that error looks like you are querying a Sybase database. is that correct?

and what driver are you using? vendor and version.

and... what are you doing with the returned value?

103 (dd/mm/yyyy)  is the wrong format for that input

101 (mm/dd/yyyy) would be the right one

and from your last post you are trying to convert from a varchar to a datetime...

if zero records is not a valid answer then kindly explain why the following code returns zero records and why, according to you, it SHOULD return something.

if...

invalid question - and adding rules afterwards to explain why your desired good answer is right is really not a good thing.

and if there are locking transactions then the query...

and while we at it - always specify the size of your varchars

CONVERT(VARCHAR, MONTH(DateOfServiceFrom))

the replacement of null's with zeros should be done by the presentation layer.

try

select [type]
     , Group1Value
     , Grup2Value
 ...

• This reply was modified 3 years, 3 months ago by frederico_fonseca.
• This reply was modified 3 years, 3 months ago by frederico_fonseca.

this is not a postgreSQL site - you better off posting elsewhere (or read documentation).

you may or not get a reply here.

process is as follows

1. create new table
2. create clustered index (if it is a PK the pk should have a new name associated with the new table)
3. insert data onto...

• This reply was modified 3 years, 3 months ago by frederico_fonseca.

In response to both frederico_fonseca and ratbak

I am using global temp tables because I need to persist the data across SPIDs.  The table in question is built from some...

you can't have 2 processes running concurrently that try to drop/create the same global temp table - you get the conflict you have.

either change it so that the 2 processes...

Steve have a look at this profile https://www.sqlservercentral.com/forums/user/papaku

I'm pretty sure I've reported them more than once - and there are similar profiles to this one - reply to one of...

you can only have exponentials if there datatype is a string...

how you see them in SSMS is not how the data is stored - that is an issue with how SSMS displays real/floats.

to see the real content you will need...

if insert is done in one go ...

adding page compression is likely to reduce the size of that table to around 90 GB - overall result is that querying the table will require less IO and less...