You can also delete the history with a maintenance plan.

There is a pre-defined task just for that, the History Cleanup Task.

50 MB may be a little small. I usually use at least 100 MB, but for a database that I expect to grow fairly often, 500 MB or more...

Piotr.Rodak (11/10/2009)

...There is no point in filtering out rows that did not change from the update statement in my opinion, as updating all rows will not change rows that are...

You can use the following script to shrink the data file.

Shrink DB File by Increment to Target Free Space

http://www.sqlteam.com/forums/topic.asp?TOPIC_ID=80355

After you shrink the datafile, you should:

1. Defragment the indexes and update...

Glad to help. As least you didn't post back to complain that my solution was too much work....

Create multiple monthly schedules for the job: one that runs on day 5, one that runs on day 12, one that runs on day 19, and one that runs on...

There may be jobs where knowing the answers to those questions would be important. I have encountered consulting companies that seem more interested in the “star quality” of their...

There is no reason to run a shrink, and that is likely the cause of the large diff backups.

A better daily maintenance would be: 1. Update statistics, 2. Integrity check,...

Put simply, testing is part of development, so if you don’t unit test, the job is not done.

The ability to properly test is probably the most important skill for a...

The typical de-normalization process is this:

1. Create database design without knowing or attempting to follow normalization rules.

2. Claim that you de-normalized the design for performance.

This script gets the file information for every database on a server, and inserts it into temp table that is queried multiple ways to give various levels of analysis of...

select

database_id,

CAST(name as varchar(66)) logical_name,

sum(size*8/1024) mb_size

from

master.sys.master_files

where

physical_name NOT LIKE '%.ldf'

group by

database_id,

name

with rollup

having

name is not nullor

database_id is null

order by

case when database_id is null then 1 else 0 end,

database_id,

name

Read this article to understand the subject:

Detecting and Ending Deadlocks

http://msdn.microsoft.com/en-us/library/ms178104.aspx

However, in many cases, changing the database to read_committed_snapshot will solve your deadlocking problems:

use master

alter database [MyDatabase] set allow_snapshot_isolation ...

You need to use decimal math, instead of integer math.

select

Wrong= (76075/119027)*100 ,

Correct= (76075./119027.)*100.

Results:

Wrong Correct ...

This works OK for me.

select

a.*

from

openrowset('SQLOLEDB','SERVER=(local);Trusted_Connection=yes;',

'

set fmtonly off;

exec master.dbo.sp_who

') a