Thanks all for replies.

Jonathan, your solution works. The only downside is that it creates blockings. But this is what was expected.

Thanks Mike!

This is exactly what I wanted.

And yes, each new row should accumulate all the previous one plus the current one.

Thanks

Here is DDL:

create table table1
(
col_to_order int identity(1,1),
column1 int,
column2 varchar(max)
)
go

insert into table1
(column1)
values
(123),
(71),
(1844),
(191)
go

No, I don't have Visual Studio. And I am looking for SQL solution.

Apparently it's XPATH. It looks like this:

SELECT ActivityLogXml.Activity.query('Comments').value('.', 'varchar(max)')
FROM @ActivityLogXml.nodes('/Activity') AS ActivityLogXml(Activity)

And the part ActivityLogXml.Activity.query is interpreted by sys.sql_expression_dependencies as db.schema.object

I tried SQL Server Logs which is under Management folder, it does not have this info. Where I can find connection logs?

Thanks

Thank you Grant, and all, for your advises.

What I also heard is that we need to collect a database workload from production and apply it on our testing environment to...

They are views? But I specified o.type_desc = 'SQL_TABLE_VALUED_FUNCTION' in Where clause.

This is not from production. This is from one lower env. to another lower env. (like dev, test, qa, like that)

Thanks all for replies. Now I am convinced, will leave the 2nd index in place.

Jet Blue.

The reason for these NOLOCK's is mostly performance, they don't want to be blocked.

Thanks both, it worked.

My goal is not a resultset.

What I have now is two queries which separately assign values to two single variables:

select @v1 = amt 
from #t1
where type =...