Viewing 15 posts - 1,411 through 1,425 (of 1,554 total)
It depends...
As you've noted, both temptables and table variables have their uses. Where the limit goes, I'd guess would depend on things as...
June 22, 2004 at 3:56 am
So, if I understand it correctly, what you want is one-statement incrementing counter-thingy...?
declare @nextid int
update table1
set @nextid = intcount = intcount + 1
where loc = 'A'
@nextid now contains the incremented...
June 22, 2004 at 3:51 am
How is SQL Servers memory configured?
Does the doc-loading app compete with SS over memory..
If memory is set on auto, try to cap the server at some level (ie 1Gig), allowing...
June 22, 2004 at 3:05 am
Have you verified that the 7 server can actually see the w2k3 server over the network at all?
ie ping, map drives, odbcping, manual QA login 7.0 -> 2k3 etc..
/Kenneth
June 22, 2004 at 3:00 am
AFAIK there's no end date determined yet.
You may look at microsoft.com for more detailed info
/Kenneth
June 22, 2004 at 2:57 am
Unless the db is audited, you can't find out.
SQL Server itself doesn't keep track of this.
/Kenneth
June 22, 2004 at 2:56 am
The thing is, if the error generated from the INSERT statement is a batchbreaking error, the proc exits immediately, and you will never come to the 'Select @errorcode = @@Error'...
June 22, 2004 at 2:53 am
You can't.
Some errors aren't trappable from within a proc - they must be handled at the calling (client) side.
/Kenneth
June 22, 2004 at 1:52 am
What you want is a PIVOT.
Unless the #items possible to buy is known in advance, this is best to do on the client side.
It gets very ugly in Transact-SQL, although...
June 22, 2004 at 12:55 am
No real ideas, sorry, but...
using the sp_OAxxx procs is probably "shaky" at best, and may also add significant overhead, especially if the data volume is great.
Is it really necessary to...
June 22, 2004 at 12:50 am
I don't quite get what you want.
Could you expand on your example and desired result, and provide a few rows with "accurate" data?
(I don't see how the example result correlates...
June 22, 2004 at 12:45 am
Seems that what you want is to get rid of the duplicate(s) jobid + status before the aggregate..?
What about first selecting the distinct JobID + Status, then aggregating that result?
Select...
June 22, 2004 at 12:39 am
I'm beginning to get the feel that the problem is basically that some values are calculated 'too many times'...
On way to get around such anomalies is to start from the...
June 10, 2004 at 8:15 am
To add to the diversity, here's another way.
create table #x ( number int not null )
insert #x values (1)
insert #x values (12)
insert #x...
June 10, 2004 at 7:58 am
There is, but those are undocumented and highly unsupported for 'user' usage.
/Kenneth
June 10, 2004 at 2:47 am
Viewing 15 posts - 1,411 through 1,425 (of 1,554 total)