Re: Visibility regression test

On Thu, 29 Aug 2002 13:27:36 -0400, Tom Lane <tgl@sss.pgh.pa.us>
wrote:
>I don't like depending on a timeout *at all* in a regression test;
>the exact value of the timeout is not particularly relevant to my
>concern about it.

I agree.  But a timeout was the only thing that came to my mind for
aborting an endless loop.  Better suggestions are welcome.  Waiting
for the disk to get full will not be accepted :-)

>It surprises me quite a bit that there aren't any existing spots in
>the regression tests that would expose a Halloween problem ...

Me too.  BTW, why is this called the "Halloween problem"?

> I guess
>my other concern is that we shouldn't need a whole new test for this.

Again I agree.  First I wanted to insert these few lines into an
existing test, but didn't find one, where it seemed to fit.  The most
suitable one seemed to be vacuum.

Servus
 Manfred