Re: replication slot stats memory bug

Andres Freund  writes:
> I saw a leak in pgstat_read_statsfiles(), more precisely:
> 	/* Allocate the space for replication slot statistics */
> 	replSlotStats = palloc0(max_replication_slots * sizeof(PgStat_ReplSlotStats));

Yeah, I just found that myself.  I think your fix is good.

> But it seems like we just shouldn't allocate it dynamically at all?
> max_replication_slots doesn't change during postmaster lifetime, so it
> seems like it should just be allocated once?

Meh.  I don't see a need to wire in such an assumption here.

			regards, tom lane