Shoot, Lynn... I knew that... you knew that... I just wanted to make sure Kung Lao knew that.   Whether anyone suggests changing the server...

???  He's already deleted the data by mistake... how is either method you suggest going to recover data that's already been deleted?  

I agree...

The other thing about Carl's trigger, is that it will actually update more than 1 row in the DCParty table whereas the original trigger would only update one row in...

Nope... not a PATINDEX bug... and you don't need to mess with the server default settings (case sensitive servers are a huge pain!).

This will do it for you...

SELECT PATINDEX...

Sure... if the @Column_Name variable contains something less than 2 characters, then because of the "-2" modifier in it, BOOM!

Can't help much more than that 'cause I don't know what...

Back button has never worked for me

Good workaround, but a bit slow and resource intensive if you have millions of rows...

Something like the following works very fast... yeah, it does use dynamic SQL but that's only...

Why do you need a white paper on the fact that if two users use a function, they get their own answers?  It's that way with any function in any...

Also, almost forgot... you can shrink TempDB (by file) as with any other DB without bouncing the server (possibly as Sugesh suggested).  It could cause a fairly long wait time, though......

If the batch size of BCP/BULK INSERT is relatively small (less than a million rows, say), then those are likely not the problem.  Rumor has it that the extensive use of...

No... not ready to make that recommendation.  We first need to identify what change was made to cause the code to suddenly go south several days ago.  Quick fix would...

Run the following...

DBCC USEROPTIONS

... which will produce something like this...

Set Option Value

textsize 64512

language us_english

dateformat mdy

datefirst 7

quoted_identifier SET

arithabort SET

ansi_null_dflt_on SET

ansi_defaults SET

ansi_warnings SET

ansi_padding SET

ansi_nulls SET

concat_null_yields_null SET

(12 row(s) affected)

DBCC execution completed. If DBCC printed error messages, contact...

Ok... not the explanation I was looking for but I guess it'll do.

If you want to round to a single decimal place, then convert it to to something like DECIMAL(19,1)...

It's ok... heck, I guess I don't blame you about the bump on a busy forum.  When you need an answer, you need an answer and a relatively harmless bump...

Heh... ok... not sure that's better or worse but at least its not a desparate human   Thanks, Rainer.