In general optimistic concurrency scales better.
In the world of multipled DB connections and short transactions, I am not
clear when you can ever use long locks effectively.
-Chris.
> -----Original Message-----
> From: Nittle Gupta [SMTP:[EMAIL PROTECTED]]
> Sent: Friday, December 03, 1999 6:53 AM
> To: [EMAIL PROTECTED]
> Subject: Re: Transaction isolation level useless at the ejb server
> level?
>
> Then which one is better pessimistic or optimistic.Then only we can
> decide for the application server.
>
> >----------
> >From: Rickard �berg[SMTP:[EMAIL PROTECTED]]
> >Reply To: A mailing list for Enterprise JavaBeans development
> >Sent: Friday, July 04, 1997 2:39 PM
> >To: [EMAIL PROTECTED]
> >Subject: Re: Transaction isolation level useless at the ejb server
> >level?
> >
> >Nittle Gupta wrote:
> >> Yes,I too want the answere to this question from sun people.
> >> >>If two client requests arrive to the same entity
> >> >>object (i.e. same primary key),
> >> >>the EJB container can do one of the following:
> >> >>
> >> >> - use a different bean class instance for each
> >> >>transaction context, or
> >> >>
> >> >> - use a single instance but serialize the use of
> >> >>the instance by the transactions
> >> >>
> >> >>If the first method is used, the underlying resource
> >> >>managers will properly
> >> >>isolate the work of the multiple clients.
> >> >
> >> >
> >> >I am a bit puzzled by this answer. In the situation
> >> >posited above, if two copies of an entity class are
> >> >used (one per transaction), and each transaction
> >> >modifies its copy, which copy is valid, which is
> >> >stored?
> >
> >This depends on the locking mechanism of the database, and which
> >transaction isolation level is set. It is my understanding that if
> >optimistic locking is used and two concurrent transactions use the same
> >rows in a read/write scenario, the one to commit last will rollback
> >(due
> >to the data changes by the first transaction). The transaction will in
> >this case have to be re-done in order to commit properly.
> >
> >AFAIK
> >
> >/Rickard
> >
> >--
> >Rickard �berg
> >
> >@home: +46 13 177937
> >Email: [EMAIL PROTECTED]
> >Homepage: http://www-und.ida.liu.se/~ricob684
> >
> >========================================================================
> >===
> >To unsubscribe, send email to [EMAIL PROTECTED] and include in the
> >body
> >of the message "signoff EJB-INTEREST". For general help, send email to
> >[EMAIL PROTECTED] and include in the body of the message "help".
> >
>
> ==========================================================================
> =
> To unsubscribe, send email to [EMAIL PROTECTED] and include in the
> body
> of the message "signoff EJB-INTEREST". For general help, send email to
> [EMAIL PROTECTED] and include in the body of the message "help".
===========================================================================
To unsubscribe, send email to [EMAIL PROTECTED] and include in the body
of the message "signoff EJB-INTEREST". For general help, send email to
[EMAIL PROTECTED] and include in the body of the message "help".
