Jagadeesh Chowdary (4/19/2014)

---

If the Database is in Suspect mode then follow the below steps to recover ur DB

>Stop the SQL Service

>>Copy the MDF and LDF file and place in other...

PDW is different and you do have to create stats manually, but that's specific to PDW and not something you typically do on normal SQL databases. You'll need to read...

You break your log chain and sit with potentially a many hour recovery after the...

Chapter 3: https://www.simple-talk.com/books/sql-books/troubleshooting-sql-server-a-guide-for-the-accidental-dba/

No, that's exactly the same as the delete without a transaction. It's a single statement, so it's implicitly in its own transaction.

Delete in batches, not in a transaction and have...

You can try that. You may have the entire DB offline for hours or days as a result.

The approximation is just that. An approximation.

If the view has an index, SQL may not need to go to the base tables at all.

Bear in mind there are a massive set of restrictions around indexed views....

The Associated_object_id is not an object ID, it's too big. It's an allocation unit ID (or HoBT). See sys.partitions and/or sys.allocation_units

Then it's probably just chugging along. Rollbacks usually take longer than the initial changes did.

Kick all other users off the server will help. :hehe: 🙂

Check what the rollback is waiting for (wait type in sys.dm_exec_requests) first.

Unfortunately the only thing you can do with a rollback is wait. It has to finish.

Don't restart SQL or the rollback will just start over from the beginning with the...

Cool, I'll scratch it off my to-write list then.

Schema changes are in the default trace, though only for a while. It keeps 5 files of 20MB each. On some servers I've seen that overwritten within 4 hours.

Otherwise, unless...

What do you mean by 'priority'? They can run concurrently.

That makes sense.