I'm sorry, I'm not sure. I just did the same thing you described and it worked fine.

Then you sir, are a gentleman and a scholar. I'm neither. Pass the hammer please.

sp_who2 is a simple way to see which processes are running. It shows the SPID. You can use that to kill a process.

I'm not sure if this is the answer...

Yeah, you'll need to change the app. The methods for accessing system metadata within SQL 2005 are different than they were in 2000. The views that look like the old...

That completely depends on your environment. For example, we have a server environment where some 3rd party apps are running that need 'sa' log-in in order to operate (can I...

Nope, sorry. Common Table Expressions are new in 2005.

I read this yesterday and wasn't sure what you're looking for. Would a common table expression work? This is derived from BOL:

WITH TableX AS

   (SELECT e.ManagerId, e.EmployeeId, 0 AS Level

   ...

If you run a query against the sys.sysobjects system view in the master db, do you still see it?

SELECT * FROM sys.sysobjects

WHERE NAME = 'sp_executesql'

If so, note the type. It's...

What are you trying to do? There is a view available, sys.foreign_keys, which shows the basic information on a given set of keys within a database. Is that what you're looking...

Yeah, especially since M:M is usually done through a third table.

Since you can only restore to a point in time you need some method of determining which transaction is your final point. You might want to check Lumigent LogExplorer or...

You have that 100% correct. As a matter of fact, the BOL says you can't mirror between two servers on the same cluster. Although, if you set up active/active clusters...

Yes, it does provide availability. The way that it was described to us when we were working with the Dell consultants on our last cluster configuration was that for high...

Are the indexes in reference used in the update queries? If so, you need those indexes. You may or may not want to rebuild them before and after (and if...

Agreed. I'd go for the partitioning first. It's complete seperate from the other architecture decisions and work.

After that, clustering gives you availability and mirroring gives you redudancy. These can be...