Tom,

Can you post the XML as an attachment?

Without seeing what you're looking for, take a look at examples that use FOR XML PATH instead of FOR XML AUTO. ...

Edwin,

Based on your architectural diagram you won't need the Fibre Channel Switch. The SAN you listed is an iSCSI solution so you only need Gigabit switched ethernet.

Also, you might want...

What does that SP do? Can you post a script of that as an attachment?

Maybe it worked at one time because a password was saved somewhere such...

Is the @username that's passed into that stored procedure a SQL login? Do all usernames come back with the same error?

Dave,

What does the stored procedure csp_AddDMUser do? Is it that you can login to the application using the sa account but that the stored procedure errors out?

Take a look at the REVERSE function. The steps would look something like this.

Initial string: you are working

Reversed: gnikrow era uoy

Extract first word: gnikrow

Reversed: working

First thought is to make your sa password complex enough to meet the password complexity requirements.

Otherwise, to disable that check; [font="Courier New"]ALTER LOGIN [sa] WITH CHECK_POLICY=OFF[/font]

Depends on if the client you're using to connect to SQL supports redirection.

Have a look at this article: Implementing Application Failover with Database Mirroring

Using FOR XML PATH, this should be close.

SELECT c.idChemical as "c/@idChemical",

(SELECT s.chemicalSubLeft as "s/@SpecDescription" FROM chemicalSubInfo s WHERE c.idChemical = s.idChemical FOR XML PATH (''), TYPE),

(SELECT s.chemicalSubLeft as "p/@measurement" FROM...

To see the DBCC PAGE output you have to enable trace flag 3604; DBCC TRACEON(3604).

https://blogs.msdn.com/sqlserverstorageengine/archive/2006/06/10/625659.aspx

So for example, if SPID 55 is killed by SPID 80 then you want to know what SPID 55 was doing?

How about starting with the replay trace template?

You should be able to get the spid and hostname for a kill using the ErrorLog class; filter on something like 'killed'.

Keith,

Yes, the Rebuild Index is probably what's causing the transaction log to grow. That rebuilds all indexes and index rebuilds are logged operations.

Something you might want to consider is...

ericzbeard (8/11/2008)

---

It was "set arithabort off". I haven't figured out why yet. That's the last thing I would have considered.

I just assumed that the defaults sent in by...

In the original query window you were probably getting a different execution plan. Then when you set the options you likely got the same plan as the client app, ...