I get really frustrated most times I use connect. (connect.microsoft.com). The latest reason is one I've hit many times.

When I do a connect visit I generally look at new feedback. Almost always I come across one like this

https://connect.microsoft.com/SQLServer/feedback/ViewFeedback.aspx?FeedbackID=368538

This person has an immediate problem, and I can't do anything to get in touch with him.

In this situation the person should have contacted Microsoft Support and not raised a connect bug.

If you have an operational issue that is causing your application to fault then you MUST raise a Support case.

The feedback loop for connect has no SLAs, guarantees and no easy ship vehicle for fixes.

You might think that you don't want to riase a support case as it will cost you money. Well if your problem is due to a bug then you WILL NOT be charged.

If the bug is non critical then fine raise a connect issue because it is likely that you will need to get a number of votes to convince the product team that the bug needs fixing.


This is cross posted from my SQLBlogcasts blog which can be found here, http://sqlblogcasts.com/blogs/simons/Should I use connect or Mcirosoft Support (PSS)?

Categories:


The create table statement doesn't allow you to create a table with more than 1024 columns.  So how do you do it.

You have to create the table and then add them after the fact.

[21/9/2008 Thanks to Kalen for pointing out that you can if you create the table with a columnset at the same time].

Even if using column sets you have to define all the columns that you want to store. The XML columnset doesn't act like an XML overflow column, it is a computed representation of the sparse columns on the table.

The following highlights that. The first three inserts fail as they are trying to populate columns that con't exist

create table test2 (

id int not null primary key

,sp1 int sparse

,cs  xml column_set for all_sparse_columns)

go

insert into test2 (id,cs)values (1,'<col1>12123</col1>')

insert into test2 (id,cs)values (2,'<col2>aaa</col2>')

insert into test2 (id,cs)values (3,'<col3>1/1/2008</col3>')

go

select * from test2

go

declare @i int = 0

while @i<2000

begin

      declare @sql varchar(100)= 'alter table test2 add col' + cast(@i as varchar(10)) + ' int sparse'

      execute (@sql)

      set @i = @i+1

end

go

 

sp_help test2

 


This is cross posted from my SQLBlogcasts blog which can be found here, http://sqlblogcasts.com/blogs/simons/Getting more than 1024 columns in a table

Categories:


If you don't believe me then come along on Friday before SQLbits and learn about the great things you can do with .net in SQL.

http://www.sqlknowhow.com/events/sqlbits20080912.aspx

During the one day seminar we will cover

  1. How CLR is hosted in SQL Server
  2. Creating CLR objects using a .Net language
  3. Comparing the performance with TSQL and other options
  4. Providing guidance on when to use CLR objects and when not to
  5. Getting the best performance from your CLR objects

I hope you will come away not afraid of using .net in SQL, it means you can write simpler code, that is performs better and in some cases achieves things not possible in TSQL.


This is cross posted from my SQLBlogcasts blog which can be found here, http://sqlblogcasts.com/blogs/simons/Using .Net in SQL is faster than TSQL in many situations

Categories:

Theme design by Nukeation based on Jelle Druyts