Viewing 15 posts - 5,356 through 5,370 (of 6,678 total)
Lynn Pettis (4/17/2009)
I think he-who-should-just-go-away may be back. I wonder how he'll respond to my comment?
Hard to tell - then again, who can even understand him when he's actually...
April 17, 2009 at 2:42 pm
Don't change the login properties using the services applet. Use SQL Server Configuration Manager which will make sure the changes are made to all of the appropriate locations.
April 17, 2009 at 2:29 pm
See this technique posted by Steve Kass
CREATE TABLE dupNulls (
pk int identity(1,1) primary key,
X int NULL,
nullbuster as (case when X is null then pk else 0 end),
CONSTRAINT dupNulls_uqX UNIQUE...
April 17, 2009 at 2:15 pm
A slightly easier, and generally accepted method is to perform a delete/insert instead of an upsert. For example:
DELETE FROM Destination
WHERE key IN (SELECT key FROM Source);
Or, using EXISTS:
DELETE...
April 17, 2009 at 1:41 pm
How are you starting the service? Since this is a cluster you should not be starting the service from the services applet. You need to bring the service...
April 17, 2009 at 1:35 pm
I am not seeing any problems with this - what are the results you are getting?
April 17, 2009 at 1:30 pm
Well - it really depends on what your table structures are, how they are related and what is the purpose of your queries.
If you want the results in a single...
April 17, 2009 at 11:48 am
Is this really the setting you have for max memory?
Max Server Memory = 14145000 meg
14 million, 145 thousand megabytes? If so, for a system with 16GB of memory available...
April 17, 2009 at 11:28 am
The only way to avoid dynamic SQL would be to create a partitioned view. You could create the view dynamically, and then reference the view in your procedures.
So, you...
April 17, 2009 at 10:11 am
One of the things you will have to consider is the additional LUN's that will need to be presented to the cluster. Remember, each instance is going to have...
April 17, 2009 at 10:05 am
No idea what you are really asking here - but, you can only have one clustered index per table.
April 16, 2009 at 5:03 pm
No - views are not materialized. The view definition is expanded into the calling query and used to generate the final query to be executed.
The view definition itself will...
April 16, 2009 at 5:01 pm
No - each job step is it's own transaction. I don't think you can use transactions because each step could also be using a new connection.
April 16, 2009 at 4:15 pm
The EXISTS function returns true if there is at least one row returned from the query. Using EXISTS will be faster than a count.
April 16, 2009 at 3:20 pm
You will see that message while the update is patching the passive node. Again, if there are any users logged into the passive node - the upgrade will hang.
There...
April 16, 2009 at 1:59 pm
Viewing 15 posts - 5,356 through 5,370 (of 6,678 total)