Re: Function to kill backend

Tom Lane wrote:
> Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
> > If there is a problem, maybe we can fix it, or perhap have the kill
> > function use SIGINT, then wait for the query to cancel, then SIGTERM.
>
> Well, if someone could prove that the SIGTERM path is equivalent to
> a transaction-aborting error followed by normal client disconnect,
> I'd feel a lot better about its reliability. We know those two code
> paths work well.

As far as I know, most database systems have the ability to kill an
individual session, and I think we need that functionality too, as I
described in my report example.

> Right now I do not think they are equivalent, but maybe they could be
> made so.

Let me look into it.

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073