Tom Lane wrote:
>
> "Hiroshi Inoue" <Inoue@tpf.co.jp> writes:
> >> Doesn't bother me a whole lot; I don't think that's what the die
> >> interrupt is for. In my mind the main reason die() exists is to
> >> behave reasonably when the system is being shut down and init has
> >> sent SIGTERM to all processes.
>
> > In my mind the main reason die() exists is to kill individual
> > backends which seems to be in trouble without causing
> > the database-wide restart.
>
> [ raises eyebrow ] That isn't recommended procedure or even documented
> anywhere, AFAIR.
I don't call it a dbms unless it has a will to limit a
trouble locally.
Anyway it seems too late to complain. I was foolish enough
to have overlooked the very significant change that introduced
the dominant ImmediateInterruptOK variable.
Sigh... Where were my eyes ?
regards,
Hiroshi Inoue