Stephen Frost <sfrost@snowman.net> writes:
> Ugh. That doesn't make it correct though.. We really should be using
> has_privs_of_role() for these cases (and that goes for all of the
> default role cases- some of which are correct and others are not, it
> seems).
I have a different concern about this patch: while reading statistical
values is fine, do we REALLY want pg_read_all_stats to enable
pg_stat_get_activity(), ie viewing other sessions' command strings?
That opens security considerations that don't seem to me to be covered
by the description of the role.
regards, tom lane