The Wiert Corner – irregular stream of stuff

Jeroen W. Pluimers on .NET, C#, Delphi, databases, and personal interests

  • My badges

  • Twitter Updates

  • My Flickr Stream

  • Pages

  • All categories

  • Enter your email address to subscribe to this blog and receive notifications of new posts by email.

    Join 1,860 other subscribers

Archive for the ‘SQL Server’ Category

SQL Server 2000 Reporting Services: copying the .RDL from your server (via: SQL Server Forums – Copy .rdl from ReportServer)

Posted by jpluimers on 2012/06/12

With old installations like Reporting Services 2000, you will often see that a few the reports on the server don’t match the ones in your version control system.

That’s why you sometimes want to download the .RDL files.

User nsaini shows on the SQLTeam forum:

It is easy …..

  1. Go to Report on Report Manager,
  2. Click properties of report,
  3. under the report definition you will see Edit and Update link…
  4. if you click on Edit you can open or Save the RDL files.

Tara Kizer shows how you get into the report manager: through this URL http://yourserver/reports

Easy :)

There are similar steps for SQL Server 2005 and SQL Server 2008.

If you have to do many of them, you can use Reporting Services Scripter to do it.

Even easier :)

–jeroen

via: SQL Server Forums – Copy .rdl from ReportServer.

Posted in Database Development, Development, Reporting Services, SQL Server, SQL Server 2000 | Leave a Comment »

The ADO.NET Entity Framework and SQL Server 2000: the ProviderManifestToken attribute and selecting it with XPath

Posted by jpluimers on 2012/05/23

Yes. Dorothy. There are people using the ADO .NET Entity Framework with SQL Server 2000  in parallel of moving towards a more modern Microsoft SQL Server version.

Entity Framework is lovely for developing data-centric applications.

By default, Visual Studio 2010 will target SQL Server 2008 as a database. That is fine, but it is kind of invisible it does: there is no property or dialog where you can change this.

What you have to change in order to have the Entity Framework send SQL Server 2000 compatible queries is to:

  1. Right click your .edmx file
  2. Choose “Open with”
  3. Choose the “XML (text) editor”
  4. Find the  ProviderManifestToken attribute
  5. Change the value (usually from “2008”) into “2000”
  6. Save the .edmx file
  7. Build and run your application

A few caveats:

Read the rest of this entry »

Posted in .NET, .NET ORM, C#, C# 2.0, C# 3.0, C# 4.0, Development, EF Entity Framework, Software Development, SQL Server, SQL Server 2000, SQL Server 2005, SQL Server 2008, SQL Server 2008 R2, SQL Server 2012 | Leave a Comment »

Checking which applications have a TCP connection to SQL Server (DTAP)

Posted by jpluimers on 2012/05/15

When in a DTAP environment, you cannot always have complete clean boundaries. Issues in production don’t reproduce in acceptance, you cannot develop in production, etc.

So sometimes you have to simulate or connect to Test or Acceptance Database Servers from a Develop workstation.

There it can get hairy to keep track of which applications connect to which database server.

That’s where the below batch file comes in handy: it scans your systems on connections to common TCP ports used by SQL server, then for each connection give you some process details (or – if you add a commandline parameter – all details that TLINK can get).

The batch file uses the built in tools tasklist, netstat, find and sc (the latter to show information on the local running SQL Services).

It also uses TLIST, which can be a bit awkward to get.

Read the rest of this entry »

Posted in .NET, Batch-Files, CSV, Database Development, Development, Scripting, Software Development, SQL, SQL Server, SQL Server 2000, SQL Server 2005, SQL Server 2008, SQL Server 2008 R2, SQL Server 2012, SQL Server 7 | Leave a Comment »

SQL Server 2012 is Generally Available! – SQL Server Team Blog – Site Home – TechNet Blogs

Posted by jpluimers on 2012/04/25

I’ve been really busy lately with a couple of migration projects, so I totally missed the various SQL Server 2012 is Generally Available! announcements.

Which is important because for one client, this opens the possibility of move from SQL Server 2000 right to SQL Server 2012  in stead of SQL Server 2008 R2 and make a more smooth path for Azure.

–jeroen

via:

Posted in Database Development, Development, SQL Server, SQL Server 2000, SQL Server 2008 R2, SQL Server 2012 | Leave a Comment »

SQL Server 2000 (and probably later) other reason for System.Data.SqlClient.SqlException: A severe error occurred on the current command. (via SQL Server Forums)

Posted by jpluimers on 2012/04/24

While transitioning from SQL Server 2000 to 2008, I recently had the “A severe error occurred on the current command. The results, if any, should be discarded.”  occurring on SQL Server 2000 in the form as shown at the bottom of this message.

Many of the search results point you into the area of atabase corruption, or in using NVARCAR parameters with SQL Server 2000 or SQL Server 2005 (the app didn’t use NVARCAR, nor did it use large VARCHAR parameters).

The cool thing on the SQL Server Forums – System.Data.SqlClient.SqlException: A severe error occurred on the current command post was that it summed up causes, and asked for more:

Posted – 06/17/2004 :  15:05:20

Rashid writes “Hi: Gurus I am getting these errors when I try to execute my application. According to MS knowledge base (http://support.microsoft.com/default.aspx?scid=kb;en-us;827366) these errors happen due to following resons

  1. You use a SqlClient class in a Finalize method or in a C# destructor.
  2. You do not specify an explicit SQLDbType enumeration when you create a SqlParameter object. When you do not specify an explicit SQLDbType, the Microsoft .NET Framework Data Provider for SQL Server (SqlClient) tries to select the correct SQLDbType based on the data that is passed. SqlClient is not successful.
  3. The size of the parameter that you explicitly specify in the .NET Framework code is more than the maximum size that you can use for the data type in Microsoft SQL Server.

None of these are true in my case. Are there any other reasons that can cause these problems..

There is one more: sending huge SQL Statements to your SQL Server is always a bad idea and gives this error too. Read the rest of this entry »

Posted in .NET, C#, C# 2.0, C# 3.0, C# 4.0, Database Development, Development, Encoding, Software Development, SQL Server, SQL Server 2000, SQL Server 2008 R2, Unicode | Leave a Comment »

Migrating SQL Server 2000 ISQLW Utility to SQL Server 2008 R2 SSMS command-line parameters (SQL Query Analyzer; SQL Server Management Studio)

Posted by jpluimers on 2012/04/17

If you followed my blog, you probably already guessed that I’m assisting a client to prepare for a big SQL Server 2000 to SQL Server 2008 R2 migration. When not, you know now :)

I’m a fan of commandline, tools, and preconfigured settings. Which means that I’m in the midst of reconfiguring all my shortcuts to SQL Server 2000 tools to SQL Server 2008 equivalents.

When inspecting and changing LNK shortcut files, two tools are important:

  • dumplnk: dumps a shortcut lnk file from the commandline
  • shurtcut: creates a shortcut lnk file from the commandline

One of tools I many shortcuts for is the ISQLW aka SQL Query Analyzer, which I use far more than the SQL Server Enterprise Manager (more on SSEM in a future post).

Depending on the mode of authentication you use to connect to your SQL Server, there basically are two forms of shortcuts:

  • SQL Server Authentication (using SQL Server username and password):
    "C:\Program Files\Microsoft SQL Server\80\Tools\Binn\isqlw.exe" /S"servername[\instancename]" /U"username" /P"password"
  • Windows Authentication (using the credentials of the currently logged in windows user):
    "C:\Program Files\Microsoft SQL Server\80\Tools\Binn\isqlw.exe" /S"servername[\instancename]" /E

The former is less secure (so better to only store those shortcuts in a place that no other users can access).

From the ISQLW.exe syntax documentation:

Syntax

isqlw
[-?] |
[
[-S server_name[\instance_name]]
[-d database]
[-E] [-U user] [-P password]
[{-i input_file} {-o output_file} [-F {U|A|O}]]
[-f file_list]
[-C configuration_file]
[-D scripts_directory]
[-T template_directory]
]

The parameters /S, /U, /P and /E are very similar to the -S, -U, -P and -E ones from SSMS.exe (SQL Server Management Studio) with one distinction: there MUST be a space between each parameter and the value:

  • SQL Server Management Studio (using SQL Server username and password):
    "C:\Program Files\Microsoft SQL Server\100\Tools\Binn\VSShell\Common7\IDE\Ssms.exe" -S "servername[instancename]" -U "username" -P "password"
  • SQL ServerManagement Studio (using credentials for the currently logged in windows user):
    "C:\Program Files\Microsoft SQL Server\100\Tools\Binn\VSShell\Common7\IDE\Ssms.exe" -S "servername[instancename]" -E

Read the rest of this entry »

Posted in Batch-Files, Database Development, Development, Power User, Scripting, Software Development, SQL Server, SQL Server 2000, SQL Server 2005, SQL Server 2008, SQL Server 2008 R2, SQL Server 2012 | 5 Comments »

SQL Server 2000: How do I register multiple databases with Microsoft SQL Server Enterprise Manager?

Posted by jpluimers on 2012/02/24

I’m in the midst of a big migration traject.

One class of systems involved in the migration is SQL Server 2000 databases (yes, I know that SQL Server 2000 has been in Extended Support mode since April 8, 2008, that’s one of the grounds for migration).

Sometimes, you need SQL Server Enterprise Manager (a plugin to the Microsoft Management Console) to work on those databases to see what needs to be migrated (yes, this should be in the documentation, of which the ultimate form is the system itself ;-).

SQL Server Enterprise Manager has a nice feature that you can group SQL Server registration into SQL Server Groups.

However, it also has a limitation: you cannot add the same SQL Server to multiple SQL Server Groups. That is: unless you create an alias on the SQL Server, as aliases are not counted in the “unique” criterion.

Below are the top 2 links I found using “multiple databases” “sql server”  “enterprise manager”. Since I’m not sure the content of those will last, below are the full quotes.

Besides adding a SQL Server Alias (which has other uses as well), the most important steps are “6” in the first quote, and “1.6” in the second quote: your alias needs to point to the same IP address (or hostname) as your regular server.

How do I register multiple databases with MS SQL Enterprise Manager?

MS Enterprise Manager does not allow you to register the same database server multiple times. It is sometimes useful to create multiple registrations, if you want to connect to different databases on the same server. To workaround this, follow the steps below.

  1. Go to Start – Programs – Microsoft SQL Server – Select “SQL Server Client Network Utility”
  2. Go to Alias Tab
  3. Click “Add”
  4. In the “Server alias:” text box, enter a name you want to use to distinguish this database server
  5. In the “Network libraries:” section, select “TCP/IP”
  6. In the “Connection parameters” section, change the “Server name:” to your database server, i.e.; mssql01.discountasp.net
  7. Click OK

Once the Alias is created, you can register it using the same method described in this Knowledge Base article.

How can I open multiple databases in SQL Server Enterprise Manager?

  1. Create SQL Server Alias
    (Please repeat the following steps if you want to open mutiple databases)

    1. Open SQL Server Client Network Utility.
    2. When Client Network Utility opens, click on the Alias tab.
    3. Click on the Add button.
    4. In the Server Alias text box enter a name for your SQL connection.
    5. Select TCP/IP under Network libraries
    6. Enter the IP address of the Server your SQL database is located on, in the Server name text box.
    7. Click on the OK button.
    8. Your Server alias will now be configured, click on the Apply button.
    9. Click on the OK button.
  2. Create SQL Server Connection in Enterprise Manager
    (Please repeat following  steps if you want to open mutiple databases)

    1. Open SQL Server Enterprise Manager.
    2. When Enterprise Manager opens, right click SQL Server Group and select New SQL Server Registration.
    3. The Register SQL Server wizard will now open.
    4. Click on the Next button.
    5. From the Available Servers drop down menu, select the server alias you created earlier.
    6. Click on the Add button.
    7. The server should now be listed under Added servers, click on the Next button.
    8. When the next dialog box appears select The SQL Server login information that was assigned to me by the system administrator see above.
    9. Click on the Next button.
    10. Enter your Login name and Password ? this will be used every time you connect to your SQL server.
    11. Click on the Next button.
    12. Ensure that Add the SQL Servers to an existing SQL Server group is selected and SQL Server Group is selected for Group name.
    13. Click on the Next button.
    14. When the next dialog box opens you can click on the Finish button.
    15. Your SQL Connection will now be created.

–jeroen

via:

Posted in Database Development, Development, SQL, SQL Server, SQL Server 2000 | 1 Comment »

Migrate/Transfer SQL Server 2008/2005/2000/7 Logins to SQL Server 2008

Posted by jpluimers on 2012/01/12

When moving databases across servers, you face the same problem as when upgrading servers: users are server specific, but permissions are databases specific. They are not bound by UserName, but through a SID (security identifier).

When adding the same UserName entries to a new server that already has the databases restored, you get error messages like these:

Error 21002 [SQL-DMO] User ‘account’ already exists

When adding the same UserName  entries, then restoring the databases, it won’t work, because the matching SIDs don’t exist.

There are many sites explaining the “Error 21002” and pointing to sp_change_users_login.

But sp_change_users_login (and the ALTER USER UserName WITH LOGINI = UserName2 in SQL Server 2005 and higher) is not the actual fix to the problem: it will re-add the user with a new SID, then correct the SID in the database currently in use.

Actually you are after “Orphaned” users: users that are defined in the various databases, but not present in the user list on the SQL Server instance. Running sp_change_users_login with the “Report’ parameter on the “new” server will show a list of orphaned users.

An even better way to show Orphaned Users is by running the ShowOrphanUsers script (thanks Vyas, wish more people republish your stuff with attribution!). In my own version (sourcecode is below), I have added an extra UserSID column of type varbinary(85).

In order to transfer users to a new server, you need a sp_help_revlogin stored procedure. Depending on the version of your SQL Server (7/2000/2005/2008), you need a slightly different version of a script that creates sp_help_revlogin for you. All versions are available at NetNerds.net.

–jeroen

via: Migrate/Transfer SQL Server 2008/2005/2000/7 Logins to SQL Server 2008.

USE master;
GO
IF OBJECT_ID ('dbo.ShowOrphanUsers') IS NOT NULL
  DROP PROCEDURE dbo.ShowOrphanUsers
GO
CREATE PROC dbo.ShowOrphanUsers
AS
BEGIN
	CREATE TABLE #Results
	(
		[Database Name] sysname COLLATE Latin1_General_CI_AS,
		[Orphaned User] sysname COLLATE Latin1_General_CI_AS,
                [UserSID]  varbinary(85)
	)

	SET NOCOUNT ON

	DECLARE @DBName sysname, @Qry nvarchar(4000)

	SET @Qry = ''
	SET @DBName = ''

	WHILE @DBName IS NOT NULL
	BEGIN
		SET @DBName =
				(
					SELECT MIN(name)
					FROM master..sysdatabases
					WHERE 	name NOT IN
						(
						 'master', 'model', 'tempdb', 'msdb',
						 'distribution', 'pubs', 'northwind'
						)
						AND DATABASEPROPERTY(name, 'IsOffline') = 0
						AND DATABASEPROPERTY(name, 'IsSuspect') = 0
						AND name > @DBName
				)

		IF @DBName IS NULL BREAK

		SET @Qry = '	SELECT ''' + @DBName + ''' AS [Database Name],
				CAST(name AS sysname) COLLATE Latin1_General_CI_AS  AS [Orphaned User],
                                SID AS [UserSID]
				FROM ' + QUOTENAME(@DBName) + '..sysusers su
				WHERE su.islogin = 1
				AND su.name <> ''guest''
				AND NOT EXISTS
				(
					SELECT 1
					FROM master..sysxlogins sl
					WHERE su.sid = sl.sid
				)'

		INSERT INTO #Results EXEC (@Qry)
	END

	SELECT *
	FROM #Results
	ORDER BY [Database Name], [Orphaned User]
END
GO

Posted in Database Development, Development, SQL, SQL Server, SQL Server 2000, SQL Server 2005, SQL Server 2008, SQL Server 2008 R2, SQL Server 2012, SQL Server 7 | 1 Comment »

SQL Server inverse of (equals) is ((not equals) or (is NULL)

Posted by jpluimers on 2011/12/07

Usually getting queries right with SQL Server comes down to handling NULL correctly.

For instance, on this table:

MYTABLE
ID LAST_NAME
6 COUNT(*)
1 FOO
2 BAR
3 FOO
4 **NULL**
5 BAR
6 FOO

What are the results of these SELECT statements

SELECT COUNT (*)
FROM MYTABLE

SELECT COUNT (*)
FROM MYTABLE
WHERE     (LAST_NAME = 'FOO')

SELECT COUNT (*)
FROM MYTABLE
WHERE     (LAST_NAME <> 'FOO')

SELECT COUNT (*)
FROM MYTABLE
WHERE     NOT (LAST_NAME = 'FOO')

You might think they are like these, as LASTNAME <> ‘FOO’ looks like the inverse of LASTNAME = ‘FOO’:

  • 6
  • 3
  • 3
  • 3

But in fact the results are these:

  • 6
  • 3
  • 2
  • 2

When you look closely to the SQL statements below, you will see that the inverse of EQUALS contains an IS NULL, the same for the inverse of NOT EQUALS:

SELECT COUNT (*)
FROM MYTABLE

-- inverse of NOT EQUALS
SELECT COUNT (*)
FROM MYTABLE
WHERE     (LAST_NAME = 'FOO') OR (LAST_NAME IS NULL)

SELECT COUNT (*)
FROM MYTABLE
WHERE     (LAST_NAME <> 'FOO')

SELECT COUNT (*)
FROM MYTABLE
WHERE     (LAST_NAME = 'FOO')

-- inverse of EQUALS
SELECT COUNT (*)
FROM MYTABLE
WHERE     (LAST_NAME <> 'FOO') OR (LAST_NAME IS NULL)

-- inverse of EQUALS
SELECT COUNT (*)
FROM MYTABLE
WHERE     (NOT (LAST_NAME = 'FOO')) OR (LAST_NAME IS NULL)
  • 6
  • 4
  • 2
  • 3
  • 3
  • 3

Lesson learned: always take into account NULL when trying to formulate your SQL comparisons well.

–jeroen

Posted in Database Development, Development, SQL Server, SQL Server 2000, SQL Server 2008 | Leave a Comment »

SQL Server 2000: export SQL Query Analyzer result data to Excel/CSV and more

Posted by jpluimers on 2011/11/16

I just visited a client which is still using SQL Server 2000, and not upgraded their tool set, nor allows different tools to be installed.

Which means back to basics, re-adjusting the project planning and frantically trying to remember things from the past.

Boy am I spoiled with a current toolset :)

This goes from simple things like saving a result set from the SQL Query Analyzer:

  1. Select a cell
  2. Press Ctrl-A to select all rows
  3. Right click in the grid and select “Copy” (to copy the cells as CSV) or “Save As” (to export the cells as CSV)

along the absence of support for XML, MARS, error handling (not even talking about spatial data!) to the way that in SQL Server 2000 DTS (Data Transformation Services) has its own mind of date/time format handing while importing stuff.

But it sure helps setting aims for the scheduled migration process to the far more current SQL Server 2008 R2 :)
(Mental note: migrating the SQL Server 2000 DTS packages to SQL Server 2008 R2 will be a challange).

Oh, some of the SQL Queries that come in handy when moving stuff around in an OTAP/DTAP environment:

1. Selecting relevant DTS packages (that are always in the MSDB database)

select
name
from msdb..sysdtspackages -- 2005: sysdtspackages90; 2008: sysssispackages
where
name like '%my-app-ID%'

2. Selecting relevant objects from a database (watch the xtype values that can exist in SQL Server 2000)

using my-database
select 
name, xtype
from 
sysobjects
where 
name like '%my-app-id%'
and xtype in ('U', 'V', 'TR', 'P', 'X') -- tables/views/triggers/procedures/xprocedures only
order by xtype, name

Blast from the past :)

–jeroen

via: export sql query analyzer data to excel SQL Server.

PS: For the statistics, SQL Server 2000 has been EOL for a while; mainstream support ended in 2008, extended support ends in 2013.

Posted in CSV, Database Development, Development, Software Development, SQL, SQL Server, SQL Server 2000 | Leave a Comment »