Subject | Re: Closing the gap between OAT and Next transaction |
---|---|
Author | alex_vnru |
Post date | 2002-01-21T11:07:30Z |
--- In ib-support@y..., guido.klapperich@t... wrote:
opened select. Any statement is performed within transaction context,
some connectivity components mask it.
http://www.ibphoenix.com/ibp_oit.html
If you'll have more questions, we can discuss it. BTW, gap you
trying to eleminate is'nt gap at all, my applications don't feel any
performance degradation with daily gap about 30 000, which I eleminate
by night cron using sweep when no one is connected. If someone work at
night this gap can be even more for next day and we neverthelles
have'nt problems. New feature of FB RC2 and IB 6.5 - read only read
commited transactions don't stop OAT/OIT moving - gives additional
advantages on this subject.
Best regards, Alexander V.Nevsky.
> I know, that the gap between Next transaction and OAT comes fromNo, this gap always present if you have connected application with
> disconnecting from the server without commiting all transactions,
opened select. Any statement is performed within transaction context,
some connectivity components mask it.
> but what is the reason for the gap between OAT and OIT ?Rollbaks. Guido, take a look at
http://www.ibphoenix.com/ibp_oit.html
If you'll have more questions, we can discuss it. BTW, gap you
trying to eleminate is'nt gap at all, my applications don't feel any
performance degradation with daily gap about 30 000, which I eleminate
by night cron using sweep when no one is connected. If someone work at
night this gap can be even more for next day and we neverthelles
have'nt problems. New feature of FB RC2 and IB 6.5 - read only read
commited transactions don't stop OAT/OIT moving - gives additional
advantages on this subject.
Best regards, Alexander V.Nevsky.