 0;
  v_change_mitarbeiter NUMBER (1): = 0;
  v_recht_mitarbeiter NUMBER (1): = 0;
  v_typ varchar2 (20);
  v_obj_key varchar2 (11): = "gugus";
  v_ph_key varchar2 (11);
  -ph_key varchar2 (11);
  v_obj_keys varchar2 (4000): = ', ';
  
  anz_num NUMBER: = 0;
  anz_num_tmp NUMBER: = 0;
  anz_st_si NUMBER: = 0;
  
  
  BEGIN
  
  
  Start
  declare
  
  cursor nums is
  
  Select obj_key
  of PSE_BKB.num_objekt_verb
  where phantom_key =: old.phantom_key;
  
  Start
  
  for the RC looping nums
  v_obj_key: = cr.obj_key;
  
  RAISE_APPLICATION_ERROR (-20099,' AGAIN 9 v_obj_key: ' | v_obj_key |) ': old.phantom_key ' | (: old.phantom_key);
  
  When the exit nums % NOTFOUND;
  
  Select count (*) in the pse_bkb.num_objekt anz_num_tmp where obj_key = v_obj_key and type = "NUM";
  anz_num: = anz_num + anz_num_tmp;
  end loop;
  
  end;
  end;
  
  Published by: user832075 on 14.01.2013 06:18

  The reason why you get count_n = 0 in tr_c is that there is no line in B with key_c = 17, to the moment you arrive at tr_c. The removal of B has already occurred, and since you're in a single transaction, the application for tr_c:

  select count(*)
    into   counter_n
    from   a aa
    where  aa.key_a in (select distinct(bb.key_a)
                        from   b bb
                        where  bb.key_c = v_ph_key
                       )
    and    aa.typ = 'n';
  

  'knows' that the row has been deleted then gets a number of 0.

  I added a set of instructions dbms_output to show the order in which triggers run, and that's what I get:

  SQL> delete from b_vw where key_a = 1 and key_c = 17 ;
  in trigger b_tr_vw
  In tr_b
  in tr_c
  delete from b_vw where key_a = 1 and key_c = 17
              *
  ERROR at line 1:
  ORA-20077: Remark: counter_n: 0 counter_s: 0 v_ph_key: 17 old.key_c: 17
  new.key_c:
  ORA-06512: at "OPS$ORACLE.TR_C", line 38
  ORA-04088: error during execution of trigger 'OPS$ORACLE.TR_C'
  ORA-06512: at "OPS$ORACLE.B_TR_VW", line 17
  ORA-04088: error during execution of trigger 'OPS$ORACLE.B_TR_VW'
  

  Thus b_tr_vw trigger is triggered first and deletes a B. trigger fires tr_b as part of this statement to remove and check against A, happening apparently so no error is triggered and succeeds the removal of B. Now tr_b_vw takes back control and deletes C. Since the B line which has key_c = 17 already deleted a control for counter_n > 0 and counter_s = 0 fails and raise you the-20002 error. Note that I'm assuming that the error of-20077 you lift is just to see the values and is not intended to be part of your actual code.

  Because I have really no idea what you're trying to accomplish here I can't offer a lot of advice that you might want to try entering two tr_c of the tr_b_vw trigger validation queries before deliver you the removal of b and put the counter_n > 0 and counter_s = 0 control under tr_b_vw to decide whether or not you want to remove.

  John

• How to export a table and the table as an image?

  My VI generates a table (formatted with different colors of text and cell), plus a text box, as well as a table (formatting with a background color).

  I can group these objects on the front panel and copy them as a single unit, and paste them into PowerPoint.

  Is it possible to do this automatically, have labVIEW group together them as a single object and export it as an image file with the name of the file in the form of control? See the example attached VI and the image below:

  

  Here's a way to do it.

  1. create a decoration on the FP that surrounds the controls

  2 get a reference to the decoration and the MasterBoundRect property to define a subset of the image of the public Service

  3. save the subset of the image

  You can color the transparent decoration to hide it if you wish.

  If the method works for you, there are more complex and yet robust methods to get the reference of the decoration.

• How exp/tables in the schema hosted Apex 4.2

  Hi all

  I created a test and schema + tables hosted apex application:

  Oracle APEX 4.2 instance hosted by Oracle, https://apex.oracle.com/pls/apex/f?p=4550:1:0

  How I can exp or expdp schema or tables that I created here in it?

  Thank you very much

  MK

  MariaKarpa (MK) wrote:

  Do the images of data type and BLOB like (photos, jpeg, pdf), the recorded data can be extracted as well?  text or xml?

  It would be much faster to try this for yourself, rather than post it here.

  Values of BLOB columns can be unloaded as XML, but not in the form of text.

• Script to export specific elements of the VM inventory, by folder

  Hi all

  I was trawl through the discussions here... and I think this has already been done before, but maybe not in the same script... and I'm not much more than a writer.

  I'm looking to export to CSV, the following information for each virtual computer, but sorted by folder:

  Folder name, VM name, no. vCPUs, memory, HDD (GB) SCSI ID/Virtual Device Node, the size of the disk in GB (size provisioned)

  that is something that would be output similar to (assuming that 'Customer ABC' is the name of the folder):

  ABC's of customer, Server1, 4, 16, SCSI (0,0) 1, 60GB, a SCSI hard drive hard drive (0.1) 2, 20GB

  Customer ABC, Server2, 2, 8, SCSI (0,0) 1, 60GB, a SCSI hard drive hard drive (0.1) 2, a SCSI hard drive, 20GB (0.2) 3 250 GB

  Who is?  I don't know that lucd has something up his sleeve for this. Not too worried if the description of the hard drive is not exactly the above (to be honest, just the SCSI ID or "Hard disk 1' etc, it would be perfect). As each VM is likely to have a different number of hard drives, and I want that all information for each VM on the same line... It is probably best to put the info hard drive at the end.

  We have about 200 VMS, and it will grow - so I wonder if something with 'Get-View "?  I read everywhere and it seems that this method is fast enough.  Furthermore, a script where we set the variable of the specific file we want to question rather than in the object of data centre?

  Any help on this is much appreciated - let me know if the above makes no sense.

  Cheers, James

  All credit goes to Lucd and Alan Renouf.

  $Information = @()
  $Vminfo = Get-VM | Select-Object *, @{N="HDDs"; E={$_ | Get-HardDisk}}, @{N="HDDcount"; E={($_ | Get-HardDisk).length}}
  $vmlist = $Vminfo | Sort-Object -Descending HDDcount
  foreach ($vm in $vmlist) {
      #$Harddiskinfo = $vm | get-harddisk
      $props = New-Object PSObject
      $props | Add-member -name "Folder" -Membertype NoteProperty -Value $vm.folder
      $props | Add-member -name "VMName" -Membertype NoteProperty -Value $vm.Name
      $props | Add-member -name "vCPUs" -Membertype NoteProperty -Value $vm.NumCpu
      $props | Add-member -name "Memory(GB)" -Membertype NoteProperty -Value $vm.MemoryGB
      $DiskNum = 0
      foreach ($hdd in $vm.HDDs) {
          $controllerinfo = $hdd.ExtensionData
          $controllerkey = $controllerinfo.controllerkey.tostring()
          $keycount = ($controllerkey |  measure-object -Character | Select-Object -expandproperty Characters).tostring()
          $virtualID = "{0}:{1}:{2}" -f $controllerkey.substring(0,1), $controllerkey.substring($keycount - 1), $hdd.ExtensionData.Unitnumber
          $props | Add-member -name "Disk$($DiskNum)scsiID" -Membertype NoteProperty -Value $virtualID
          $props | Add-member -name "Disk$($DiskNum)Harddisk" -Membertype NoteProperty -Value $hdd.Name
          $props | Add-member -name "Disk$($DiskNum)Size(GB)" -Membertype NoteProperty -Value $hdd.CapacityGB
          $DiskNum++
      }
       $Information += $props
  }
  $Information | Out-GridView
  

  Reference URL

  http://www.virtu-al.NET/2009/11/17/PowerShell-gotcha-Export-CSV-out-GridView-and-others/

• Export a table for the flat file and must insert sysdate in flat file column

  Hi, I created an interface allowing to export an oracle table to a csv file. All the columns in the table work well. Then I need to insert the sysdate in a column in the csv file.

  I made the map as working in the transit area, implementation is to_char(sysdate,'dd/mm/yyyy'). But the result is insert in column 14.

  I tried to create a variable refreshing that choose double to_char(sysdate,'dd/mm/yyyy'), maps then as the csv file column, but he only insert 1 row and the format is YYYYMMDD.

  I tried to use SELECT '< % = odiRef.getSysDate ("YYYYMMDD") % >' double to the variable and it also only insert a line in the flat file.

  I used the same methodology in ODI10g, it works fine.

  So, I wonder how it can be implemented in 11g.

  Thank you

  It turns out that I used the snapshot_date as a field name, it may be a reserved word. After that I changed the name of the field, it works fine. Thank you.

• Export specific tables, with or without data

  Hello
  I am writing a script to export some data tables and other tables WITHOUT. I want a reproducible process (script, procedure, or function) if I don't want to use the Wizard SQL Developer. Can I just run a command EXP in SQL Developer? I'm running just a simple export with the following command but get an error:
  
  Start
  tables test/test@prod exp is TMP_TEST FILE=C:\Oracle\test.dmp LOG=C:\Oracle\test.log ROWS = N;.
  end;
  /
  
  Any help is greatly appreciated.
  Thank you
  John

  You can call from PL/SQL is the DBMS_DATAPUMP package. Here's the documentation: http://docs.oracle.com/cd/E11882_01/appdev.112/e25788/d_datpmp.htm