SQL Kiwi (9/7/2012)

---

I don't think it proves anything.

Logical/physical reads/writes are about pages read/written.

There was never an argument that SQL Server finds pages affected by the UPDATE not changed and...

SQL Kiwi (9/7/2012)

---

I have said, over and over again, that I don't know whether the in-memory page is written to or not. A page that is exactly the...

MissTippsInOz (9/7/2012)

You have had numerous genuine responses from people who are trying to help you to understand the behaviour of a 'non-updating update'. You have been given code (and...

Key word - changes.

No matter if it's rolled...

SQL Kiwi (9/7/2012)

---

And that is exactly the problem. My question to...

So, there are modifications?

SQL...

What exactly is logical update operation and what other kind of update operatios are there?

You may...

Sergiy (9/6/2012)

---

SQL Kiwi (9/6/2012)

---

No idea, actually. Perhaps it's just simpler to always take an exclusive lock prior to possibly changing data. Perhaps it wouldn't be safe to allow...

SQL Kiwi (9/6/2012)

---

I thought you...

SQL Kiwi (9/6/2012)

---

No idea, actually. Perhaps it's just simpler to always take an exclusive lock prior to possibly changing data. Perhaps it wouldn't be safe to allow people...

I've come accross a very ineresting thing.

I was trying to test for some other case, but the original test geve me some different result.

I was running this:

BEGIN TRANSACTION

UPDATE dbo.test

SET...

SQL Kiwi (9/6/2012)

---

The two lock types are different,...

SQL Kiwi (9/6/2012)

---

By updating you're losing data.

You might think you don't need it anymore, but still - you're losing data previosly recorded in the database.

Which...

Can you bring sopme details please?

Eugene Elutin (8/29/2012)

I would probably agree on that with one small addition: such verification should be made in a client app not in the update statement itself, otherwise you may...