Seems your mixing some concepts?

Snapshots are an easy escape if you want to revert a database update. So if you upgrade your application, make a snapshot of your database first....

I agree, the only (holy) answer to this question is "it depends"

for example, if you have a mission-critical database, I recommend a dedicated databaseserver just for that database. With this...

A full backup is just a copy of every used page in your database. So if you have some sort of fragmentation in your database, it's there (again) after a...

Make sure you do a DBCC UPDATEUSAGE first, in order to remove inaccuracies in the statistics.

After that, run sp_updatestats tot update your statistics.

Note: There are some discussions on the...

Are you sure the path for the attached files is also available on the 2005 server? It seems it's using the original filelocation when attaching the file, which is a...

"This instance of SQL Server has been using a process ID of 4368 since 04/09/2008 16:56:33 (local) 04/09/2008 15:56:33 (UTC). This is an informational message only; no user action is...

Do you know why you have/want to change the recovery model? It seems you're stuck in a very strange situation.

The recovery model has to met with the company's recovery strategy....

Hi,

You can change the recovery model "on the fly", users do not have to log off during the change. No restart of MSSQL is required.

Make sure you implement a transaction...

I understand the concepts of the retention time. I'm interested in how SQL decides it's time to do a cleanup. It this check done every friday at 2:00, or sunday,...

Not that I know. I assume your real statement is much more complex, if not, add "where id=10" 🙂

My wish is to use CTE as a cursor replacement, but this...

Logon trigger?

Generally, you'll get the most benefit of a (clustered) index if you add columns which are most used in your where clause.

In your case EMPLID. But Clustered indexes have to...

If you want the text which belongs to an error number (if that's the case), try:

select *

from master.sys.messages

where message_id = ####

(replace #### with the error number)

If you want to simulate...

Are you sure? We are running Axapta 3 on a 2K5 Cluster. If failover happens, all connections are broken. So AFAIK your application also needs to be cluster aware.

You're right, my specification was not accurate enough.