Viewing 15 posts - 421 through 435 (of 13,460 total)
files are actually zip files with a sqlplan extension.
On Plan1_server1 you can see the estimated and actual number of rows are way off(estimated = 91.2K, actual = 1.8K,...
Lowell
July 17, 2017 at 6:45 am
variable, string OR [COLUMN NAME] of type varchar/nvarchar is accepted for the parameter!
so if you choose a column, or the converted value of a column that is NOT a...
Lowell
July 17, 2017 at 6:31 am
For the RevenueAmt, if the datatype is MONEY, the convert function has an additional parameter and will insert your commas for you. if it si a decimal, you have to...
Lowell
July 17, 2017 at 6:27 am
I did this once, when we identified that developers were using the application login as their dev login to do work. I got the management buy in that this should...
Lowell
July 14, 2017 at 8:16 am
As John Mitchell noted in the other thread,,all you can do is reasonably infer the person, based on the IP address or hostname(whihc technically can be spoofed in the connection...
Lowell
July 14, 2017 at 8:07 am
the first table should be sysjobs, as some jobs may exist but have NEVER been executed,and thus have no history.
also, the joins should be LEFT JOINs, since jobs could...
Lowell
July 14, 2017 at 6:57 am
Robert, are you talking about each Server in an Always On Availability Group, or a Windows Server Failover Cluster with shared storage and a cluster aware SQL instance?
You...
Lowell
July 13, 2017 at 8:27 am
I've seen issues like this in the GUI scripting; when you use an older version of SQL Server Management Studio(lets say 2008 or 2012) and are connecting to a higher...
Lowell
July 12, 2017 at 6:26 am
you could do it as a string representation like this:
/*--results
ElapsedString Days Hours Minutes Seconds Milliseconds
3:03:30:11 3 3 30 11 446
*/
WITH MySampleData
AS
(
SELECT
[Days] = datediff(day,0,ET-ST),
...
Lowell
July 11, 2017 at 3:06 pm
ok since you are inserting into a table, you do not need the ORDER BY in the end.
You could fiddle with changing it to use EXISTS instead of...
Lowell
July 11, 2017 at 2:57 pm
Lowell
July 11, 2017 at 12:58 pm
i would think about slicing the new SAN into three drives; two big ones for the SQL stuff for db and logs, and a new,smaller one for the disk-based packages. Lowell
--help us help you! If you post a question, make sure you include a CREATE TABLE... statement and INSERT INTO... statement into that table to give the volunteers here representative data. with your description of the problem, we can provide a tested, verifiable solution to your question! asking the question the right way gets you a tested answer the fastest way possible!
July 11, 2017 at 11:50 am
I've done this with a script task in an SSIS package. it is a good size snippet of code, with lots of caveats.... the end result is exactly what everyone...
Lowell
July 11, 2017 at 9:43 am
Yes, you'll end up needing bunch of case statements, one for each column.
This could also be done int he presentation layer, so for example SSRS could use an expression...
Lowell
July 10, 2017 at 5:14 am
With the exceptions of indexes created as part of a primary key or unique constraint, then no. SQL does not save the creation/modify date of indexes. you can get the...
Lowell
July 9, 2017 at 8:56 pm
Viewing 15 posts - 421 through 435 (of 13,460 total)