OK, so you want to concatenate all the preceding ReportingGroups into one column. I've got to ask - how many rows do you or will you have? If...

Yes, if you have multiple matches in the second table then you'll get one row for each in your result set. Now, what results are you expecting to see?

John

If there are only three columns, you might want to take a look at today's headline article[/url]. Of course, if this is something you're going to do regularly on...

Does Luis's solution get you anywhere close? If not, you might try including database name (or id) and date (with time portion, if any, stripped out) in your join...

So you have a table of DBCC check results and a table of backup results, and you want to join them? I'm not sure why you'd want to do...

Yes, and furthermore, a table variable only lives for the duration of the batch in which it is created, so you may find it difficult, if not impossible, to use...

Run that second subquery on its own, and you'll see it returns a different result for each of the two guises. That's why the whole thing returns a different...

Sounds like an extension of the UPDATE...FROM issue, where you can get unreported cardinality errors. If you update TableA based on a join to TableB, and there are multiple...

Is it the exact same message (unpersonalised) that you want to send to each recipient. You can send the same message to multiple recipients, so why not build the...

I've had this problem before. If I remember correctly, I solved it by creating stored procedures - one for each action (sp_start_job etc) and assigning the user permission to...

Ian

What happens if you CAST after doing the SUM?

John

Yes - every time you use dm_db_index_physical_stats, SQL Server has to go away and examine the indexes to find out their fragmentation level. Even though you've filtered on those...

No, do you have plenty of space on the disk(s) on which the msdb data and log files are located?

John

In that case, it's even simpler.

SELECT

id

,product

,MIN(start_date) AS StartDate

,NULL AS EndDate

FROM @t

GROUP BY

id

,product

HAVING COUNT(*) > COUNT(end_date)

But what happened to the DATEDIFF requirement?

John

That any integer is just a placeholder. As you've already noticed, you'll get the same result whatever argument you put in there. What you do have to be...