Viewing 15 posts - 7,561 through 7,575 (of 7,614 total)
That's for sure. I love "experts" who don't know how to tell what the current SQL activity on the box is.
March 5, 2008 at 9:43 am
Interesting. I guess my concern is that if a (very) large table(s) need updated, you could see a significant hit to your performance from autoupdate of stats. Also,...
March 5, 2008 at 9:20 am
First, 100% on the honesty thing. I would detect you and not hire you. Moreover, I would add you to my personal file of "never hires". Just...
March 5, 2008 at 8:49 am
Hmm, but happens if/when your requirements change and a table that was a "prime" table becomes an intersection table, or vice versa? Do you go back thru and rename...
November 20, 2007 at 4:12 pm
Regarding table names, I now prefer singular but can live with plurals ... the main thing is consistency. Whichever you choose, use it thru-out, without regard to whether it's...
November 20, 2007 at 3:58 pm
I've read everything: interesting article and comments!
FWIW, here's my take on a first-cut design *for just books* [personally I would adjust it to allow reviews on other objects -- such...
November 20, 2007 at 10:54 am
I need to find the record in the DB that has a start time >= the time of my call and on or after the day of the week...
September 26, 2007 at 7:50 am
I think you need to check the current date instead of automatically using s.enddate in the calcs. For example:
DATEDIFF(DAY, s.startdate, CASE WHEN GETDATE() < s.enddate THEN GETDATE() ELSE s.enddate...
September 26, 2007 at 7:45 am
Essentially, @tbl is not getting populated with the first "insert to @tbl"
The INSERT will not, and should not, be reflected in the table until the statement has finished....
September 26, 2007 at 7:37 am
If it's not running from SQL, how does it even have a spid? I thought spid was used by SQL, kpid was used by Windows.
May 24, 2007 at 8:38 am
>> I'm pretty sure that when OSQL is used, any procs called by it are given the same spid. <<
Yes, of course.
But the osql task itself will have a...
May 24, 2007 at 8:13 am
The index should be on the search column(s), not on an id column.
May 23, 2007 at 12:41 pm
Yes, if you want to add an index to a small table, and you should for many of them, always make it a unique, clustered index. For example, for the sample...
May 23, 2007 at 10:21 am
I thought the osql was a separate process and so would get its own spid. How will you know which row in the table the child process needs to get?
May 23, 2007 at 10:15 am
Probably easiest is to do something like this (handles up to 24000, assuming PRINT can handle that):
print space(@hangingIndent) + replicate(@startCharacter, CASE WHEN @widthOfPattern > 8000 THEN 8000 ELSE @widthOfPattern END)...
May 23, 2007 at 8:28 am
Viewing 15 posts - 7,561 through 7,575 (of 7,614 total)