> On 20 Aug 2025, at 00:55, Tom Lane <t...@sss.pgh.pa.us> wrote:
>
> Andrey Borodin <x4...@yandex-team.ru> writes:
>> I believe there is a bug with PageIsAllVisible(page) &&
>> visibilitymap_clear(). But I cannot prove it with an injection point test.
>> Because injections points rely on CondVar, that per se creates corruption in
>> critical section. So I'm reading this discussion and wonder if CondVar will
>> be fixed in some clever way or I'd better invent new injection point wait
>> mechanism.
>
> Yeah, I was coming to similar conclusions in the reply I just sent:
> we don't really want a policy that we can't put injection-point-based
> delays inside critical sections. So that infrastructure is leaving
> something to be desired.
>
> Having said that, the test script is also doing something we tell
> people not to do, namely SIGKILL the postmaster. Could we use
> SIGQUIT (immediate shutdown) instead?
I'm working backwards from corruptions I see on our production.
And almost always I see stormbringers like OOM, power outage or Debian scripts
that (I think) do kill -9 when `service postgresql stop` takes too long.
Best regards, Andrey Borodin.
