Hi Steve,

I usually have no problem with a somewhat vague question if I can at least dedduct what line of thinking the author had. In this case, I had no...

Hi Adam,

My build is reported as 9.00.3054.00 by @@VERSION.

The correct answer is not listed. There *IS* a way to test this reliably, but testing for membership of the db_owner role is not near enough. It is in fact...

Hi Adam,

Even though you made a typo, the question was correct. I just tested it and for this query:

SELECT *

FROM n

WHERE

id = 1

AND id2 IN (2)

the execution plan was generated for

Hi Al,

>>In theory the above INSERT statement should throw a foreign constraint error because when it inserts the very first record (Paul, Alice) there is no record with 'Alice' primary...

Good point, Kelvin!

Hi,

The explanation of this question says:

"This statement works like the ROLLBACK TRANSACTION statement, but it rolls back to the outer transaction and resets @@trancount to 0"

Whereas it should say:

"This statement...

I hate questions like these, because they test my ability to read irrelevant details rather than my understanding of important issues.

Important issues are knowing who is and who isn't allowed...

I like the question, but I fail to see why it's included in the "SQL2005 - T-SQL" category. There's not a single word of T-SQL in the question, and cohesion...

Steve: Once more (and than I'll drop this). BOL is confusingly worded. The counters are *NOT* reset. You don't need to reword the question (but adding a bit to the...

TDuffy: See my previous reply to Tao. Query this view on a just recycled server, and you get (almost) nothing.

Tao: BOL is indeed misleading. You can try it if you have a test server at your disposal - run some queries, check that various indexes used show up in...

I really have to disagree here (edit - I mean that I disagree with the article, not with the comments). People should always take responsibility for what they do. Doing...

Hi Steve,

Thanks for addressing this (and good to see you back online).

>>The write performance of RAID 5 is usually slower because the information has to be written + parity calculated...

Hi Craig,

Not really an answer to your question, but I don't think adding countless extra disks to a RAID 5 array is wise. The prime motivation of RAID is reducing...