You haven't provided any sample data for the second table, so it is not clear to me what you are asking.

Have you tried a solution? What did you come...

It's the total length that counts, not the individual columns - as it says in the error message.

If you still can't see the problem, post the query plus the column...

Hey,

You say that this table has heavy insert/update activity - inserts are not a problem, but I am at a loss to understand how updates can be performed in reasonable...

RBarryYoung (6/4/2009)

---

No, I was just referring to the fact that there has to be cross-CPU synchronization of some kind in order to correctly implement a sequential counter, whether it's mutexes,...

No worries - you helped me earlier!

Sounds intriguing - the thing I...

On a separate note, it's kinda sucky that NEWSEQUENTIALID has to be protected by a mutex don't you think?

I don't suppose having all calls to UuidCreateSequential serialized causes many people...

Jeff Moden (6/4/2009)

---

I agree... I must think differently and use the wrong search words. It's always a PITA for me to do a search on such a thing. ...

Just...

Very cool stuff Jeff. And amusing too.

I do hope the term "passive pork chop" sticks. I like it.

Helped. I think. 😀

Ok. So when you start the snapshot agent for a merge-replicated table, master.[sys].[sp_MSprepare_mergearticle] is called - which contains:

if @snapshot_ready = 2 or (@article_status 2...

Apparently, merge replication changes to sequential by default - could this have happened in your case?

I can't find a great resource to detail under what circumstances newid changes to newsequentialid,...

Thanks! I should have checked the site myself - apologies for being lazy 🙁

Well that sucks - I just got the May edition in my inbox yesterday. Where's my June copy??? Grrr!

Thanks Wayne, I will look forward to receiving the...