On Fri, Nov 10, 2023 at 08:32:34PM +0900, torikoshia wrote:
> On 2023-11-10 13:18, Andres Freund wrote:
>> I see no reason to not include slrus. We should never have added the
>> ability to reset them individually, particularly not without a use
>> case - I couldn't find one skimming some discussion. And what's the
>> point in not allowing to reset them via pg_stat_reset_shared()?
>
> When including SLRUs, do you think it's better to add 'slrus' argument which
> enables pg_stat_reset_shared() to reset all SLRUs?
I understand that Andres says that he'd be OK with a addition of a
'slru' option in pg_stat_reset_shared(), as well as including SLRUs in
the resets if everything should be wiped.
28cac71bd368 is around since 13~, so changing pg_stat_reset_slru() or
removing it could impact existing applications, so there's little
benefit in changing it at this stage. Let it be itself.
> As described above, since SLRUs cannot be reset by pg_stat_reset_shared(), I
> feel a bit uncomfortable to delete it all together.
That would be only effective if NULL is given to the function to reset
everything, which is OK IMO, because this is a shared stats.
--
Michael