Numerous...

There are a lot of things that can affect query plan reuse, and most have some level of nuance or persnickety thing about them. With your stuff I would look...

YES...

ffarouqi (1/29/2016)

---

Come on people, forums are meant to ask questions and help each other in a good way. I know you all have vast amount of experience that does not...

GilaMonster (1/29/2016)

---

Since I wrote that article, it's already in my words.

Skim the first section and read the second section. You can skip the rest of it for now. The...

You didn't include any query, nor any logic describing the output.

Help us help you by providing CREATE TABLE, INSERTS of data and said logic explaining your expected output.

Right now plan cache issues is the LEAST of your problems, assuming your two fields in that update are not actually VARCHAR(8000)!!! Someone has set the database execution setting in...

Jason A. Long (1/28/2016)

---

I just wanted to give a HUGE thank you to Kevin. He gave his "Common SQL Programming Mistakes" presentation for out company today.

It's an outstanding presentation. If...

curious_sqldba (1/28/2016)

---

I am big fan of Ola's script too, the only problem is it doesn't create multiple jobs. I used his script for DBCC but i had to make...

i_am_moose (1/28/2016)

---

I'm working as a consultant at a company which is running their production SQL Server (2012) on a VM. I don't know which VM platform is being used....

I STRONGLY recommend everyone use Ola.Hallengren.com's full maintenance suite. FREE, incredibly flexible and powerful, supported, does ALL of your MX needs.

A key point though is that YOU must understand...

Oh, another thing I do when batching bulk stuff like this is to drop in a WAITFOR DELAY '00:00:0.x' or something appropriate to give the server some breathing room and...

Did you observe this same behavior when you tested the SP3 upgrade BEFORE you rolled it into production? You did DO that, right?!? 🙂

Did you review the complete fix list...

There is nothing before the @@ROWCOUNT check in the WHILE loop, so clearly one connection does SOMETHING on the spid that gets you an @@ROWCOUNT return other than 0.

If...

1) WHILE looping is essentially the same as a cursor - Row By Agonizing Row processing either way.

2) In this case it was appropriate to do that because you WANT...