This is a question concerning query optimisation. Sorry if it's a bit long,
but thanks to anyone who has the patience to help - This is my first post
here...
If I have two tables: 'tblContact' and 'tblCategory' where categories are
like:
Code Name
010101 Short
010102 Fat
010103 Stupid
The junction table 'tblConCat' has fields CctConID, CctCatCode to tell me
which contacts have which category codes. These are nchar(6) fields, if it
makes any difference.
If I need to find all people who are short, fat and stupid I can see two
ways:
Solution One:
SELECT tblContact.* FROM tblContact WHERE
ConID IN (SELECT CctConID FROM tblConCat WHERE CctCatCode='010101') AND
ConID IN (SELECT CctConID FROM tblConCat WHERE CctCatCode='010102') AND
ConID IN (SELECT CctConID FROM tblConCat WHERE CctCatCode='010103')
Solution Two:
Build a helper table which contains the codes I'm looking for
SELECT tblContact.* FROM tblContact
WHERE ConID IN
(SELECT CctConID
FROM tblConCat INNER JOIN tblHelper
ON tblConCat.CctCatCode = tblHelper.HlpCatCode
GROUP BY CctConID HAVING Count(*)=3)
I have tried them both out and although I looked at the query analyzer it
provided more information than I knew what to do with. In practise they
both provide similar working times (I wait about a second) but I thought the
first looked rather inefficient and thought my helper table might help.
There are about 30,000 contact records, 180 category records and 120,000
junction table records.
All I am looking for comments on any pros and cons of these two approaches -
does one look that bad? The database was migrated from Access where the
helper table really did help, but using SQL Server I might not need it.
Thank you for taking the time to read this, I need all the advise and help I can get on this ... so please post anything you think would work ... A little confused I am: Have a database table called "people" with "person name" and "ID" field. My ASP.NET application mainly stores articles in article table. An article's Article text mentions various people's names in different combinations (e.g. John, Smith, John Smith, Smith John, etc) Is there any way, I could compare the article text stored in article table with people table and get the people from people table along with their ID's who have been mentioned in that article? ... so in an article "i love john smith ... and i think Mr smith has always been helpful", I get John Smith back... Not too sure being honest, what is the best way of implementing this, looking for the most efficient way, probably using XML? SQL Query or may be ASP.NET's code behind? Thanks once again for taking the time. Cheers, Tyro
Maintenance Plans Broken Database Mail Broken Reporting Services Report Viewer web part displays US Format dates instead of report design English-Australian. . In Detail:
Microsoft SQL Server 2005 SP2 was downloaded from Microsoft.com. This download occurred after 5 March 2007 and the file version was 9.0.3042.1 (SQLServer2005SP2-KB921896-x86-ENU.exe) During the Installation of SP2 by a Domain Administrator an error occurred on the Database Engine. A Hotfix Exception was raised and the setup completed with errors. The Error Log reported a problem: ---------------------------------------------------------------------------------- Product : Database Services (MSSQLSERVER) Product Version (Previous): 2047 Product Version (Final) : Status : Failure Log File : C:Program FilesMicrosoft SQL Server90Setup BootstrapLOGHotfixSQL9_Hotfix_KB921896_sqlrun_sql.msp.log Error Number : 29534 Error Description : MSP Error: 29534 Service 'MSSQLSERVER' could not be started. Verify that you have sufficient privileges to start system services. The error code is (1067) The process terminated unexpectedly.
The SQL Server service could not be manually started. An error was reported in the event viewer: One or more files do not match the primary file of the database. If you are attempting to attach a database, retry the operation with the correct files. If this is an existing database, the file may be corrupted and should be restored from a backup.
It was noted that a new file mssqlsystemresource.ldf had been copied to the %data% folder. Note: As part of separating data and log files, both the master and mssqlsystemresource databases had been previously moved to another volume %data% whilst their respective log files had been moved to a separate volume also %log%. This had occurred after SP1 was initially installed and months prior to the SP2 install using Microsoft KB articles to detach and reattach the databases. It appears that the SP2 installer does not work correctly if mssqlsystemresource mdf and ldf files are not in the same location. Whilst the Engine could not be started €“ the new mssqlsystemresource.ldf file was manually moved to the %log% file, overwriting the previous version. The SQL Server Service was now able to start without any errors. Since the SQL Server Service has been started two major errors have become apparent: The following error is recorded when trying to create a new maintenance plan: TITLE: Microsoft SQL Server Management Studio ------------------------------
Failed to retrieve data for this request. (Microsoft.SqlServer.SmoEnum)
For help, click: http://go.microsoft.com/fwlink?ProdName=Microsoft+SQL+Server&LinkId=20476
An exception occurred while executing a Transact-SQL statement or batch. (Microsoft.SqlServer.ConnectionInfo)
------------------------------
Invalid column name 'from_msx'. Invalid column name 'has_targets'. (Microsoft SQL Server, Error: 207)
For help, click: http://go.microsoft.com/fwlink?ProdName=Microsoft+SQL+Server&ProdVer=09.00.3042&EvtSrc=MSSQLServer&EvtID=207&LinkId=20476
Also when using the maintenance plan wizard the following error is recorded when the Wizard attempts to save the plan at the end: TITLE: Maintenance Plan Wizard Progress ------------------------------
Failed to retrieve data for this request. (Microsoft.SqlServer.SmoEnum)
For help, click: http://go.microsoft.com/fwlink?ProdName=Microsoft+SQL+Server&LinkId=20476
------------------------------
An exception occurred while executing a Transact-SQL statement or batch. (Microsoft.SqlServer.ConnectionInfo)
------------------------------
Invalid column name 'msx_job_id'. (Microsoft SQL Server, Error: 207)
For help, click: http://go.microsoft.com/fwlink?ProdName=Microsoft+SQL+Server&ProdVer=09.00.3042&EvtSrc=MSSQLServer&EvtID=207&LinkId=20476
------------------------------
Database Mail is also broken on the server. The Database Mail Log reports an €œActivation Failure€? when a test email is sent. (the email does not arrive). The SQL Log and event viewer also report the following message: The activated proc [dbo].[sp_sysmail_activate] running on queue msdb.dbo.ExternalMailQueue output the following: 'Error executing extended stored procedure: Invalid Parameter'
Another error affects Reporting Services reports that are accessed using the SharePoint V2.0 Web parts ( rsWebparts.cab) . The date picker in the Report Viewer web part now only recognises US format dates. It is important to note that the same reports run through the Report Manager web site work correctly using DD/MM/YYYY (English-Australian ) date format for the report picker. It is noted that the rsWebparts.cab file was not updated by SQL 2005 SP2 install. However, the behaviour of the Report Viewer web part has definitely been affected by the SP2 install. Are the first two issues related? Are they due to the fact that mssqlsystemresource mdf and ldf files had been moved to an unexpected location prior to SP2? How can this be fixed? Are the Reporting Services SharePoint (V2) web parts going to be updated? We are not in a position to upgrade our Portal 2003/SharePoint Servcies V2 installation at the present time.
Here is a fairly basic question in which I'd like assistance. In determining a standard for a schema, for simplicity, my thinking is that using the long data type would meet our purposes.
Are there any thoughts on mapping out where we would not need to take advantage of long vs short? I realize the space but for this particular table, space won't an issue.
Is there a way to short cut through the fields in a table when updating them or inserting? ie. I have a table with 78 fields and could grow with time. Is there a way to run the UPDATE or INSERT TO [table](blah....) without listing all the table fields?
Hello, i have database front end is Access, and backend is sql server. The front end, there are two dropbox, called 'CountyLongName', and 'CountyShortName'. I want to: when i select a countyLongName, there it will pop up the exact CountyShortName in anohter field.
For example, when i click Howard Couty in the 'CountyLongNmae', it will automatically display "HW" in the 'CountyShortName" Field. How can i do that? Thanks.
CREATE PROCEDURE MyBooks_Selling ( @MemberID SMALLINT ) AS
SELECT * FROM v_BookInfo_Sellers_Extended WHERE MemberID=@MemberID
GO GRANT EXEC ON MyBooks_Selling TO bto GO
CREATE PROCEDURE MyBooks_Buying ( @MemberID SMALLINT ) AS
SELECT * FROM v_BookInfo_Buyers_Extended WHERE MemberID=@MemberID
GO GRANT EXEC ON MyBooks_Buying TO bto GO
Is there a way to make it so I could combine those 2 prcedures and choose which table i would like to select from based on another input parameter? I tried it that way but it didnt work...so im asking here to make sure
Hi all, I have a view with a datetime column, when viewing the column data the format is:
" 2006-06-12T12:00:00.0000000+03:00 "
while i'm only want to see the date part
(that is, 2006-06-12, and if possible in the format: dd/mm/yyyy) now, when i'm using convert( varchar(10), [column_name],101) i'm getting the format i need but the problem is that i'm "losing" the date type, is it possible to view it as the format i stated and still as a datetime type ??
I have a suite of several SSRS reports that differ essentially on the stored procedure called. All have very the same one-page header block and all have a footer. Those where a large number of pages are produced, page without fault when printed or converted to PDF.
Two page reports often break down on printing, so that page 1 only has the header and a footer and all the detail lines are printed on page2 (together with the footer). However on the screen the report paged correctly.
I have checked all the obvious problems such as page dimensions exceeding that of the print medium. (A series of such problems was fixed earlier in development.)
I have observed this behaviour with SQL2005 SP1 and SP2 (i.e. SP2 did not provide any fix in this area). The host OS for development and production is Windows 2003 server. The tests were done by viewing in IE6 and the fault only became apparent when going to print preview, printing or in PDF production.
This has the appearance of a bug in SSRS hereby intermediate length reports are not handled correctly. It appears to happen when the amount of detail lines just fills page 2. As soon as the report has sufficient lines to be 3 pages long (page 1 is header plus some details lines plus footer, page 2 is full of detail lines plus footer and page 3 is remaining detail lines plus footer then the problem is no longer seen! Where the detail lines will all fit on page 1, then the problem is not seen.
I observe this variation in behaviour since I am able to switch between different sized datasets and thus switch between the different scenarios described above. I should also note that the reports are in landscape mode.
Hello to all,Below the sample code:declare @arg_szMsgText ntextset @arg_szMsgText = isnull(@arg_szMsgText, N'unknown message')Now the error message I get:"The assignment operator operation cannot take a ntext data type as anargument."Explanation:I'm using it in a SP. I was trying thousand things to make it running,and the only way I maked was, I get it when I set the return value fromisnull fucntion to nvarchar or varchar variable.The problem is, the field in DB, where the value of @arg_szMsgText iswritten) is ntext, so I need to use this format.In the description of isnull function (help) is written:----------------------------------------------------Argumentscheck_expressionIs the expression to be checked for NULL. check_expression can be ofany type.replacement_valueIs the expression to be returned if check_expression is NULL.replacement_value must have the same type as check_expresssion.Return TypesReturns the same type as check_expression.-----------------------------------------------------so ite means, I should be able to use this in such a way ...I know I can assign the value in other way (like by if @variable isnull ....), but if there is already function doing this, why not to useit, if it helds the code more clear?Any ideas? Maybe I'm doing something wrong.Matik
Hello all, this is my first post, and I think I've found something interesting between sql 2005 and 2000. Thanks in advance for any help you can provide.
I am attempting to migrate a system from SQL 2000(v 8.00.2039 - Enterprise Edition) to SQL 2005 (v9.0.2047 - Standard Edition ), and a few of my most complex queries are generating errors. I was able to isolate the problem to a fairly simple query that works in SQL 2000, but generates an error in SQL 2005.
I believe that the problem has something to do with how short-circuiting is handled in the WHERE clause. I'm not sure if short-circuiting is really going on when these statements are parsed, but it seems to describe the situation well.
To replicate the problem, first create the following table and fill it with values:
insert into orgview (orgname, qid, answer) values('org1', 120, '1') insert into orgview (orgname, qid, answer) values('org2', 94, '2006-06-06') insert into orgview (orgname, qid, answer) values('org3', 98, 'free kevin') ----------------------------------------------
The second value (org2) contains a date, but the rest do not. Under both systems, the following contrived query works fine:
SELECT orgname FROM OrgView WHERE (QID = 94 AND CONVERT(DATETIME, Answer) = '2006-06-06')
It would seem, that when QID = 94, the expression short circuits, and the CONVERT statement doesn't run. This is good, because the CONVERT statement would throw an error whenever 'Answer' doesn't contain a date. However, if modify the query by adding an OR to the end like this:
SELECT orgname FROM OrgView WHERE (QID = 94 AND CONVERT(DATETIME, Answer) = '2006-06-06') or QID=98
then the query fails on SQL 2005 with a 'Conversion failed when converting datetime from character string.' error. On SQL 2000, the short circuiting works fine, and the above query returns 'org2' and 'org3'. I know shortcircuiting may not be the best way to describe the set calculus going on in the background, but can anyone explain what is going on here? Is there any way to make SQL 2005 work like SQL 2000 in this specific case?
Hi!We have SQL Server 2000 on virtual machine (slowly). We run a pair ofstored procedures from ADO.NET 2.0 (WebService):- first procedure: insert into table A a data row with correlationID- next procedure: search for inserted record in tab A bycorrelationID and insert value into tab B with constraint to tab A.The problem is when we try to do this in short time (few minutes)thousands times (5000/6000 pairs in 5/6 minutes) in 100 parallelthreads:query in second procedure cannot find inserted record in firstprocedure.Everything is dirty read. no transactions.Have you any ideas?edi
I feel like ssis encryption model has a serious flaw. Especially when linked to SQL Agent jobs.
I have posted and others have posted messages about this. Something is plain wrong with ssis encryption keys and password protection. Also, you do not have the choice not to protect the packages. In my case, protecting packages is completely useless.
I created config files for al my packages connections passswords.
Now, by our IT Policy, I had to change again my password and of course, all packages now return multiple errors when I open them.
Hopefully, the config file did its job and the packages are ran anyways by SQL Agent, however, having to manually retype and resave all packages not to have the errors is just a plain hassle. Not to speak about people not using the config files and the correct "Run As" sql agent account.
I stress the fact that in a real world production environment all packages are driven by SQL Agent jobs and MUST run automatically.
Here is the error I get after opening a package after changing my password:
Error 1 Error loading Constants05.dtsx: Failed to decrypt protected XML node "DTS:Password" with error 0x8009000B "Key not valid for use in specified state.". You may not be authorized to access this information. This error occurs when there is a cryptographic error. Verify that the correct key is available. c:projectsssis packagesssis constantsConstants05.dtsx 1 1
So Why is'nt this key automatically adjusted after Windows NT Domain password Change?
How can I refresh the key, not to have to reype all the packages connections passwords and rebuilding, Checkin-in again all the stuff?
I do not think the solution is "Use an application account which password never changes when you create your ssis packages" however at this time, this is the only solution I can think of.
How do you guys deal with this problem?
I still do not understand the ssis security model I feel it is diconnected from the reality and unpracticable in a production environment like mine.
I think I know the answer to this but thought I'd ask anyway.
I have a conditional split to check a column for null values or empty string values. It looks like this:
(!ISNULL(Ballot)) || (LEN(TRIM(Ballot)) > 0) My question is: Are both sides of the expression evaluated? My testing says yes, because a Null value causes an error. Is there a way to short circuit the evaluation like the || operator in C# or the (less than elegant, and seemingly threatening) OrElse operator in VB? Whats the best alternative:
A slightly more complex expression that turns a null value into an empty string
A script component
Two conditional splits
Two paths out of one condtional split
I went with the first option, here is the expression I came up with:
The problem I met is that SQL server seems not woring with too short keywords (with FREETEXT syntax). The following is the detailed description.
I index two columns, JobDes(ntext) and JobTitle(nvarchar), which mainly store Traditional Chinese(Big5). If I search with select * from JobWanted where FREETEXT(JobTitle,'asp'), it goes well. But if I replace the 'asp' with 'as', nothing returns. What is the reason? Can I eliminate this limination?
Currently, I have set up a SQL 7.0 database backup job and have one person notified if the job succeeds, but HOW do I choose an option to notify a different person if the job fails ?
hello,My friend and I are working on a project involving oracle 9i and an oldUnify database. We got the oracle to see the unify (oracle running onwindows) by using the heterogenous services on oracle.However, that was a test environment. The production environment hasoracle running on linux. The problem is that the odbc drivers for theold unify database are only available on windows!So, I was wondering about something.In the production environment, they also have a MS-SQL 2000 database.Does that database have a function similar to oracle's heterogenousservices that will let us connect to the old unify database and passthat data to oracle?We're building custom apps. on oracle 10g application server. Runningoracle on windows in the production environment is not going to happen.Thanks,Dave
I have a report that need to be delivered to two different persons.
It has 10 columns. But one of them is not supposed to be seen by one person. Can I do this without creating another report and delivering two different reports although they are almost the same?
Some categories have sub-categories. Therefore, some entries will have null for parentID while others will point to another category.
I need to select all of the categories, but have them returned in the correct parent/child order (parents BEFORE children). I can't seem to get this to work correctly. Is this not possible or am I just being stupid?? :)
