Paralell query using db links

We use Sparc servers T2-based (CMT) for some of our Oracle 10g databases. Their performance is excellent when your query can be paralleled. But when we run queries (select sth from ...) from others Oracle databases using dblink we can't use parallel queries. Does anybody know why is that ?

Did you enable parallel dml on the local session? Do: "*alter session enable parallel dml*" and theninsert /*+ parallel */ into my_table
(branch, date, value1, value2)
select 1, sysdate, number1, number2
from huge_table@remote_database_db_linkOne thing you can do if the remote query is not running in parallel is create a view on your remote database with a parallel hint in the view definition. Something like this:create view huge_table_v
as select /*+ parallel */ number1, number2
from huge_tableAnd then insert from that view@db_link. Having the hint on the other side of the db link helps tell the remote database optimizer how to run that query.

Similar Messages

Can a report be developed using free hand SQL(Sql Query) using a DB Link

Hello All,
A report to be developed based on query in OBIEE dashboards
Select Query -----1
UNION
Select Query------2
First select query is coming from one database schema and second is coming from the other.
Is there any way to develop the report using free hand SQL based on a DB link?
Is there any feature to use Database link (DB link)?
Thank in Advance,
SMA

What you ask is the direct database request capabilities :
http://gerardnico.com/wiki/dat/obiee/presentation_service/obiee_direct_database_request
But you can also use the federated query capabilities of the repository to model a star schema with two star schema from two different data source.
Cheers
Nico

How to query a distant database without using db link in pl/sql

Hi
I have two different distantly located databases say d1,d2
There is a table dept which had deptno (primary key),department name as columns .
I need to query database d2 on the basis of the deptno present in database d1.
In database d2 , there are some tables which have columns as deptno.
I cannot use DB link .
Is there any other method that I can use from Pl/sql ?
Is it possible thru XML ?
PLs let me know in details .
regards
SHUBH

You need to have an Oracle Apache Server installed. It can be on any platform that has connectivity to the Oracle Server. Preferably on the same server as Oracle itself.
The Apache Server is installed in its ORACLE_HOME/Apache/Apache. It should have been configured by the installer. All you need to do is to create a DaD (Database Access Descriptor) for MOD_PLSQL. This is a custom Apache Module that provides direct access to PL/SQL via OCI. DaDs are defined in the ORACLE_HOME/Apache/modplsql/conf directory. There are README files, sample files and a dads.conf file in which you need to define a URL to assign as the service for your database, the Oracle username and password to use to logon, and the database connection details (e.g. TNS alias).
The Apache Server needs to be restarted for the changes to take effect (using either opmnctl or apachectl).
Let's say the URL path defined is /mydatabase. When Apache receives a URL that starts with /mydatabase, it passes it to MOD_PLSQL. MOD_PLSQL looks in the dads.conf for the database to connect to. The rest of the URL contains the PL/SQL package or function to call, e.g.
/mydocuments/scott.p123.procABC&emp_id=123
Where procedure procABC is called in packahe p123 in the Scott Schema. The variable EMP_ID is passed with value 123. Thus the procABC definition will look as follows in the package:
procedure procABC( emp_id number ) is
begin
.. code..
end.;This proc can write HTML data - i.e. dynamically create a HTML response page. Or it can create an image or any other Mime type that the browser can handle.
Oracle provides a bunch of PL/SQL packages for that. E.g. HTP and OWA packages.
This in a nutshell is how MOD_PLSQL is used to allow a web browser URL to be passed directly to a PL/SQL procedure.
The details are under Oracle Application Server PL/SQL (http://download-west.oracle.com/docs/cd/B14099_12/web.htm). Two manuals are applicable:
* mod_plsql User's Guide
* Web Toolkit Reference
It is not complex to use. In fact, the easiest to get all this going is to install HTMLDB 2. Download the Oracle 10G Companion CD for your platform (even if you're running 9.2). Install HTMLDB (this will be in a seperate ORACLE_HOME and will not touch your existing Oracle installation). This will install the latest Apache and MOD_PLSQL software from Oracle. Also HTMLDB is built entirely on MOD_PLSQL and with PL/SQL procedures - and serves as Microsof Access type web interface for developing and deploying web applications on Oracle using PL/SQL. Very powerful and flexible and very easy to use.

Using query strings in links mysite.verizon web pages

Has anyone had trouble using query strings in links to other web pages your personal web pages? I have a link to another web page where I pass an ID using query string but when I click on the link I get a 404 error even though the web page exists (the link appears fine).

Hello jillibee,
Maybe this link will provide some assistance for you. http://forums.verizon.com/t5/FiOS-Internet/Personal-Website-Access/m-p/277209/highlight/true#M19266
Thanks,
Shamika_Vz
Verizon Support
Notice: Content posted by Verizon employees is meant to be informational and does not supersede or change the Verizon Forums User Guidelines or Terms or Service, or your Customer Agreement Terms and Conditions or Plan.

Query using link server to an Oracle database failswhile similar query works

I have two queries almost the same that use a link to an Oracle database. One works ok, the other is a copy of the first with a few fields selects removed. The first one works fine, the copy fails with: Cannot get the data of the row from the OLE DB
provider "OraOLEDB.Oracle" for linked server "linkname".
They are so similar, why would one work but not the other?
First query, this one works:
SET
ANSI_NULLS ON
GO
SET
QUOTED_IDENTIFIER ON
GO
CREATE
PROCEDURE [dbo].[procedure_name]
-- Add the parameters for the stored procedure here
@Begindate Datetime,
@Enddate Datetime
AS
BEGIN
--SET NOCOUNT ON added to prevent extra result sets from
--interfering with SELECT statements.
SET NOCOUNT
ON;
-- Insert statements for procedure here
select p.patientid, p.testid,
mdx."Dx_4", td.acquisitiondatetime_dt, td.democompletedatetime_dt,
td.editdatetime_dt,
CAST(STUFF(STUFF(mn."Arrival_Time_9",
9, 0,
' '), 12, 0,
AS datetime)
as Arrival_date,
CAST(STUFF(STUFF(mt."Comp_D_T_22",
9, 0,
' '), 12, 0,
AS datetime)
as Order_date, td.confirmdatetime_dt
(CAST(DATEDIFF
(MINUTE,
CAST (STUFF
(STUFF
(MN."ARRIVAL_TIME_9", 9,0,
' '), 12, 0,
AS DATETIME), TD.DEMOCOMPLETEDATETIME_DT)
as char(10))
+ 'min')
AS "ARRIVAL TO COMPLETE",
(CAST(DATEDIFF
(MINUTE,
CAST (STUFF
(STUFF
(MT."COMP_D_T_22", 9,0,
' '), 12, 0,
AS DATETIME), TD.DEMOCOMPLETEDATETIME_DT)
as char(10))
+ 'min')
AS "ORDER TO COMPLETE"
from linknametosqldb.sqldbname.dbo.tstpatientdemographics
as p inner
join linknametosqldb.sqldbname dbo.tsttestdemographics
as td on p.testid
= td.testid
left
outer join linknametooracledb."M01Tab01_Main" mn
on td.visitnumber
= mn."Pt_Acct_5"
inner
JOIN linknametooracledb."M01Tab06_Ord" mt
ON mn."Master_Recid_1"
= mt."Master_Recid_1"
inner
join linknametooracledb."M01Tab09_Dx" mdx
on mn."Master_Recid_1"
= mdx."Master_Recid_1"
where td.location
in (2, 4)
and mt.order_4
like 'EKG%'
and mdx."Dx_4"
like '%Chest Pain%'
and
(td.acquisitiondatetime_dt
>= @Begindate and td.acquisitiondatetime_dt
< @Enddate)
order
by p.patientfullname_last, p.patientfullname_first
END
GO
Second query, copied from first with some fields selects removed, this does not work:
SET
ANSI_NULLS ON
GO
SET
QUOTED_IDENTIFIER ON
GO
CREATE
PROCEDURE [dbo].[procedure_name_2]
-- Add the parameters for the stored procedure here
@Begindate Datetime,
@Enddate Datetime
AS
BEGIN
--SET NOCOUNT ON added to prevent extra result sets from
--interfering with SELECT statements.
SET NOCOUNT
ON;
-- Insert statements for procedure here
select p.patientid, mdx."Dx_4",
td.democompletedatetime_dt,
CAST(STUFF(STUFF(mn."Arrival_Time_9",
9, 0,
' '), 12, 0,
AS datetime)
as EDIS_Arrival_date,
(CAST(DATEDIFF
(MINUTE,
CAST (STUFF
(STUFF
(MN."Arrival_Time_9", 9,0,
' '), 12, 0,
AS DATETIME), TD.DEMOCOMPLETEDATETIME_DT)
as char(10))
+ 'min')
AS "ARRIVAL TO COMPLETE"
from linknametosqldb.dbo.tstpatientdemographics
as p inner
join linknametosqldb.dbo.tsttestdemographics
as td on p.testid
= td.testid
left outer
join linknametooracledb."M01Tab01_Main" mn
on td.visitnumber
= mn."Pt_Acct_5"
inner
JOIN linknametooracledb."M01Tab06_Ord" mt
ON mn."Master_Recid_1"
= mt."Master_Recid_1"
inner
join linknametooracledb M01Tab09_Dx" mdx
on mn."Master_Recid_1"
= mdx."Master_Recid_1"
where td.location
in (2, 4)
and mt.order_4
like 'EKG%'
and mdx."Dx_4"
like '%Chest Pain%'
and
(td.acquisitiondatetime_dt
>= @Begindate and td.acquisitiondatetime_dt
< @Enddate)
order
by p.patientid
END
GO
Thanks very much for any help with this.

* Consider putting parentheses around the 3 oracle tables so the 3 of them can possibly be treated as a unit. Assuming there's some performance impact due to the linked server, this MIGHT help, may do a little joining over on the Oracle side of
the world. In either case, it's a little cleaner. They syntax below, including a few extra line breaks for clarity, should be correct. (try the change to the ON clause suggested above too, though).
The parentheses will not help much. SQL Server will remote (as the term is) the query to the Oracle instance if this appears to have a lower cost and there is nothing the blocks remoting.
On the other hand, OPENQUERY can be an alternative to ensure that this happens:
left outer join
OPENQUERY (linknametooracledb,
   'SELECT ...
FROM   "M01Tab01_Main" mn
JOIN   "M01Tab06_Ord" mt ON mn."Master_Recid_1" = mt."Master_Recid_1"
JOIN   "M01Tab09_Dx" mdx on mn."Master_Recid_1" = mdx."Master_Recid_1"') mn on td.visitnumber = mn."Pt_Acct_5"
But this presumes that you there are no columns from the SQL Server side that are used to select data from the Oracle side. That is, the above could lead to far much data being brought across the wire.
Then again, this does not really seem to relate to the original problem where a stored procedure was not working at all. That problem is not very easy to assist with since code post is the not the authentic code as pointed out in the thread. For instance
inner JOIN linknametooracledb."M01Tab06_Ord"
Has nothing to do with Oracle, but accesses a table in the schema linknametooracledb.
I think to help dmcm, you need a really good crystal ball - at least as long he/she does not share the actual code.
Erland Sommarskog, SQL Server MVP, [email protected]

Ps query results to xml using xml link function registry

In People tools version 8.46 we need something like where the ps query result output is in the form of XML so that it can be used for external system. The catch I cannot use webservices. It is something like opening peoplesoft queires form excel using hyperlink but in this case we need xml. I looked at David Vandiver's Excel XML libraries which open the data in excel but is there something where the hyperlink output is xml using xml link function registry. I have capatured the data in rowset and now need to create mime type of xml to write the data
thanks in advance
vinn

If I understand correctly, you want to use something like Microsoft Excel's web queries to get data from a PeopleSoft query in XML format. I have never done this before, but if I were to implement this, here is what I would do in PT 8.46:
1. Create an unstructured message in app designer.
2. Create a user that only has access to the queries you want to expose to Excel
3. Write synchronous message handler PeopleCode to call SwitchUser to switch to a specific query user rather than the generic integration broker user and execute a query, returning the results in XML (see [Query.RunToRowsest|http://download.oracle.com/docs/cd/E13292_01/pt849pbr0/eng/psbooks/tpcr/htm/tpcr31.htm#d0e113933] for an example).
4. Call the message through IB using the HTTP GET URL format as defined in PeopleBooks: [http://download.oracle.com/docs/cd/E13292_01/pt849pbr0/eng/psbooks/tibr/book.htm?File=tibr/htm/tibr33.htm#H4026|http://download.oracle.com/docs/cd/E13292_01/pt849pbr0/eng/psbooks/tibr/book.htm?File=tibr/htm/tibr33.htm#H4026].

Find out query using specific DB link

Hello,
Environment: 11G and Ksh script
Currently our Client is upgrading to new environment, so I need to find out which query in our Database use DB link to pull data from that client will be impacted.
for example,
select * from table_1@DB_link_1All our SQL are embedded in UNIX script, so I use command line to search any scripts contain '@DB_link_1' but not find anything.
is it possible to search any dynamic view or any suggestion to search inside ORACLE database to see if any query using this 'DB_link_1'? I tried 'dba_db_links' but this does not show which query is using which DB link.
Thanks,
Thomas

frank wrote:
Hello,
Environment: 11G and Ksh script
Currently our Client is upgrading to new environment, so I need to find out which query in our Database use DB link to pull data from that client will be impacted.
for example,
select * from table_1@DB_link_1All our SQL are embedded in UNIX script, so I use command line to search any scripts contain '@DB_link_1' but not find anything.
is it possible to search any dynamic view or any suggestion to search inside ORACLE database to see if any query using this 'DB_link_1'? I tried 'dba_db_links' but this does not show which query is using which DB link.
Thanks,
Thomashttp://download.oracle.com/docs/cd/E11882_01/server.112/e17110/statviews_1068.htm#REFRN20053
REFERENCED_LINK_NAME VARCHAR2(128) Name of the link to the parent object (if remote)
"

Issue using data link in XML Publisher

Dear All,
I am new to XML Publisher, for the first time trying data link concept in XML Publisher tool. The issue is very simple, I have two queries Q1(say it fetches sales_order_number, application_name ) and Q2 (say it fetches sales_order_number , error_number, error_description). And sales_order_number is passed as input using data link to Query Q2 from query Q1 and Q2 will retrive more than one row for a given sales_order_number.
Say for a given dept_name 1001 the output will be something similiar like
Q1:
SO# | Appl.Name
1001 | 'Invoice '
Q2:
SO# | Error.# | Error Desc
1001 | 1512 | No Data Found
1001 | 1513 | Invalid Entry
1001 | 2674 | Error while inserting the record
and the query which I have used in XML template goes like this
<dataQuery>
<sqlStatement name="Q1">
<![CDATA[select sales_order_number, application_name
                  from T1
                 where sales_order_number = :in_sales_order_num]]>
</sqlStatement>
<sqlStatement name="LINES">
<![CDATA[
select sales_order_number , error_number, error_description
from T2]]>
</sqlStatement>
<link name="Dummy_LINK" parentQuery="Q1" parentColumn="sales_order_number" childQuery="Q2" childColumn="sales_order_number" condition="="/>
</dataQuery>
<daraStructure/>
And the grouping is as follows
<group name="G_Q1" source="Q1">
     <element name="sales_order_number" value="sales_order_number"/>
     <element name="application_name" value="application_name"/>
<group name="G_Q2" source="Q2">
     <element name="sales_order_number" value="sales_order_number"/>
     <element name="error_number" value="error_number"/>
     <element name="error_description" value="error_description"/>
</group>
</group>
And in the RTF template I have properly used <?for-each:G_Q1?> <?for-each:G_Q2?>- --
<?end for-each?><?end for-each?>
Ideally the header (Q1) should have one row and the lines (Q2) should have 3 rows. But the issue here is even Q1 produces 3 rows.
If in case Q2 has 4 records, then Q1 also plots 4 records in the report output. Hope I am explaining the issue correctly. Any help to resolve the issue would be a great. Thanks in advance.
Regards,
Antony.

try like this
<?if: (TYPE='A') or (TYPE= 'B')?><TABLE STRUCTURE><?end if?>
or
<?if:contains('A~B', TYPE )?><TABLE STRUCTURE> <?end if?>

Using database links in a plsql page process

I am trying to use a database link in a pl/sql page process.
It works fine when I use the link name in the plsql like this:
BEGIN
SELECT
CARD_ID
into
:P12_HDR_CARD_NUMBER
from     [email protected];
But the link name will be coming from a page item (P12_DBLINK) populated as follows:
select db_link d, db_link r from user_db_links;
I have tried the following and it's not working:
DECLARE
l_link VARCHAR2(30);
BEGIN
l_link := :P12_DBLINK;
SELECT
CARD_ID
into
:P12_HDR_CARD_NUMBER
from     fusion.EXM_CC_COMPANY_ACCOUNTS@l_link;
this gives me:
ORA-04052: error occurred when looking up remote object FUSION.EXM_CC_COMPANY_ACCOUNTS@L_LINK.WORLD ORA-00604: error occurred at recursive SQL level 3 ORA-02019: connection description for remote database not found
I thought that maybe the link name was getting appended with .WORLD automatically but the link name already has .WORLD at the end. I tried stripping that off first but the error is the same.
Is there some synatx for this that will work for this? Will I need to use dynamic sql?
Any suggestions are most appreciated.
thanks,
john

Hi John,
check the following example to use the execute immediate statement for your purpose
sql - Variable for Database Link Name - Stack Overflow
let me know if this answers your query in the current thread

How to find the list of existing tables in a schema using DB link?

Hi
I know how to find the list of existing tables in a schema using the following query
SQL> select * from tab;
but, how to list the tables using a DB link?
For Example
SQL> select * from tab@dblink_name;
why this doesn't work?
Pl advice me
Thanks
Reddy.

ORA-02019: connection description for remote database not foundHave you used this database link successfully for some other queries?
The error posted seems to indicate that the DB Link is not functional at all. Has it worked for any other type of DML operation or is this the first time you ever tried to use the link?

Cannot kill a session that is running a query on a linked server in SQL Server 2008 R2

Hi,
Cannot kill a session that is running a query on a linked server in SQL Server 2008 R2.
When I try to kill, it status shown as "KILLED/ROLLBACK"
I have facing the issue from long back. I searched many and none of them worked for me
Any help is greatly appreciated
Thanks
Jaison
Carver

I guess its stuck, can you see some blocking or any other process blocking the rollback.
can you check rollback status using below command. Or you can use sp_who2 and check status column
select percent_complete,estimated_completion_time from sys.dm_exec_requests
where spid=xxx
Please mark this reply as answer if it solved your issue or vote as helpful if it helped so that other forum members can benefit from it.
My TechNet Wiki Articles

Error while accessing Query using Query Analyzer

dear experts...
while accesiing the query using query analayzer...
we are getting below error...
What has happened?
URL http://xxx.xxx.xxx.xx:XXXX/irj/servlet/prt/portal/prtroot/pcd!3aportal_content!2fcom.sap.pct!2fplatform_add_ons!2fcom.sap.ip.bi!2fiViews!2fcom.sap.ip.bi.bex call was terminated because the corresponding service is not available.
Note
The termination occurred in system BI1 with error code 404 and for the reason Not found.
The selected virtual host was 0 .
What can I do?
Please select a valid URL.
If you do not yet have a user ID, contact your system administrator.
ErrorCode:ICF-NF-http-c:001-u:ANAND-l:E-i:PSRCCPRDA003_BI1_00-v:0-s:404-r:Notfound
HTTP 404 - Not found
Your SAP Internet Communication Framework Team
thanks for helping me...
anand

Hi friends,,
instead of getting the link like
http://128.222.125.57:9000/sap/bw/bex?cmd=ldoc&infocube=ZMC_SRH1&query=AGINGV21A&sap-language=EN(working link)
am getting below link...which is getting an error....
http://128.222.125.57:9000/sap/bw/://:/irj/servlet/prt/portal/prtroot/pcd!3aportal_content!2fcom.sap.pct!2fplatform_add_ons!2fcom.sap.ip.bi!2fiViews!2fcom.sap.ip.bi.bex?QUERY=AGINGV21A
above marked is the unwanted thing so how can i change my link in query designer???
please suggest me firneds

Unable to create new query using query manager

Hi friends,
I have been trying to create a query using query manager for couple of hours but still not able to.I am following the instructions given in oracle Peopletools 8.52 : Peoplesoft query
Chapter - creating new queries.
Below are the steps I am going through to create new query:
Step - I open the component Reporting Tools-> Query->Query Manager , I don't get the tabbed pages one for search an existing query and another for create new query.
Please follow this link to see how the page is displayed in first step:
http://uploadpic.org/v.php?img=EvMvVAXX1E
Step 2 - When I click on create new query link, I am redirected to a page where it asks for record name to add in the query. But, even this page is not displayed as how it supposed to be
http://uploadpic.org/v.php?img=GzHh3f6krU
Step 3 - Following the above step, when I click on Person record to add into my query I am asked to select the fields that I want to display in the output.
But I somehow do not get the proper tabbed pages where individual pages are there to add the attributes to complete the query like adding multiple records, fields, query, expressions, prompt...etc
http://uploadpic.org/v.php?img=Wbbla3Q3jE
I am neither able to select multiple records in my query nor able to customize my query to get the desired results.
Below is the query that I want to create using query manager:
SELECT P.EMPLID,P.BIRTHDATE, N.NAME, A.ADDRESS1, A.ADDRESS2, A.CITY
FROM PS_PERSON P, PS_NAMES N, PS_ADDRESSES A
WHERE P.EMPLID = N.EMPLID AND
N.EMPLID = A.EMPLID AND
P.BIRTHDATE BETWEEN to_date('1990/1/1','yyyy/mm/dd') and to_date('1991/1/1','yyyy/mm/dd');

Hi,
As I cannot access your screenshot by the blocking of company firewall rules.
I'm guessing currently you are using PT 8.52.00 version, correct?
This should be a bug, that you need to apply the 8.52.0X patch, not sure which patch fix this, you can apply the latest one to solve this issue for no tab page of query manager. (PT 8.52.06)
Hope this helps.
Thanks,
Saxon SI

Why Using Top Link is best in DB Adapter?

Hi All,
Can any one suggest , Why Using Top Link (Build-in Insert, Select etc. operation) is best in DB Adapter over using custom query?
Thanks

Hi Vikky,
for insert/select it depends on what kind of user you are. TopLink lets you browse and click on a tables and have everything generated for you. If you are more a DBA or show me the SQL type then you can just type SQL directly.
Some advantages of TopLink would be:
-The range of SQL generated by TopLink is limited, but if you hard code complex SQL into your service you need to maintain it.
-TopLink can generate at runtime the correct SQL for a given database, making switching from say DB2 to Oracle easy.
-The merge operation will compare the input XML to the columns on the database and update only what has changed. It can also do a sparse merge. If only 4 columns in the XML were set, only those 4 columns in the database will be updated.
-For inbound polling the strategy used (LogicalDelete, Sequencing Table, etc) is a configuration property and then at runtime multiple SQL statements are generated. The SQL also takes advantage of advanced syntax like the Oracle-only FOR UPDATE SKIP LOCKED, writing it all yourself may be tedious and error prone.
-The main benefit of TopLink is when you go beyond thinking about a single table. If you import multiple related tables at once, TopLink will generate the SQL to select from and maintain multiple tables, establish a commit order, and generate a hierarchical XSD. With custom SQL the matching XSD is always flat. I.e. if you just need to insert an emp you could get away with custom SQL. If you need to insert a dept and emps, I would use TopLink.
-This is also when the intermediary abstraction of an object/table makes more sense, as you only need to import a complex relational schema once, then generate inserts, selects, etc.
So I hope that helps. They are each equally viable and can do something that the other can't. Where you see an overlap I would go with personal preference.
Thanks
Steve

Accessing Tables using DB link .

Hello Friends,
I have a development and production env.
I created a public DB link in Development say mylink .
Trying to access the developement table from production using the following command say to access table x
From PROD accessing DEV table x using dblink
select * from x@mylink ;
Got the error as :
ORA-02019: connection description for remote database not found
02019. 00000 - "connection description for remote database not found"
Any idea on this why this error is coming ..
thanks/kumar

I have a development and production env.
I created a public DB link in Development say mylink .
Trying to access the developement table from production using the following command say to access table x
From PROD accessing DEV table x using dblink
select * from x@mylink ;You are trying to execute the query in PROD database. This query uses a DB link which points to the DEV database.
Very simple to understand : Since you are using the DB LINK in PROD database, obviously it should be in PROD area... right?
Just imagine you are using an object (the 'DBLINK') in PROD and it doesn't exist here. Got it?
SELECT * FROM x@mylink;
Here,
x -- table in DEV database
mylink -- DB LINK in PROD database (pointing to DEV)
Please try this.
HTH
Edited by: ranit B on Dec 25, 2012 12:24 AM

Paralell query using db links

Similar Messages

Maybe you are looking for