Re: timeout implementation issues

Bruce Momjian wrote:
> 
> Hiroshi Inoue wrote:
> > Bruce Momjian wrote:
> > >
> > > Hiroshi Inoue wrote:
> > > > > > Why should the timeout be reset automatically ?
> > > > >
> > > > > It doesn't need to be reset automatically, but the problem is that if
> > > > > you are doing a timeout for single statement in a transaction, and that
> > > > > statement aborts the transaction, the SET command after it to reset the
> > > > > timeout fails.
> > > >
> > > > As for ODBC, there's no state that *abort* but still inside
> > > > a transaction currently.
> > >
> > > Yes, the strange thing is that SET inside a transaction _after_ the
> > > transaction aborts is ignored, while SET before inside a transaction
> > > before the transaction aborts is accepted.
> >
> > What I meant is there's no such problem with psqlodbc
> > at least currently because the driver issues ROLLBACK
> > automatically on abort inside a transaction.
> 
> If it does that, what happens with the rest of the queries in a
> transaction?  Do they get executed in their own transactions, or are
> they somehow ignored.

They would be executed in a new transaction. Queries shouldn't
be issued blindly(without error checking).

regards,
Hiroshi Inoue