Are you sure the log backups and the full/differential are from the same database on the same server? There has to be a log that overlaps with the LSN on...

You could try doing the insert into the new table without having the key in place and then apply the key after the data is loaded. In some situations you...

sys.dm_exec_requests will show the execution of the query and any waits that it's experiencing as well as any blocks.

I'd suggest looking at the combined execution plan. It might be different...

If you aren't using them, cleaning them up isn't going to hurt anything.

Is the book by Denny Cherry? If so, I'd take it's word as pretty near to gospel.

sdownen05 (4/4/2014)

---

Keith,

First, thanks for pointing me to the other scripts. I will definitely start using them instead.

As to your query, it shows that the table exists. However, when I do...

It's also not a question of web servers vs. something else. Query behavior is query behavior. Nothing all that special about queries coming from a web interface vs. queries coming...

I'd suggest using extended events for this. They're more light weight than trace events and their filtering is much, much better, so you'll only be capturing the data you want....

If you have the clustered index key value in the WHERE clause and you're still getting a scan, the suggestion is that you must have one of two situations, your...

That's one way, sure. I'd probably use extended events, or, on a 2008 system, trace....

But you can see an aggregation of the queries currently in cache in sys.dm_exec_query_stats. It shows execution times, counts, other stuff like that, but not who did the executing or...

It goes back to the histogram on the statistics. That is only build on the first column in the index key. The others are used to help establish the density...

Query performance measures alone won't show you what is blocking or being blocked. Since you're on 2012 you can use the blocked_process_report extended event. That will tell you what is...

Best thing to do is get that thing local. Any cross network process you go through is going to be extremely painful.

You measure the speed at which queries that access that index are performing before and after you defragment the indexes.

patrickmcginnis59 10839 (4/3/2014)

---

Grant Fritchey (4/3/2014)

---

It's an approach taken by a developer or development team that doesn't understand SQL Server, or, just didn't want to take the time to do things...