Re: fix stats_fetch_consistency value in postgresql.conf.sample

On Wed, Mar 29, 2023 at 11:03:59PM -0500, Justin Pryzby wrote:
> On Wed, Jul 13, 2022 at 04:49:00PM -0700, Andres Freund wrote:
> > On 2022-07-14 08:46:02 +0900, Michael Paquier wrote:
> > > On Wed, Jul 13, 2022 at 12:30:00PM -0500, Justin Pryzby wrote:
> > > > How did you make this list ?  Was it by excluding things that failed for you ?
> > > > 
> > > > cfbot is currently failing due to io_concurrency on windows.
> > > > I think there are more GUC which should be included here.
> > > > 
> > > > http://cfbot.cputube.org/kyotaro-horiguchi.html
> > > 
> > > FWIW, I am not really a fan of making this test depend on a hardcoded
> > > list of GUCs.
> > 
> > I wonder if we should add flags indicating platform dependency etc to guc.c?
> > That should allow to remove most of them?
> 
> Michael commented on this, but on another thread, so I'm copying and
> pasting it here.
> 
> On Thu, Mar 23, 2023 at 08:59:57PM -0500, Justin Pryzby wrote:
> > On Fri, Mar 24, 2023 at 10:24:43AM +0900, Michael Paquier wrote:
> > > >> * Check consistency of GUC defaults between .sample.conf and pg_settings.boot_val
> > > >   - It looks like this was pretty active until last October and might
> > > > have been ready to apply at least partially? But no further work or
> > > > review has happened since.
> > > 
> > > FWIW, I don't find much appealing the addition of two GUC flags for
> > > only the sole purpose of that,
> > 
> > The flags seem independently interesting - adding them here follows
> > a suggestion Andres made in response to your complaint.
> > 20220713234900.z4rniuaerkq34s4v@awork3.anarazel.de
> > 
> > > particularly as we get a stronger
> > > dependency between GUCs that can be switched dynamically at
> > > initialization and at compile-time.
> > 
> > What do you mean by "stronger dependency between GUCs" ?
> 
> I'm still not clear what that means ?

Michael ?

This fixes an issue with the last version that failed with
log_autovacuum_min_duration in cirrusci's pg_ci_base.conf.

And now includes both a perl and a sql-based versions of the test - both
of which rely on the flags.