Syncronized database tables (different databases)

Hi,
I'm dealing with a problem as I have multiple databases (instances) that have common tables that need to be the same all the time for all the databases.
DML on these tables will be issued from one point usually but the main problem is that all databases won't be up and running all of the time.
This causes problems, because databases that would be out of service
a) have to be updated immediatelly for any inserts or updates upon their reconnection in order to have synchronized data, and
b) deletes would not be able to be confirmed upon constraints of offline databases (which makes deletes with offline databases impossible to my eyes at this time).
I'd like to ask for any ideas on the implementation of this.
Thanks in advance,
teo

Well, this sort of problem is tackled by Oracle's Advanced Replication option. You don't say which version of the database you are using, but you can find the documentation for most versions online. For instance, the 9i stuff is here.
Cheers, APC

Similar Messages

Export data from database table before database migration

Hello,
We are planning to migrate our SAP ERP 6 Ehp4/NW7.01 from Oracle 11.2 to IBM DB2 v. 9.7 database. During test migrations I have established that we spend a lot of time for one particular table (COEP). Because we donu2019t have possibility to archive this table before migration I have an idea to export data from previous years from this table to the file system (using an ABAP report), delete those data from table before migration and then after migration, import back to the database from the file system.
Does anybody have any concerns or suggestions about this idea?
Thank you for your answers
Andrej

Hello Andrej,
I strongly do not recommend to do so.
I am not sure whether technically this could work at all..
Even it if would work .. In order to really save time, the export and the import would have to be a dirty one (meaning the system is operational and in production). With this there is a high risk to produce inconsistencies on this table. And you most likely will receive no support if something unforeseen happens and you end with problems.
Also your approach (if it should work at all) , would have to be tested thoroughly by you , also protecting the table from any changes.
I do not believe that this can save any effort compared to implementing advanced migration techniques like table splitting.
On top, you would go high risk to end with an unsupported system, with not using official migration procedures
Hans-Juergen

Oracle SOA Suite 10.1.3.4 Polling database table with Database Adapter

We are trying to use the database adapter to poll for new and changed data in a database. To do that we try to use a sequencing table on an external database. But for that we need to specify the data-source name for the external sequencing table. Where do we need to define this data-source? We have tried with data-source defined in the Enterprise Manager, but it's not right. Also tried with the name of database connection in JDeveloper, still not working.
Any ideas?

Hi,
If you are trying to understand how to create a connection pool for a data source, then this document might be helpful
http://blogs.oracle.com/fusionmiddlewarereallife/soa_11g_bpel_composite_and_database_adapter.html

Need to insert into NVARCHAR2 column in a database table

I need to insert into a table with column type NVARCHAR2(2000) in java.
Cant use normal setString on that column. How can I do this using PreparedStatement in Java?

The scenario is:
I have to read a CSV file which contains a column in Urdu language, and show it on the JTable first.
Then I have to import the JTable contents into a database table.
Database Table structure is:
CREATE TABLE IMPORT_TMP (
ctype          VARCHAR2(3),
urdu_name NVARCHAR2(2000)
);My java code which inserts into the database table is:
                Vector dataVector = tableModel.getDataVector();
                rowVector = (Vector) dataVector.get(row);
              cType = "";
                if (rowVector.get(INDEX_BANK) != null) {
                    cType = rowVector.get(INDEX_CTYPE).toString();
                urduName = "";
                if (rowVector.get(INDEX_URDU_NAME) != null) {
                    urduName = rowVector.get(INDEX_URDU_NAME).toString();
                statementInsert.setString(1, cType);
                statementInsert.setString(2, urduName);I also applied Renderer on the table column, my renderer class is:
public class LangFontRenderer extends JLabel implements TableCellRenderer {
    private Font customFont;
    public LangFontRenderer(Font font) {
        super();
        customFont = font;
        System.out.println("font = " + font.getFontName());
        this.setFont(font);
    @Override
    public Component getTableCellRendererComponent(JTable table,
            Object value, boolean isSelected, boolean hasFocus, int row, int column) {
        if (value != null) {
            if (value instanceof String) {
                setText((String) value);
                return this;
        return this;
    @Override
    public Font getFont() {
        return customFont;
    // overriding other methods for performance reasons only
    @Override
    public void invalidate() {
    @Override
    public boolean isOpaque() {
        return true;
    @Override
    public void repaint() {
    @Override
    public void revalidate() {
    @Override
    public void validate() {
}I applied the renderer on the column as:
        TableColumn col = IATable.getColumnModel().getColumn(INDEX_URDU_NAME);
        LangFontRenderer langRenderer = new LangFontRenderer(new java.awt.Font("Urdu Naskh Asiatype", 0, 11));
        col.setCellRenderer(langRenderer);It does not give any error but when i query on the database table it does not show the column data in Urdu language.
Also it does not show the correct value on JTable column, some un-identified symbols.
Furthermore when I open the CSV file in notepad, it shows the correct language in that column, as I have installed the Urdu language and font on my system.

Data structure similar to database table

i have the need for a data structure that will essentially be the same as a database table. this "table" can be queried against and sorted.
currently, i am using an interface to abstract this "table" datastructure, and implementing the interface via jdbc calls.
may someone point me in the right direction?
in .NET, they have the DataTable and DataSet data structures, which do exactly as i require. These are objects essentially correspond to a database table and database, respectively. They are used as in-memory database.
i have already tried using HSQL, but was wondering if there was a "better" way (i.e. more light-weight, intuitive, easier to use).
thanks.

Use an array and put your records in it.
Efficient querying (simple queries) requires you to build indices for each element you wish to query on. The type of index depends on the kind on search you want to perform. Exact matches can use Hashes. Sorting, prefix or postfix matching can be performed using various tree indices.
Complex queries (i.e. those that query on multiple parameters generally require query optimisers to figure out how to use the indies you have in the most efficient manner possible (well as close as the optimizer can get to it).
cheers
matfud

How to create a view consisting of data from tables in2 different databases

Using Oracle 10.2g
I have 2 databases Gus and haggis on Comqdhb schema.
glink indicates a databse link between Haggis and Gus
In Gus there are tables student,subject,grade,school containing columns like upn...
STUDENT
upn
academicYear
SUBJECT
subject
GRADE
examlevel
grade
SCHOOL
sn
In HAGGIS there are tables student,grade,teacher containing columns upn...desc below.
STUDENT
upn
GRADE
grade
upn
academicyear
level
Create view in your HAGGIS database which will join all of the exam grades together. You should have one view which will produce the following relation :
examGrade(upn, subject, examlevel, sn, grade,academicYear)
so I need to create a view which gets the data from both the tables in both the databases.
create view as examGrade(upn, subject, examlevel, sn, grade,academicYear) as select s.upn
But i am not getting how to select a column from 2 tables in different databases
I mean if i said
select upn from comqdhb.student@glink,comqdhb.student;
select upn from comqdhb.student@glink,comqdhb.student
ERROR at line 1:
ORA-00918: column ambiguously defined
help me out,Thank you.

Thank you for the reply will follow up the code format
Create views in your HAGGIS schema database which will join all of the exam grades together. You should have one view which will produce the following relation :
examGrade(upn, subject, examlevel, sn, grade,academicYear)
I understand that there wont be duplication when we use conditions
If i query
select count(upn)
from   comqdhb.student@glink I get 9000
but after the union
create view examGrade(upn, subject, examlevel, sn, grade,academicYear)
as
select distinct s.upn as upn
,                  g.subject as subject
,                  g."LEVEL" as examlevel
,                  g.grade as grades
,                  '9364097'
,                  to_number(g.academicyear) as academicyear
from             comqdhb.student s
,                   comqdhb.grade g
where           s.upn=g.upn
union
select            s.upn
,                   sb.subject
,                   g.elevel
,                   g.grade
,                   s.acyr
,                   sc.sn
from              comqdhb.subject@glink sb
,                   comqdhb.student@glink s
,                    comqdhb.gradevalues@glink g
,                    comqdhb.school@glink sc,
,                    comqdhb.studentingroup@glink sg
,                    comqdhb.teachinggroup@glink tg
where            sb.sid=tg.sid
and                tg.gid=sg.gid
and                sg.upn=s.upn
and                g."LEVEL"=tg.elevel
and                s.school=sc.id
and                sc.id=tg.id; returns
count(upn) from exam gradeIt gets stuck actually sometimes it returns
932002 some results.
2:
Another problem i am having which i am trying to solve and written up my ideas but haven't been getting the expected results.Hope you can help.Thank you.
Information:
=======
All children take exams at the age of 16 called a General Certificate of SecondaryEducation (GCSE).
They have to study and take exams in Mathematics, English and Science, and can take other subjects such as History, French, Art etc. Most students will study between 5 and 10 different subjects before taking their GCSEs.
For each exam, a student is awarded a grade from A*, A, B,C,D,E,F,G,U,X An A* grade is the best grade achievable and an X is the worst grade.
In order to analyze how students have performed, each grade is mapped to a numeric value as follows:
Grade Numerical score
A* 8
A 7
B 6
C 5
D 4
E 3
F 2
G 1
U 0
X 0
Now why i need this avgGCSE is because i have to create a view containing avgGCSE of the students it is used in the next question where a condition is avgGCSE is between 6.5 and 7
In order to calculate the avgGCSE the idea is to calculate the grades of the students and map the grades to their corresponding scores/values
add them all up and div by the total no of grades to get the avg.
desc comqdhb.STUDENT@glink;
STUDENT
=======
UPN
FNAME
CNAME
DOB
GENDER
PREVIOUSSCHOOL
XGCSE
SCHOOL
ACYR
STUDENTINGROUP
=============
UPN
GID
STARTDATE
ENDDATE
GRADE
GRADEVALUES
===========
GRADE
LEVEL
VALUE
I have a opinion that xgcse in STUDENT table refers to the avgGCSE which i want to calculate as when i asked my professor as to what xgcse he said that he forgot to take it out of the table and it is not necessary while creating avggcse.
select *
from comqdhb.student@glink
where xgcse<6.5; Displaying a result
returns:
UPN FAMILYNAME COMMONNAME DATEOFBIR GENDER PREVIOUSSCHOOL XGCSE SCHOOL ACYR
===========================================================================
1011 KIMBERLY ABBOT 07-JUL-79 f none 3.93500948 2 2
select *
from comqdhb.student@glink
where xgcse between 6.5 and 7 and upn = 1386; Displaying a result
returns:
UPN FAMILYNAME COMMONNAME DATEOFBIR GENDER PREVIOUSSCHOOL XGCSE SCHOOL ACYR
===========================================================================
1386 STEPHANIE AANNESSON 15-JAN-79 f none 6.88873 2 2 so if xgcse is the avgGCSE then upn 1011 has avggcse<6.5 and 1386 has avggcse >6.5
my idea was backward strategy like so now if we find out upn 1368 has suppose xgcse(avggcse)>6.5 how to extract the avggcse for the particular upn We need to map grades from GRADEVALUES to grade in STUDENTINGROUP and map upn from studentingroup to upn in student to output the values for the corresponding grades from GRADEVALUES
select grade
from comqdhb.studentingroup@glink
where upn = 1011;
Result:
GRADE
=====
D
F
B
E
C
E
E
B
8 rows selected. Mapping each grade to the corresponding value and calculating we get
32/8=4 total(values to corresponding grades)/no of grades.
But the xgcse for upn 1011 is 3.935 and i am getting 4!! maybe xgcse isn't avggrade but ? is the procedure by me correct for calculating avggcse
select grade
from comqdhb.studentingroup@glink
where upn = 1386;
Result:
GRADE
======
A*
A*
A*
A*
B
A*
A*
A
B
B
B
11 rows selected. grade to the corresponding value and calculating we get
79/11=7.12 total(values to corresponding grades)/no of grades.
But the xgcse for upn 1011 is 6.88... and i am getting 7.12!!
But another problem
when i say
select   g.value,g.grade
from     comqdhb.gradevalues@glink g
,        comqdhb.studentingroup@glink sg
where    g.grade=sg.grade
and      sg.upn=1011;
result:
======
VALUE GRADE
===========
100 B
100 B
   80 C
   60 D
   40 E
   40 E
   40 E
   20 F
    6 B
    6 B
    5 C
VALUE GRADE
=============
    4 D
    3 E
    3 E
    3 E
    2 F
16 rows selected.
select   distinct g.value,g.grade
from     comqdhb.gradevalues@glink g
,        comqdhb.studentingroup@glink sg
where    g.grade=sg.grade
and      sg.upn=1011;
result:
======
VALUE GRADE
============
     2 F
   100 B
     6 B
     3 E
    60 D
     5 C
     4 D
    80 C
    40 E
    20 F
10 rows selected. I am getting only 8 for the query
select grade
from comqdhb.studentingroup@glink
where upn = 1386; here its becomming 10 and also its displaying values as 100 and ...
select distinct *
from   comqdhb.gradevalues@glink;
GRADEVALUES
===========
LEVEL      GRADE           VALUE
================================
a          A                 120
a          B                 100
a          C                  80
a          D                  60
a          E                  40
a          F                  20
a          U                   0
a          X                   0
g          A                   7
g          A*                  8
g          B                   6
LEVEL      GRADE           VALUE
================================
g          C                   5
g          D                   4
g          E                   3
g          F                   2
g          G                   1
g          U                   0
g          X                   0
18 rows selected. I was hoping if i could map the grades and get the values and calculate avggrade by total(values)/count(values)that would be it but here there are values like 100...
select sum(g.value)/count(g.grade) as avggrade
from    comqdhb.gradevalues@glink g
,         comqdhb.studentingroup@glink sg
where g.grade=sg.grade
and     sg.upn=1386;
avggrade
========
37.4375 the avggrade cant be this big and when i map each grade i obtained for 1368 like a to 7+b to 6 so on i get avggrade 7.12
kindly help.
Edited by: Trooper on Dec 15, 2008 4:49 AM

How to use %Type declaration with table residing in a different database

How can I use the %TYPE declaration if the table is from a different database.
E.g
v_business_unit ps_jrnl.header.business_unit%TYPE;
In the above declaration statement,the table 'ps_jrnl_header' resides in a different database(Database A) from the one I am currently in( Database B).(This is because data needs to be extracted from Database A into Database B).

1. Create a database link to the other database (this probably already exists since your proc is interacting with that database)
2. Create a synonym for the table in the other database
create synonym foo for ps_jrnl.header@database_a;3. Reference the synonym in your variable declaration:
v_business_unit foo.business_unit%TYPE;

Table transformation from two different database

Hi all,
I am trying to load table from one database table to another database table in ODI 10g.I have created two physical and logical topology.
But when in interface design i have drag the source and target tables in the diagram window, but here it doesnot shows the auto mapping options, when i try it manually by dragging a field, i have failed. when i execute the operation the following error occured:
create or replace view db1."C$_0 W_INT_ORG_D"
as select
from     db1.W_INT_ORG_DS W_INT_ORG_DS
where     (1=1)
here both the source and target db are same..
How can i resolve it

Hi,
Thanks it worked.
I am getting another error on Insert flow into I$ table step.
I am connecting two database from two different hosts, both are in oracle.
I have two logical schema prepared for both database.
should i have to mentioned these schema in one physical topology? what should be placed in schema and work schema option.
I have used LKM Oracle to Oracle(DBLINK) as KM.
Please help
Best regards,
manish

Compare Table Data on 2 different databases having same schema

I need to compare data in all the tables in 2 different databases having same schema.
If there is any difference in data in any table on Database1 and Database2 then I need to update/insert/delete the rows in my table in Database2.
So Database1 is my source database and Database2 is my sync database. I cannot use expdp tables as I am not having sufficient privileges to the database server.
Also I cannot drop and recreate the tables as they are huge.
Can anyone please guide me how to compare data and to write a script to comapre the changes in say Database1.Table1 and Database2.Table1 and then accordingly do inserts/updates/deleted on Database2.Table1?
Thanks

Karthick_Arp wrote:
Do you have a DBLink? If youes you can do this.
1. Login into the Database-2 and run this code.
begin
for i in (select table_name from user_tables)
loop
execute immediate 'truncate table ' || i.table_name;
end loop;
end;This will empty all the tables in your Database-2. Now what you need is to just populate the data from Database-1This might result in error, if any of the tables have referential integrity on them.
From 10g documentation :
Restrictions on Truncating Tables
You cannot individually truncate a table that is part of a cluster. You must either truncate the cluster, delete all rows from the table, or drop and re-create the table.
You cannot truncate the parent table of an enabled referential integrity constraint. You must disable the constraint before truncating the table. An exception is that you can truncate the table if the integrity constraint is self-referential.
If a domain index is defined on table, then neither the index nor any index partitions can be marked IN_PROGRESS.I would go for normal MERGE. Also change the cursor to select table names by first modifying the child tables and then the parent table.

Select query based on joining of tables from different database is taking too long

Hi Team,
Select query on table with millions of records is taking very long time. It took 50mins, the below query is joining on multiple table from two databases DB1.dbo.Table1 contains 100 million records and also Table3 and Table4 (of different database) might
contain close to 1million records.
Select T1.*
From DB1.dbo.Table1 T1
Join DB1.dbo.Table2 T2 on RTRIM(T1.Col3) = RTRIM(T2.Col3)
Join Table4 CA on RTRIM(T1.Col1) + T2.Col2 = CA.Col1
Join Table3 U on CA.Col2 = U.Col2 AND RTRIM(T2.Col2) = U.Col3
Where U.Col4 NOT IN ('A1', 'A2', 'A3', 'A4', 'A5', 'A6','A7','A8','A9')
And (T1.flg IS NULL OR T1.flg = 'N')
And LTRIM(RTRIM(T2.Col2)) NOT IN ('B1','B2')
How can i improve the performance of this query. Actual thing is update the data in Db1.dbo.table1 based on the conditions but if the select is taking close to 1hr then update will take hours together. Indexes already implemented on all the tables.
Thanks,
Eshwar.
Please don't forget to Marked as Answer if my post solved your problem and use Vote As Helpful if a post was useful. It will helpful to other users.

Thanks.
Join Table4 CA on RTRIM(T1.Col1) + T2.Col2 = CA.Col1
This join is not working with spaces if Rtrim is not used.
Thanks! Eshwar.
Please don't forget to Marked as Answer if my post solved your problem and use Vote As Helpful if a post was useful. It will helpful to other users.

Using Dense/Sparse Lookup on Tables or files from different Databases

Hi,
I have a Subject Area which is built on Essbase cube. I have a requirment where I have to use a Lookup on Flat file and/or Database Table.
I followed below steps
1. Created a Logical Column in BMM Layer for a cube dimension
2. Opened the LTS Properties --> Column Mapping and gave below formula
Lookup(SPARSE "Function1"."E:\502125500".""."Function2.txt"."ALIAS", 'NA', "TNWD06613.genpitfi01.og.ge.com"."GOGREP".""."GOGREP"."Gen5,Functions" )
where
Function1"."E:\502125500".""."Function2.txt"."ALIAS" ---> Column from File (Created File DSN and imported in RPD)
"TNWD06613.genpitfi01.og.ge.com"."GOGREP" ---> Essbase Cube
3. Dragged the Column to Presentation for Subject Area on Cube
When I try to create a report on this column I get below error.
Error Codes: OPR4ONWY:U9IM8TAC:OI2DL65P
State: HY000. Code: 10058. [NQODBC] [SQL_STATE: HY000] [nQSError: 10058] A general error has occurred. [nQSError: 43113] Message returned from OBIS. [nQSError: 16001] ODBC error state: 22005 code: -3030 message: [Microsoft][ODBC Text Driver] Expression too complex. [nQSError: 16015] SQL statement execution failed. (HY000)
SQL Issued: SELECT 0 s_0, "Essbase Test"."Functions"."Alias" s_1, "Essbase Test"."Functions"."Gen5,Functions" s_2, SORTKEY("Essbase Test"."Functions"."Gen5,Functions") s_3 FROM "Essbase Test" FETCH FIRST 65001 ROWS ONLY
Can anybody help me on the same?

You can script schemas from all DBs and in case of conflicts you will have to decide what to do.
For example if you have the same SP or UDF on two DBs that does different things, you need to decide what to do with them.
Only if the conflict is tables with the same structure and without KEY constraints and you can also script data and go on.
If you have PK-FK relationships (which I bet is the case) you really need to design your changes based on them and I suspect you will end up scripting instead of putting boxes on SSIS.
It really is a migration job for an experienced DBA and am afraid SSIS can do that much as
Integrate.. not Migrate.

How to create User and Database in different Table spaces

How to create User and Database in different Table spaces using oracle 10g
Regards
daya

I am sorry but your question does not seem to make much sense.
Can you please rephrase your question?

Joining tables in different databases

Hi,
I know we can access multiple databases like Oracle, Informix etc. through a single toplink session using session broker. Also in toplink documentation it says there is a work around for joining tables that exists in different databases i.e. joining an Oracle table with an Informix Table. If anyone had tried this or if it is possible, please let me know how to do this.
Thanks..

If I could know, how to perform or some sample code, for the following two steps it would be great.
Write descriptor ammendment code to bind together that go from (A) to (B) and (B) to (A).
Be sure to map the ammendment descriptors in the mapping workbench.Or please point me to the right toplink documentation or examples.
Thanks a lot for your help.

Bug: showing data in tables with same name, 2 different databases

Hi,
two connections, to databases "a" & "b". Both logged into the same schema, "s".
If you select table "t" in db "a" and then select the same table in db "b", the tab is not refreshed - you have to either select a different object first, or explicitly hit refresh.
regards,
William

I can reproduce this consistently on 9.2.0.1.
You need to have connections to the same schema in two different databases and the schemas must have the same table.
1 Open connection in schema 1
2 Select the table in the Tables node
3 Select the Data tab
4 Open connection to schema 2
5 Select the table in the Tables node (without going to any other tabs or objects)
What should happen is the details of the second table should occupy the main display tab and the data tab should be refreshed to show the data of the new table.
What actually happens is the main display tab apparently switches to the table in the new connection as shown by the tooltip (table@connection_name), but the data in the data tab stays the same. Refresh doesn't seem to work, only switching to a different object and back again.
The columns tab does seem to refresh properly.

How to integrate different tables structures from different databases?

Hi
I'm new to oracle ..kindly help me with the below issue...
I have same set of inter-related attributes (a,b,c,d) in two different databases A and B.
The structure of tables in both the database are different but they carry same attributes in same relation.
I have to integrate these attributes into single db object such that any update to base tables updates my object.How can i do this and what object will be helpful?
Thanks

Arun Rajesh wrote:
Hi
I'm new to oracle ..kindly help me with the below issue...
I have same set of inter-related attributes (a,b,c,d) in two different databases A and B.
The structure of tables in both the database are different but they carry same attributes in same relation.
I have to integrate these attributes into single db object such that any update to base tables updates my object.How can i do this and what object will be helpful?
ThanksWelcome to Forum!!!
Please elaborate your question.
Arun Rajesh wrote:
I have same set of inter-related attributes (a,b,c,d) in two different databases A and B.1. What are the ATTRIBUTES?
Arun Rajesh wrote:
The structure of tables in both the database are different but they carry same attributes in same relation.2. Post the Table Structures.
Arun Rajesh wrote:
I have to integrate these attributes into single db object such that any update to base tables updates my object.How can i do this and what object will be helpful?3. What is the end result that you desire? A sample output shall help in our understanding.
PS:- Please go through this to understand SQL and PL/SQL FAQ.
Regards,
P.

Syncronized database tables (different databases)

Similar Messages

Maybe you are looking for