Different query plan on different schema

Hi gurus
i have a problem witha a select: i explain the same select on differnt schema and i obtain a different query plan.The tables have the same key,index and data.
which are the possible reason for this behavior ?

This is the first select
- <select-statement>
- <sort type="order-by">
- <nested-loops type="outer">
- <nested-loops type="outer">
- <hash-join>
- <inlist-iterator>
- <table-access object="tbl_0009" type="by-index-rowid">
<index object="ky0009_1" type="range-scan" />
</table-access>
</inlist-iterator>
- <table-access object="tbl_0210" type="by-index-rowid">
- <nested-loops>
- <table-access object="tbl_0010" type="by-index-rowid">
<index object="ky_0010_2" type="range-scan" />
</table-access>
<index object="ky0210_1" type="range-scan" />
</nested-loops>
</table-access>
</hash-join>
- <table-access object="tbl_0008" type="by-index-rowid">
<index object="ky0008_1" type="unique-scan" />
</table-access>
</nested-loops>
- <table-access object="tbl_0008" type="by-index-rowid">
<index object="ky0008_1" type="unique-scan" />
</table-access>
</nested-loops>
</sort>
</select-statement>
The second
- <select-statement>
- <sort type="order-by">
- <nested-loops type="outer">
- <nested-loops type="outer">
- <nested-loops>
- <nested-loops>
- <table-access object="tbl_0010" type="by-index-rowid">
<index object="ky_0010_2" type="range-scan" />
</table-access>
- <table-access object="tbl_0210" type="by-index-rowid">
<index object="ky0210_2" type="range-scan" />
</table-access>
</nested-loops>
- <inlist-iterator>
- <table-access object="tbl_0009" type="by-index-rowid">
<index object="ky0009_1" type="range-scan" />
</table-access>
</inlist-iterator>
</nested-loops>
- <table-access object="tbl_0008" type="by-index-rowid">
<index object="ky0008_1" type="unique-scan" />
</table-access>
</nested-loops>
- <table-access object="tbl_0008" type="by-index-rowid">
<index object="ky0008_1" type="unique-scan" />
</table-access>
</nested-loops>
</sort>
</select-statement>
The select is the same
SELECT datareg,decode(dare,null,0.00,dare) dare,decode(avere,null,0.00,avere)avere,protocollo,contocontr,numdo,datad,cocp,ccde9,conto codiceconto,contocliprincipale contocli, 0.00 dareSaldo, 0.00 avereSaldo , null conto_saldo FROM (SELECT /*+ ordered use_nl ( tbl_0210)*/ tbl_0210.coso210 conto, id10, datar10 datareg, proto10 protocollo, ' ' || pianoconti.prefi8 || '.' || LPAD(pianoconti.mastro8, 4, '0') || '.' || LPAD(pianoconti.soco8, 6, '0') || ' ' || pianoconti.nome8_1 contoContr, tbl_0008.prefi8 || '.' || LPAD(tbl_0008.mastro8, 4, '0') || '.' || LPAD(tbl_0008.soco8, 6, '0') || ' ' || tbl_0008.nome8_1 || ' ' || tbl_0008.nome8_2 || ' ' || tbl_0008.via8_1 || ' ' || tbl_0008.loca8_1 contoCliPrincipale, (ccde9 || ' ' || decode(tbl_0210.testo210,null,'',tbl_0210.testo210)) ccde9 , COCP10 cocp, numdo210 numdo, datad210 datad, DECODE(dav210, 'D', impe210, 0.00 ) dare, DECODE(dav210, 'A', impe210, 0.00 ) avere FROM tbl_0010, tbl_0210, tbl_0009, tbl_0008, tbl_0008 pianoconti WHERE datar10 BETWEEN TO_DATE('01-01-2007','dd-mm-yyyy') AND TO_DATE('31-12-2007','dd-mm-yyyy') AND id10 = tbl_0210.id210 and tbl_0210.coso210 between 10230000100 and 10230001000 AND tbl_0008.coso8(+) = tbl_0210.coso210 AND pianoconti.coso8(+) = conc210 AND cccg9 = caus210 and tipo9 in (1, 2) ) principale ORDER BY codiceconto ,datareg ,protocollo
I do this on different schema because the customer have one database with 2 factory schemas

Similar Messages

Same query at same time, but different execution plans from two schemas

Hi!
We just had some performance problems in our production system and I would like to ask for some advice from you.
We had a select-query that was run from USER1 on SCHEMA1, and it ran a table scan on a huge table.
Using session browser in TOAD I copied the Sql-statement, logged on SCHEMA1 and ran the same query. I got a different execution plan where I avoided the table scan.
So my question is:
How is it possible that the same query get different execution plans when running in two different schemas at the same time.
Some more information:
The user USER1 runs "alter session set current_schema=SCHEMA1;" when it logs on. Besides that it doesn't do anything so session parameter values are the same for USER1 and SCHEMA1.
SCHEMA1 is the schema owning the tables.
ALL_ROWS is used for both USER1 and SCHEMA1
Our database:
Oracle9i Enterprise Edition Release 9.2.0.8.0 - 64bit Production
PL/SQL Release 9.2.0.8.0 - Production
CORE 9.2.0.8.0 Production
TNS for Linux: Version 9.2.0.8.0 - Production
NLSRTL Version 9.2.0.8.0 - Production
Anybody has some suggestions to why I experience different execution plan for the same query, run at the same time, but from different users?

Thanks for clarification of the schema structure.
What happens if instead of setting the current session schema to SCHEMA1, if you simply add the schema name to alle tables, views and other objects inside your select statement?
As in select * from schema1.dual;I know that this is not what you want eventually, but it might help to find any misleading objects.
Furthermore it is not clear what you meant with: "avoided a table scan".
Did you avoid a full table scan (FTS) or was the table completely removed from the execution plan?
Can you post both plans?
Edited by: Sven W. on Mar 30, 2010 5:27 PM

Dbms_sql in a different schema from query table-error ** ORA-00942

Oracle Experts,
I think I am having problems with using DBMS_SQL in which the function was created in one schema and the query table was created in a different schema.
We have 2 schemas: S1, S2
We have 2 tables:
T1 in Schema S1
T2 in Schema S2
We have a function F1 created by DBA in schema S1 that uses the dbms_sql as:
CREATE OR REPLACE FUNCTION S1.F1(v1 in VARCHAR2) return NUMBER IS
cursor1 INTEGER;
BEGIN
cursor1 := dbms_sql.open_cursor;
dbms_sql.parse(cursor1, v1, dbms_sql.NATIVE);
dbms_sql.close_cursor(cursor1);
return (0);
EXCEPTION
when others then
dbms_sql.close_cursor(cursor1);
return (1) ;
END;
I am using jdeveloper 11G. We have an Oracle DB 11g.
We have a java program which uses jdbc to talk to our Oracle DB.
Basically, in my java program, I call function F1 to check if the query is valid.
If it is, it returns 0. Otherwise, returns 1:
oracle.jdbc.OracleCallableStatement cstmt = (oracle.jdbc.OracleCallableStatement) connection.prepareCall ("begin ? := S1.F1 (?); end;");
cstmt.registerOutParameter (1, java.sql.Types.INTEGER);
cstmt.setString(2, "Select * from S2.T2");
cstmt.execute ();
Since the table that I run the query is T2, created in different schema than F1 was created in, I have the error:
** ORA-00942: table or view does not exist
So my questions are these:
- I am using Oracle DB 11g, if I run the query on a table that created in a different schema from the one that the function (which uses dbms_sql) was created in, I would get the error ORA-00942?
- If I runs the query on table T1 in the same schema as the function F1, would I have the same problem(The reason I ask is I cannot create any table in schema S1 because the DBA has to do it; I am not a DBA)
- This is not a problem, but a security feature because of SQL injection?
- How to resolve this issue other than creating the table in the same schema as the function that utilizes DBMS_SQL?
Regards,
Binh

Definer rights (default) stored objects run under owner's security domain and ignore role based privileges. So regardless what user you are logged in as, function S1.F1 always executes as user S1 and ignores user S1 roles. Therefore exeuting statement within S1.F1:
Select * from S2.T2requires user S1 to have SELECT privilege on S2.T2 granted to S1 directly, not via role.
SY.

Sinlge select query in different schemas for same table(Indentical Structu)

Scenario :
Table XYZ is created in Schema A
After an year, the old data from the previous year would be moved to different schema. However in the other schema the same table name would be used.
For eg
Schema A contains table XYZ with data of 2012 yr
Schema B contains table XYZ with data of 2011 yr
Table XYZ in both the schemas have identical structure.
So can we fire a single select query to read the data from both the tables in effective way.
Eg select * from XYZ where date range between 15-Oct-2011 to 15-Mar-2012.
However the data resides in 2 different schema altogether.

Thanks for the reply
Creating an view is an option.
But my problem, there is ORM layer(either Hibernate or Eclipse Top Link) between the application and the database.
So the queries would be formed by the ORM layer and are not hand generated.
So i cannot use view.
So is there any option that would allow me to use single query on different schema's ?

SQL Query Template to a different schema

Hi all
I am on MII 12.2, connecting to Oracle through a data source which references user/schema "Foo"
I would like to perform queries involving tables from schema "Bar" .
As a consequence, when I need to access Bar tables I must reference them like "Foo.<TableName>"
If I use an SQL query with mode Command I can freely write the prefix but no return is allowed
If I use mode = FixedQuery(With Output) I can't prefix "Foo." because the editor prevents me from freely entering SQL code
How can I do this? Thanks regards
Vincenzo

Thanks for your reply!
Actually there is a primary key, but Streams seems to ignore it. I've also tried the SET_KEY_COLUMNS way plus supplementary logging on the source DB, but this didn't help at all. I think this is happenening because the tables are not in the TEST_HQ schema, there are only synonyms to the real tables contained in TEST_CO.
Is there any other easy way to get Streams working between two different schemas?
Thanks again for your help!!
Max

Need query to compare the columns of 2 diff tables of 2 different schemas.

There are two different tables(sample1, sample2) in different schemas(s_schema1, s_schema2).
I want the query to compare the columns of two different tables of two different schemas and provide whether the data as well as the count of data in
the column are same .
if not provide the data which is not similar in the columns of two different table.
NOTE:
I need queries for both the cases.
(i) The datatypes in columns of two different tables are same.
(ii) The datatypes in columns of two different tables are diffrent.

Welcome to the forum!
Whenever you post provide your 4 digit Oracle version.
>
I need queries for both the cases.
>
Great - write the queries!
The forum is not a coding service where you ask people to write code for you for free.
YOU need to write the code. Then if you have a problem with the code you have written post the code you have written (using \ tags) and explain the problem you are having.
Read the FAQ about how to ask a question on the forums.

Query from different schema

I am using oracle sqldeveloper. I am trying to run a query on a different schema. Whenever I run the following:
Select *
From SchemaA.Users us
Order by US.last_name asc
I get the following error:
US.last_name: invalid identifier.
I would get the same error when I place a where clause on the end as well. I am sure it is an error in syntax. Nonetheless, I am not sure how to correctly write the query. Any help on this topic would be great. Thanks in advance.

Hi,
NewApexCoder wrote:
I am using oracle sqldeveloper. I am trying to run a query on a different schema. Whenever I run the following:
Select *
From SchemaA.Users us
Order by US.last_name asc
I get the following error:
US.last_name: invalid identifier.
I would get the same error when I place a where clause on the end as well. I am sure it is an error in syntax. Nonetheless, I am not sure how to correctly write the query. Any help on this topic would be great. Thanks in advance.
I don't see anything obviously wrong with the syntax, or ianything else in that code. You didn't post CREATE TABLE and INSERT statements for the USERS table, so I can't actually test it, but I can run very similar queries, like this:
SELECT    *
FROM      scott.dept sd
ORDER BY sd.dname ASC
Are you sure there's a column called LAST_NAME in the SCHEMAA.USERS table? If I spell DNAME wrong, then the query above gets the "invalid identifier" error, just like yours.
USERS is an Oracle keyword, so it's not a very good table name. That's almost certainly not what's causing this problem, however.

Query to list the synonyms in one schema to different schema

Hi
How can i check the list of synonyms in one schema pointing to different schemas?
Also query to list the table , foreign key constraint and reference table..
Thanks

Manikandan wrote:
Hi
How can i check the list of synonyms in one schema pointing to different schemas?
Also query to list the table , foreign key constraint and reference table..
Thanks
1) DBA_SYNONYMS or ALL_SYNONYMS
2) You can check DBA_CONSTRAINTS and DBA_CONS_COLUMNS.
Aman....

Import of BaseView to a different schema in database

Hi All,
We have developed some bam reports. Now the base view, meta view are referring to a schema name (say SchemaName1) in database.
In the plans we are using "SQL query" to get data from database. So we have queries like "Select SchemaName1.table1.column1 ... from ......".
Now we have taken a export of the base view, meta view ,plans and all other components of BAM.
All is fine till now... But now i need to import all these stuff in a different database. My problem is that in the new database instance i cannot use the same schema name (i.e. SchemaName1 ). I have a different schema name(say SchemaName2) there and i need to use that......There is no way i can use SchemaName1.
Will the import of Base view/Meta view/plans work or we need to do some configuration changes during import??
Do we have to create Base view for the new schema names from scratch or the imported code can be used??
The plans have sql queries referring to SchemaName1 but we need it to refer to the Other schema(SchemaName2 )...What is the possible way of referring to the new schema without developing the plan again??
Waiting for your valuable feedback...

***Backup your repository database***
***Please review all steps before attempting***
Sarputil Export
1.     Start | Run, type sarputil
2.     Do a partial export. (Note down directory that will contain files to be exported.).
3.     Select your Baseview, Metaview, Security:Login Profile, and Plan. (Note, the check boxes are in front of each object name and might appear very light on some clients.)
4.     On the XML dialog and with the question on what you would like to do select Execute Now.
5.     Finish the wizard and from the directory noted above, verify .csv files were created. (Typically C:\OracleBAM\EnterpriseLink\Data\rp\export).
6.     Copy all .csv files and paste to different temporary folder on 2nd server.
Sarputil Import
1.     Start | Run, type sarputil on the 2nd server where youll import into.
2.     Partial import
3.     Under Repository Information dialog, set the Data Source information on the right and remember to change the directory now containing your .csv files.
4.     Like before, select Execute now and Finish the wizard.
Oracle BAM Admin
Modify connect string (host string or tns name) if you need to:
1.     Connect in Oracle BAM Admin
2.     Expand Baseviews and select the imported Baseview
3.     Check under Server on the left if the connect string is correct.
4.     If not, click Modify button and change.
Modify Baseview Login if you need to:
1.     With your Baseview still selected on the left side, click on Baseview Logins tab
2.     Logins on the left correspond to actual database user ids and password. Add a new Login with a database User ID and Password to access the new schema.
3.     Associate or Set the new Login (right) to the BAM User on the left.
Sarpbv Modification
Use sarpbv utility to change references of the old schema name to a new schema name.
General syntax for Oracle:
sarpbv /R"username:pwd:Oracle:TNS Name::DB UserID:DBUser pwd /B"BaseView name" /O"NewSchema"
**Use Capital letters for New Schema name!
**Notice 2 colons (::) after TNS Name (because you do not have a database name in Oracle).
1.     Open DOS
2.     Type your sarpbv command. Below is an example where I changed a Baseview called scott to use a new schema name called Jack.
sarpbv /R"sa::ORACLE:baminst::sagent:sagent" /B"scott" /O"JACK"
3.     Open Design Studio, locate the plan and check the SQL Query now reflects the new schema.
4.     Test the plan to ensure its pulling data correctly from the new schema.

Any way to use a single application but point it to different schemas?

From the searching I have done, it appears that when an end-user runs an application, it can only be associated with a pre-defined schema, which I guess I just need confirmation of. What I was hoping to be able to do was either dynamically change to a different schema at run-time, or create an end-user that is associated with a different schema than the one the application is associated to so the user could use the one application but access a given schema.
The scenario is that in our production environment, we need to maintain a separate schema for each client we manage data for. But we'd like to maintain one application end users would use but could run against any one of the client schemas. However, it seems that we willneed to make a copy of the application within the production workspace for each client schema that would be the owner schema if I understand how this works. Thus, if we have 7 different schemas we would have to have 7 copies of the application, one associated with each application.
Thanks in advance!
Monte Malenke

Thanks Scott for quick response.
We will go with different workspace for each schema.
Just to give you quick background of my requirement. We have a 3 oracle E-Business versions (11.5.8, 11.5.9 and R12 in future) and APEX installed on another 11g database. We don't want to installed APEX on EBS database because of DB patching issues, our EBS 11i database version is 9i DB, future oracle EBS supoort and at the same time we want to use APEX 3.1.2 with 11g. We want use APEX for custom EBS Web UI development.
We are planning to create separate schema on APEX database for each EBS instance and DB link which points to EBS database. Under APEX schema we are planning to create a view (on our custom table which are same across all EBS instances) using DB link under each schema.
We will develop APEX application under one of the workspace and they deploy to other workspaces. We have also looked into creating APEX pages based on pl/sql procedure and we can do dynamic sql to query data from EBS instances. But its lots of code (create pl/sql api for every table) and we can not take a advantages of some of the APEX build-in wizard like master-detail, APEX record locking etc
Based on your APEX experience; Do you think this the way we should go? OR Is there any better way?.
Thanks in advance
RK

Materialized view with tables in different schemas

Hello,
I want to create a materialized view with a table from a different schema in the SELECT statement. For materialized view I would like to apply the "REFRESH COMPLETE ON COMMIT" option.
Here the code:
CREATE MATERIALIZED VIEW S1.MV_EXAMPLE
TABLESPACE TS1
PCTFREE 0
BUILD IMMEDIATE
REFRESH COMPLETE ON COMMIT
AS
SELECT T.COLUMN1 COLUMN
FROM S2.TABLE1 T
I can't execute this SQL because I get an "insufficient privileges" error to this table:
FROM S2.TABLE1 T
FEHLER in Zeile 9:
ORA-01031: Insufficient privileges
User S1 has the following privileges:
CREATE SESSION
CREATE SNAPSHOT
CREATE TABLE
CREATE QUERY REWRITE
SELECT ANY TABLE
User S2 has the following privileges:
CREATE SESSION
CREATE SNAPSHOT
CREATE TABLE
CREATE QUERY REWRITE
ALTER ANY SNAPSHOT
Which privileges are missing?
Thanks, Mathias

Thanks Kamal for your answer!
S1 has the grant select directly. But I solveld the problem. The system privilege "ON COMMIT REFRESH" was missing for S1. This has to be set if any of the tables are outside the owner's schema of the materialized view (ORACLE documentation - Data Warehouse Guide).
But one thing is not clear to me yet and the ORACLE documentation doesn't give me an answer. I can set the refresh-attribute ON COMMIT on a materialized view containing only joins when a group by clause is set. If the group by clause is missing I can't! Why?
Regards, Mathias

How to change all fk constraints in schema to refer to different schema

Hi,
We wish to change all the fk constraints which point to tables on one schema to same table names but owned by a different schema.
Is this possible?
Thinking of query user_constraints
select *
from user_constraints r
where r.constraint_type = 'R'
and r.r_owner = 'REF'
can put this inside loop but can you modify fk constriant to refer to table owned by different schema
11.2.0.3
Don't really want to need to drop and recreate.
Thanks
Edited by: user5716448 on 06-Apr-2012 04:26

Yes. You need to drop the constraint and recreate it using the clause 'new_schema.objct_name' in the constraint creation statement.
For example, if you want to change the constraint cons_1 to point to some column col_1 in schema sch_2 then the sequence of statements will be like,
ALTER TABLE TAB_1 DROP CONSTRAINT CONS_1;
ALTER TABLE TAB_1 ADD CONSTRAINT CONS_1 FOREIGN KEY COL_1 REFERENCES SCH_2.TAB_1(COL_1);
Remember, you have to have all the privileges required to use/reference the table in schema_2.

Compare table structures in different schemas. help please

Hi all,
I have a question...
I have tables on different schemas some thing like
Schema A
Table 1
Table 2
Table 3
Schema B
Table 1
Table 2
Table 3
Now situation is Table 1 and Table 2 will have similar structure or Table 1 in Schema B will have additional columns.
like so... on... for all other tables...
example !
Schema A:
Desc Table 1;
Name                                  Type            Null
No                                Number            Notnull
Name                           Varchar2(10)     Not null
Fee                              Number (10,2)   Not null
Scheam B;
Desc Table1;
Name                                  Type               Null
DX_No                                Number            Notnull
DX_Name                           Varchar2(10)      Not null
DX_Fee                              Number (10,2)   Not null
comments                          Varchar(2)        Now I need to write a procedure sort of thing to compare these tables which are in different in column names in both schema and get it exported to Excel sheet.
and here thing is first three columns SHOULD BE TREATED AS SAME even though prefix of DX_ exist since REST OF THE PART OF COLUMN NAME IS SAME.
and in same way commets is new coloumn in schema B only..So that should be highlighted excel sheets..
I am not sure how OAD or SQL Developer handle this...Is there any plsql block I can write to do above..
Thanks in advance..

An example of one method (this is just a starting point, you'll have to build it out to suit your needs).
create table emp as select * from scott.emp where 1=0;
alter table emp add (junk number(1));
select
      t1.column_name as t1_column_name
   , t2.column_name as t2_column_name
from
   select column_name
   from
      dba_tab_columns
   where
      table_name = 'EMP'
   and
      owner       = user
) t1
full outer join
   select column_name
   from
      dba_tab_columns
   where
      table_name = 'EMP'
   and
      owner       = 'SCOTT'
) t2
on (t1.column_name = t2.column_name)
where
   t1.column_name is null
or
   t2.column_name is null;
T1_COLUMN_NAME                 T2_COLUMN_NAME
JUNK
1 row selected.
TUBBY_TUBBZ?If you need to remove prefix's ... you would just do that in the select list (using regexp_replace, translate, substr, whatever you think is appropriate for your situation).
Edited by: Tubby on Nov 6, 2010 12:38 PM
Forgot to mention that this code will report differences between two schemas, regardless of where the additional columns reside ... if you have a 'master' schema that is the definitive source, then you would want to use the query provided by Xtender.

Analytic View with tables from different schema

Hi,
I'm curious if I make something wrong or if this is not possible:
I want to use two tables each in a different schema. I can drop them into the DataFoundation, can connect them, assign them some measure. But activation fails due to insufficient privileges. When I use the tables from the same schema activation works well. With the insufficient privileges I would guess HANA cannot create a View (failed with the statement CREATE COLUMN_VIEW "_SYS_BIC"."USER/ANALYTIC_VIEW_NAME"....).
Since I tried the same assignment with the System-User and failed I wonder what kind of permissions are required. Or is this generally not allowed (resp. prohibited)?
If not allowed, could somebody give me a hint, why its not possible?
If it is possible, what kind of rights are required to get it working?
roland

Hm', I have allowed SYS_REPO the rights to the table and was able to activate the view. But now I have the problem, that the DataPreview is not working (again insufficient privileges). What user has to have additional permissions for the JDBC-query?
Do I have to take care more on the privileges for other things?

Performance with having tables in different schema's

Hi all,
We have a requirement to segregate tables into different schema's based on the sources. But everything is going to be on the same instance and database.
Is there a performance hit (querying between tables) by having tables in different schema as apposed to having them in the same schema?
Thanks
Narasimha

Most likely there is bit of performance impact if your application refers the tables from different schemas. You need to use database link to access the other schemas. Even you schemas may in instance but when you use database link the network also comes into picture. even queries on same instance will routed through network connect and get the data. Distributed transaction also have issues some time. So as far as possible the distribution of objects should be avoided into diffrent schemas.

Different query plan on different schema

Similar Messages

Maybe you are looking for