Or at least post the execution plan.

Assuming you give 2008 a different instance name, no you don't need to uninstall 2005. They run side-by-side just fine. Not that I'd run them side-by-side on a production server.

Do you have traceflag 1222 and/or 1204 enabled? That will output the deadlock graph to the error log when the deadlocks occur. It will specify exactly where the deadlock is...

I think the one thing you're missing is, by and large, you don't need to worry about lock management. SQL Server manages locks for you. You do need to understand...

It really depends on what you mean by "auto"

There is the startup proc, and there can be only one. You can also set up triggers to respond to ddl or...

Nuts. I was going to put Collosus up. The Forbin Project.

It's also a series of books; Collosus, Fall of Collosus, Collosus and the Crab.

I suspect they're VERY dated now.

Oh, hey,...

Now see, we're still stuck because I thought that translated out as "Time for a beer."

Yeah, I'm not so sure about that one. I was helping someone from Spain and the Spanish column names made things more than a bit confusing.

Do you have access to a trace or, assuming it's not production, can you run Profiler against it?

Without more info, I'm flummoxed. Sorry.

You can simply restore the database backup from the 2000 machine to the 2005 machine. You can also detach the database files from the 2000 server and attach them (after...

Where are you planning on running it? On production? NO!!!!!! curse words removed.

On a sandbox or development server, sure.

If you make him owner of the database, he can run the...

Hmmm. I'm still with my original thought. You had a high severity error.

Any chance other people are inserting to that global temp table at the same time?

Snorkel.

But seriously, Profiler might not give the answer. What if there's a problem with memory or tempdb or disk space and there are waits occurring because page latches or...

actually, that query wouldn't work. It puts 32767 times the same drop statement for each database until it runs out of databases then it replicates null and since it's only...

Check the error log. You might be generating an error of sufficiently high a severity that it's just dropping the connection.