so this is a hiarchy.

With SQL2005 you can use a CTE with a recursive part

With SQL2000...

- avoid non-clustering-indexes at load time (consider drop before / re-create after load)

- sort you data according to the clustering index of your objects before loading.

- use bulk-insert whenever possible

I...

thanks for the feedback.

These kind the tiny brainteasers being solved make our day

I didn't encounter this problem, but I'm using SQL2005EE and I'm installing the client-software at the C-drive ! (default).

I've set up an Active/Active cluster with 3 instances hosted on 2...

there is offcourse http://msdn2.microsoft.com/en-us/library/ms143799.aspx.

But it does not include any reference to the created local users or groups.

Suppose I create a ghost-image after install, apply the ghost to a new...

IMO the JOIN-syntax using the on clause differantiates better wat's a join-predicate (on-clause) and what's a filter-predicate (where clause).

This makes is more ease for another person to understand your queries.

Check this...

check books online

DBCC CHECKIDENT ('table_name', NORESEED)  just for reporting

or

DBCC CHECKIDENT ('table_name', RESEED) reporting and correction !

What error is checkident giving ?

do you have scheduled jobs for DBCC DBReindex ?

or indexdefrag ?

If not, reorganize the data and test again.

Maybe it's time to do this on a scheduled/planned basis.