We have a MS SQL database with an Oracle linked server 'ALTTEST'
We can Select, Insert, Delete and Update tables on the Oracle Db using OpenQuery, but how do I apply a table lock with a transaction?
I've tried applying the code below, but it doesn't work.
Any help appreciated.
BEGIN TRAN
SELECT * from openquery(ALTTEST,'select LAST_PIN_NUMBER from sys_params') WITH (TABLOCKX)
COMMIT
Hi All -
My Set up:
Server A - Oracle 10g Database
Server B - SQL2005
Client PC - Sql Express
Server A holds all data. I am using a linked server to connect server A and B. I use a set of stored procedures containing the openquery() syntax to get data from Server A to Server B. These stored procedures run every 20 minutes. I then create a publisher on Server B. I subscribe from the client PC to publisher to get data down from Server B to client(Download only subscription).
When I fire up the stored procedures and attempt to replicate, everything works fine. It appears after about 4-5 hours of the stored procedures running replication begins to hang more and and more until eventually it hangs for about 10 minutes and I recieve the following error:
Command attempted:
{call sp_MSreleasemakegenerationapplock}
Error messages:
The merge process was unable to create a new generation at the 'Publisher'. Troubleshoot by restarting the synchronization with verbose history logging and specify an output file to which to write. (Source: MSSQL_REPL, Error number: MSSQL_REPL-2147200994)
I'm not sure if there is an error with openquery() locking? There is some locking activity going on with the TempDB and and Server B database. I've also come across some threads talking about the agent profiles. I'm very new to replication and very confused by all of the options in the agent profiles. Any help would be greatly appreciated!
I created a stored procedure like this:CREATE PROCEDURE SPASBEGINCREATE TABLE #T( C INT )INSERT INTO #T(C) VALUES (1)SELECT * FROM #TENDWhen I call it this way: EXEC SP, it works ok.But when I do it like this:SELECT * FROM OPENQUERY( MYSERVER, 'EXEC SP')I receive an error: Invalid object name '#T'Why?...*** Sent via Developersdex http://www.developersdex.com ***Don't just participate in USENET...get rewarded for it!
View 2 Replies View RelatedHi,
I have two queries as under:
QUERY 1:
SELECT * FROM
OPENQUERY(MYSERVER, 'SELECT * FROM SCOPE() WHERE FREETEXT(''Text to Search'')') AS Docs
QUERY 2:
SELECT MediaID, LawID, LawDate, Agreement, Name, NameSearch, LawType, LawNo, RegID, IssueNo, Attachment
From Dept_LegalLaw
INNER JOIN Dept_LegalMinistries ON Dept_LegalLaw.RegID = Dept_LegalMinistries.RegID
INNER JOIN Dept_LegalLawType ON Dept_LegalLaw.LawID = Dept_LegalLawType.LawID
WHERE 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1
both queries are working fine separately and I can generate the desired results separately, but I cannot merge them to get one single result for example, after a FREETEXT search I want to get some values from my local table (Query2) to display as one result. Like in Query1 FREETEXT will search the web page based on the given text and in Query2 will select the remaining data from the local database (Title, ID, Name etc etc).
Try many thing but no success yet.
Need urgent help.
Thanks
Hi,
I have two queries as under:
QUERY 1:
SELECT * FROM
OPENQUERY(MYSERVER, 'SELECT * FROM SCOPE() WHERE FREETEXT(''Text to Search'')') AS Docs
QUERY 2:
SELECT MediaID, LawID, LawDate, Agreement, Name, NameSearch, LawType, LawNo, RegID, IssueNo, Attachment
From Dept_LegalLaw
INNER JOIN Dept_LegalMinistries ON Dept_LegalLaw.RegID = Dept_LegalMinistries.RegID
INNER JOIN Dept_LegalLawType ON Dept_LegalLaw.LawID = Dept_LegalLawType.LawID
WHERE 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1
both queries are working fine separately and I can generate the desired results separately, but I cannot merge them to get one single result for example, after a FREETEXT search I want to get some values from my local table (Query2) to display as one result. Like in Query1 FREETEXT will search the web page based on the given text and in Query2 will select the remaining data from the local database (Title, ID, Name etc etc).
Try many thing but no success yet.
Need urgent help.
Thanks
Hello,
I have an application that uses a MS SQL 2005 database. When data is changed in certain tables, that data needs to be pushed to a MySQL box. I've added the MySQL server as a linked server in SQL 2k5 and I can delete and insert data with no problem, however when I try to update I get the following error (query included):
Code Snippet
with RemoteTable(r_AccountID, r_Name)
as (select AccountID, Name from openquery(RACS_TEST, 'select AccountID, Name from accounts'))
update RemoteTable
set r_Name = ex_Name
from Export_RACS_Accounts join remotetable
on r_AccountID = ex_AccountID
OLE DB provider "MSDASQL" for linked server "RACS_TEST" returned message "Row cannot be located for updating. Some values may have been changed since it was last read.".
Msg 7343, Level 16, State 4, Line 33
The OLE DB provider "MSDASQL" for linked server "RACS_TEST" could not UPDATE table "[MSDASQL]". The rowset was using optimistic concurrency and the value of a column has been changed after the containing row was last fetched or resynchronized.
What am I doing wrong ?
I'm running a query on Oracle using OPENQUERY.
I need to pass a parameter.
Which means I'm having to use dynamic sql.
In turn this requires a ##table due to the connection issue of using EXEC.
There is surely a better way of doing this, but what?
Something Like:
DECLARE
@TSQL VARCHAR(8000)
DECLARE
@VAR TABLE (VAR1
VARCHAR (2))
INSERT
INTO @VAR
values ('CA'),('OR')
[Code] ....
Hi, I have two queries as under:QUERY 1:SELECT * FROM OPENQUERY(MYSERVER, 'SELECT * FROM SCOPE() WHERE FREETEXT(''Text to Search'')') AS DocsQUERY 2:SELECT MediaID, LawID, LawDate, Agreement, Name, NameSearch, LawType, LawNo, RegID, IssueNo, AttachmentFrom Dept_LegalLawINNER JOIN Dept_LegalMinistries ON Dept_LegalLaw.RegID = Dept_LegalMinistries.RegIDINNER JOIN Dept_LegalLawType ON Dept_LegalLaw.LawID = Dept_LegalLawType.LawIDWHERE 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1 AND 1=1both queries are working fine separately and I can generate the desired results separately, but I cannot merge them to get one single result for example, after a FREETEXT search I want to get some values from my local table (Query2) to display as one result. Like in Query1 FREETEXT will search the web page based on the given text and in Query2 will select the remaining data from the local database (Title, ID, Name etc etc).Try many thing but no success yet.Need urgent help.Thanks
View 2 Replies View RelatedHello Everybody / Anybody,
Sorry but exiting problem!
The Problem: OPENQUERY throwing error [Error 7357]when the source SP uses temporary table.
Description : Need to validate data against master list. My combo on UI has a source Stored Proc(contains a temp table in it).
I'm importing data from Excel. Before import, I want to validate it against my master list values.
[say field Priority has master values "High, Medium,Low".] and in excel user has added 'ComplexHigh' under priority field]
In this case, my import validator StoredProc should not accept value 'ComplexHigh' as it is not present in my Priority master list]
I'm preparing a temp table tabName containing o/p of SP, it works fine zakkas if my SP usp_SelectData does not contain temp table.
I think you got what the situation is!! Woh!
Note : I have searched net for this and found nothing! So its challenge for all of us. TRY OUT!!
------------------------------------- The Code ----------------------------
create proc usp_SelectData
as
create table #xx (FixedCol int)
insert into #xx select 1 union select 2
select * from #xx
drop table #xx
create proc usp_SelectData2
as
create table xx (FixedCol int)
insert into xx select 1 union select 2
select * from xx
drop table xx
-- Please replace MyDB with your current Database
SELECT * INTO tabName FROM OPENQUERY('EXEC MyDB.dbo.usp_SelectData')
-- Throws Error 7357 : [Could not process object 'EXEC MyDB.dbo.usp_SelectData'. The OLE DB provider 'SQLOLEDB' indicates that the object has no columns.]
SELECT * INTO tabName FROM OPENQUERY('EXEC MyDB.dbo.usp_SelectData2') -- Works fine
Thanks in advance...
One of our Oracle Tables changed and I am wondering if there's any way that I can query all of our Stored Procedures to try and find out if that Oracle Table Name is referenced in any of our SQL Server Stored Procedures OPENQUERY statements?
View 2 Replies View RelatedGurus,
I am trying to execute this stored procedure when I try to change all occurences of a field in a table.
(
@Dept char(8),
@DDept char(8)
)
As
Set NoCOUNt On
Begin
Select '@DDept'
update phodept set fo_dept = @Ddept
where fo_dept = @Dept
update phone set fo_dept = @Ddept where fo_dept = @Dept
End
GO
The table/database is being used by others, generally in a read only mode.
via a VB 5.0 FE program.
The Stored procedure, when it is invoked, just hangs like it is waiting for exclusive use of the table.
Is there a way around it, without doing major surgery on the VB code?
Thanks.
Sam
Hello,
I am working on the implementation of a database for my company and I have
a simple (I hope) question to ask.
I have the following stored procedure that will be running when I want
(actually when I want to delete a customer).
Do not mind if you do not understand what this procedure does... ;-)
Actually the Level8View is a VIEW of a nested table
(CustomerData->CustomerData).
CREATE Procedure DeleteCustomer
@ClientID INT
AS
UPDATE Level8View
SET UpCode = Level2, Level1=Level2, Level2=Level3, Level3=Level4, Level4=Level5, Level5=Level6, Level6=Level7, Level7=Level8
WHERE Level1=@ClientID
UPDATE Level8View
SET Level2=Level3, Level3=Level4, Level4=Level5, Level5=Level6, Level6=Level7, Level7=Level8
WHERE Level2=@ClientID
UPDATE Level8View
SET Level3=Level4, Level4=Level5, Level5=Level6, Level6=Level7, Level7=Level8
WHERE Level3=@ClientID
UPDATE Level8View
SET Level4=Level5, Level5=Level6, Level6=Level7, Level7=Level8
WHERE Level4=@ClientID
UPDATE Level8View
SET Level5=Level6, Level6=Level7, Level7=Level8
WHERE Level5=@ClientID
UPDATE Level8View
SET Level6=Level7, Level7=Level8
WHERE Level6=@ClientID
UPDATE Level8View
SET Level7=Level8
WHERE Level7=@ClientID
DELETE FROM Customers
WHERE ClientID=@ClientID
GO
I checked this procedure and works perfectly.
What I want to do is to somehow lock the WHOLE TABLE (CustomerData) or the
view (Level8View) before the Update statements and unlock it after the
delete statements.
I do not want to do a Row by Row lock, or Page by Page lock since the updates
in this table occur in the whole table and during that operation I do not
want other threads to issue SELECT, INSERT or UPDATE statements.
Can someone answer me how I can lock and unlock the whole view or table
please?
I will appreciate it for your answer. Thanks.
Yours, sincerely
Efthymios Kalyviotis
ekalyviotis@comerclub.gr
I have a SP for an e-commerce site that creates an order doing the following fashion:
begin
-Fill out some variables
*Critical Section
-Create Order number
-While loop select on order table and recreate if ordernumber already exists
-Insert into order table
*End Critical Section
-Insert into order lines table, address table, etc.
commit
The problem is the Create Order number and While loop used to be at the top and occasionally I would get duplicate order numbers if two submitted at the same time. I rearranged it to the above and it happens less now.
My question is if there is a way to put an XLOCK on the orders table only during the critical section lines. Also my understanding is that once the insert into orders is done that the server will hold a lock until the commit in case of rollback.? I don't want to XLOCK the whole SP if I don't have to. It is quick though.
Thanks
I need 2 unique sequential numbers to be stored in the same table. Obviously, for the first number I can use an auto-number field. For the second number, I would like to use a 2nd table with just 1 field that would only ever store 1 record. The idea being that if a user needs to generate a new unique number, they lock the 2nd table, lookup the current value, store the value in a variable for use, add 1 to the value of the number in the table, then unlock the table.
Presumably, I'd use something like this:
SELECT OrderNo
FROM tblUniqueOrderNumber (WITH TABLOCK)
Does this lock the table so no other users can read/write the table? If so, how do I unlock the table once I've updated OrderNo? What happens if another user then reads the table while it has a TABLOCK applied? I'm using SQL Server 2005 and the database is a multi-user database. I have a feeling I'm not going about this the best way possible, but I'd like to know how the table lock works eitherway.
hi
can i lock my table while doing an updation so that all others in n/w can insert into the table only after my operation is completed?
if a user attempts to save/update the 'locked'table, will he get an error message or will his application (using ADO) will wait till the table gets unlocked?
Can any one please tell me how to lock & unlock a table a from inside a store procedure using T-SQL
thanks
hi,
I am running a sql job that select data from one database (multiple table) and inserts into another database (1 table)
I have a problem with the way it is locking the tables I am selecting from !
my query is like :
insert into testdb.dbo.tabletest
select * from
livedb.dbo.tablex
inner join liveb.dbo.table1 on tablex.id=table1.xid
It seems to be locking the table and blocking other users from accessing the table.
suppose my question is why is the select locking the table ?
I want to insert values into a table and the same time select a valuefrom the same table.I insert a record of a parent type and use a function to create asubscriberType for the record.Next i insert a record of child type and want to select thesubscriberType from the parent record.This do not work when i useINSERT TABLESELECT ,Name,"function to generate subscribertype",.............I DO NOT WANT TO USE CURSOR!!
View 1 Replies View RelatedSo randomly every 1 to 6 days queries start timing out and I'm almost positive it's from an improperly terminated transaction
Is there a way to snoop this out the next time it happens? Like when a table's locked I can look and see yea this is the transaction it's in the middle of?
Hi,
im developing an application for my university final year project and i want to lock the table before the query executes to ensure that no other changes are allowed (sorry if im confusing)
So my question , really does anyone know how to do this and is it possible ?, for example in coldfusion a query is surrounded with cflock or something similiar.
Appreciate any and all help guys
Many thanks in advance
andy
Hi
Is a TABLOCK HOLDLOCK in a transaction the best way for locking a table for a brief period of time?
We need to get generate a unique serial number each time a certificate record is printed. We have in place a field called Field_SERIAL_NUMBER for each certificate record, which may be printed 1 time or 100 times. Each time it prints, we will insert a unique number into the Field_SERIAL_NUMBER field.
This is what I'm thinking to approach this.
- Create a new table called Table_SERIAL with one field called Field_NUMBER which has only 1 record set to 1.
- Every time a certificate is printed, we
1. Begin transaction
2. Select @number = Field_NUMBER from Table_SERIAL with a TABLOCK HOLDLOCK
3. Insert that number into the Field_SERIAL_NUMBER field in the certificate table
4. Increment @number by 1
5. Update Field_NUMBER with @number in Table_SERIAL
6. Unlock table
So, everytime the Field_NUMBER is pulled, it should be unique and incremented by 1.
Does this approach sound good?
Joyce
4. Increment
Hi,
is there anyway i can run delete statements on an audit table that will not lock the whole table?
As it is an audit table, when i run the delete query, the whole application freezes until the delete query is complete.
the process is taking several minutes as the table is approx 40M rows, the where clause in the delete is indexed but thats it really.
any thoughts greatly appreciated by my system users!
Hi,
I want to insert a record in a table having an identity column as primary key. I want to lock the table while inserting. so that no one should be able to insert, select, update, delete from the table. and once my insert is over, then will release the lock.
Can I have the code for the same. M using SQL SERVER 2005.
Thanks,
Rahul Jha
Hi All,I am using a SQLTransaction to insert records into a table. At onetime, there are 5000 or more records to be inserted one by one. Ittakes some 20-25 mins for the entire process to run.Another application accesses the same table.As long as the insert process within the transaction isn't completed,the second application is not getting any response from the server. Ieven tried to run a SELECT on the table in SQL Query Analyzer while theinsert process was running and it also did not respond till the timethe insert process finished!Is this normal that a transaction is locking up a table? How do Iovercome it? I am using IsolationLevel.ReadUncommitted for thetransaction.If I do not run the process within a SQLTransaction, the second processor running the SELECT in Query Analyzer does not hang.Thanks,Sanjeev Mahajan
View 3 Replies View RelatedSQL Server 2000, MSDE 2000I have a procedure in my application that I would like only one user at atime to be able to run. Is there a TSQL command I can run that willesentially lock a set of tables so others cannot access the table until theuser is done with the procedure or until the user disconnects from thatsession (in case of a hung app, I would like to lock released)TIA--Tim Morrison--------------------------------------------------------------------------------Vehicle Web Studio - The easiest way to create and maintain your vehiclerelated website.http://www.vehiclewebstudio.com
View 3 Replies View RelatedI feel like I read about a new ability in SSRS 2005 to be able to lock the position of a table's header row (when viewing through the report viewer) so that it does not scroll out of view when scrolling down long report. Now that I need it, I have not been able to find any information about it. Does anyone have any information about this feature or know if this feature exists?
View 1 Replies View RelatedHi,
I have the following doubt about table lockinglocking in case of partitioning:-
Say we have 5 partition on the table Employee on the key Joining_Date and when we run 5 select queries on each of the parition in parallel will there be locking on the table when the 1st query is running or all the 5 queries can run in parallel. Basically, I am trying to see if parallelism and partitioning can work in sync or there will be locking at the table level if I don't specify any query hints?
Any help in this regard is highly appreciated.
Thanks,
Ritesh
We have a nightly script that drops and rebuilds a table. Problem is before the script has had a chance to complete the building of the primary key, it fails due to some other process gaining access to the table.
My question is this: how do I lock a table for the duration of the transaction.
Here is what I think would work:
SET ISOLATION LEVEL SERIALIZABLE
BEGIN TRANSACTION
DROP TABLE <table>...
CREATE TABLE <table>...
INSERT INTO <table> WITH (TABLOCKX) ...
ALTER TABLE <table> ADD CONSTRAINT PRIMARY KEY ...
COMMIT
Any insight as to whether or not I am correct, or if I am not, what would be the correct way.
We currently have a large ETL import each night. There are lots of tables and some are quite large. If there is a problem with the nightly import, it is rerun during the day while users are running reports. The only updates to the tables is the nightly import.Currently the import does a BULK Insert and several INSERT INTOs. We are going to try to improve performance by adding a WITH (TABLOCK). But wonder if the TABLOCK will cause more locking, less locking, or be about the same. We do not have transactions batched. We insert all the records for a table in a single transaction.
Have found these references regarding TABLOCK, but I am confused what they mean, and how they differ from doing a BULK INSERT or INSERT INTO without TABLOCK.
TABLOCK - Specifies that a shared lock is taken on the table held until the end-of-statement. [URL] ....
Using INSERT INTO…SELECT to Bulk Load Data with Minimal Logging [URL] ....
Lock Modes [URL] ....
Hi all,
We have an SSIS package which runs for around 30 min and it does a lot of things, now whenever this SSIS Package is run it is suspected that it is locing the table or records from the DB which it is reading the records, and indeed that system is having some fault that it is never able to remove that lock unless the DB restarts. In this source we only have a SQL Command and not a Table selected. Have anyone come across this problem? Any answers will be of great help.
Hi all,
We have a scenario where we want to lock and unlock manually some rows in a table in SQL CE database .First the rows we need should be locked. By using a select query the locked rows will be read and processed in front end. Only if the process is completed in front end, manually I need to unlock rows which I have locked earlier. When rows are in locked state no body should be allowed to access that rows . ( should not allow anybody even to put a select query on that locked rows).
At the same time appending rows in the table should always be allowed.
Is it possible to achieve the above scenario?
Hi all,
We have a scenario where we want to lock and unlock manually some rows in a table in SQL CE database .First the rows we need should be locked. By using a select query the locked rows will be read and processed in front end. Only if the process is completed in front end, manually I need to unlock rows which I have locked earlier. When rows are in locked state no body should be allowed to access that rows . ( should not allow anybody even to put a select query on that locked rows).
At the same time appending rows in the table should always be allowed.
Is it possible to achieve the above scenario?