Are you altering the view using query analyzer, enterprise manager,...?

What error do you get?

There isn't a performance hit (or a small one) to have multiple databases. This changes when they are on separate servers. In that case sql server might decide to pull the...

SELECT @fullname = (ISNULL(LastName,'') + ', ' + ISNULL(FirstName,''))

NULL+something=NULL

I've enabled the following traceflags so the deadlock situation appears in the logs

dbcc traceon (1204, 3605, -1)

go

dbcc tracestatus(-1)

go

http://support.microsoft.com/kb/832524/

an extra vote for Andrew's solution since it allows index usage

is connection pooling on/off?

We had troubles with connection pooling off resulting in error messages like server does not exists... because the machine couldn't generate connections fast enough (or something alike)

functions on columns that participate in the join can make the optimizer decide not to use an index (not sargeable)

like convert(varchar(10),mydate1,102)=convert(varchar(10),mydate1,102)

adding Trusted_Connection=true  means it needs to be windows authenticated.

Second, connecting as sa is a very serious security risc. I hope sa has a password.

found a kb article that might apply: (for standard edition)

http://support.microsoft.com/kb/316749/

Otherwise, are you profiling any performance counters?

found a kb article that might apply: (for standard edition)

http://support.microsoft.com/kb/316749/

Otherwise, are you profiling any performance counters?

select ...

from table1

where not exists (select PERMIT_ID from table1 table_bad where table1.PERMIT_ID=table_bad.PERMIT_ID and table_bad.PERMIT_EVENT_DT is null)

If could be rewritten to replace not exists, by some form of counting those whose...

Thanks for posting the solution.

Is it possible to post a link describing the issue with the cpu?

In such situations it is always helpfull to compare the execution plans of the queries, you can do this via the query analyzer: (Query) show estimated execution plan.

Possible reasons: the...

Looks your tempdb is running full or it can't grow fast enough.

Can you split your command in batches of 100 000? With a counter or something to remember the highest...

Raid 5 isn't particulary recommended if there is a lot of writing activity. Looks splitting it up would be ideal. Can't say more since the largest raid I have are...