Re: autovacuum and TOAST tables

Alvaro Herrera <alvherre(at)commandprompt(dot)com> writes:
> Note that this patch allows a toast table to be vacuumed by the user:
> I don't have a problem with that, but if anyone thinks this is not a
> good idea, please speak up.

The permissions on pg_toast will prevent anyone but a superuser from
doing that anyway, so it's no big deal.

Possibly more interesting is what happens if someone drops the parent
table while VACUUM is working independently on the toast table. Does
DROP take exclusive lock on a toast table? Probably, but it needs
to be checked. I think preventing that scenario was one reason why
the vacuuming was tied together way back when.

(The same goes for any other parent-table DDL action that would affect
the toast table; CLUSTER or TRUNCATE for instance.)

regards, tom lane