you are getting a conversion error because DATENAME returns nvarchar, not datetime.

nchar/nvarchar columns are for storing unicode strings.

maybe so, but double precision looks better. like you are doing super-fine carpentry work or something, so you need to be doubly...

I came to programming from computational physics. In calculations I often used doubles instead of floats. double is probably the least used datatype in sql server.

create table test...

OPENROWSET does not require a linked server. requires just the connection string to the remote.

you could use OPENQUERY or OPENROWSET

oops, sorry. what I suggested won't work in this case because sp_spacused returns two result sets. you can't use insert/exec if the proc returns two result sets.

another option...

4) how to do my own exams?

if you create a table with the same columns as the result set of sp_spaceused, you can insert into it like this:

insert mytable exec sp_spaceused

also float is really fun to use for surrogate keys.

😉

besides returning data that is bogus, reading NOLOCK can also simply error out. this happens if the process that has the exclusive lock moves the row on a page...

you can select from the table while the update has an exclusive lock only if you do a dirty read. This is done with the NOLOCK keyword. Have...

Please don't write code like this unless you want to be hacked. this proc is a prime target for a sql injection attack.

for dynamic search conditions, you will benefit...

I think OP must be talking about a function in client code.

is this a homework or interview question? 🙂