Subject | Re: [ib-support] OIT/OAT question |
---|---|
Author | Doug Chamberlin |
Post date | 2002-01-29T11:32:34Z |
At 01/29/2002 03:29 AM (Tuesday), Claudio Valderrama C. wrote:
number of the most recent transaction which used snapshot transaction
isolation level? Or is it the oldest transaction of that type which is
still uncommitted? Looks to me that it is the oldest uncommitted.
I understand all the isolation levels. I'm just looking for a definition of
what numbers are being reported here.
>"Doug Chamberlin" <DChamberlin@...> wrote in messageSo you are saying that what is being reported here is the transaction
>news:5.1.0.14.2.20020128162853.026d1a38@......
> > Database "d:\interbase\database\KSGBUD06.gdb" Taken at 15.40
> >
> > Oldest transaction 75409
> > Oldest active 75410
> > Oldest snapshot 73862
> > Next transaction 75731
> > Bumped transaction 1
> >
> > The spread on the transactions isn't too bad, but I am suspicious about
>the
> > "Oldest snapshot". Does anyone know what it is?
>
>Indeed, the snapshot number looks strange, but what should the header say
>instead if there's no snapshot transaction newer than the oldest txn left by
>the last sweeping? It may have been the case that only read committed txns
>where used all the time.
>
>Snapshot, concurrency or repeatable read are equivalent names for the
>default isolation level used in the engine, contrary to read committed,
>that's the default other engines use. Don't know is that was your question.
number of the most recent transaction which used snapshot transaction
isolation level? Or is it the oldest transaction of that type which is
still uncommitted? Looks to me that it is the oldest uncommitted.
I understand all the isolation levels. I'm just looking for a definition of
what numbers are being reported here.