On 2/8/22 10:07, Robert Haas wrote:
> On Tue, Feb 8, 2022 at 10:00 AM Joshua Brindle
> <joshua.brindle@crunchydata.com> wrote:
>> 4 predefined roles currently use has_privs_of_role in master.
>>
>> Further, pg_monitor, as an SQL-only predefined role, also behaves
>> consistently with the INHERIT rules that other roles do.
>>
>> In order for SQL-only predefined roles to ignore INHERIT we would need
>> to hardcode bypasses for them, which IMO seems like the worst possible
>> solution to the current inconsistency.
>
> I agree we need to make the situation consistent. But if you think
> there's exactly one problem here and this patch fixes it, I
> emphatically disagree.
If we were to start all over again with this feature my vote would be to
do things differently than we have done. I would not have called them
predefined roles, and I would have used attributes of roles (e.g. make
rolsuper into a bitmap rather than a boolean) rather than role
membership to implement them. But I didn't find time to participate in
the original discussion or review/write the code, so I have little room
to complain.
However since we did call these things predefined roles, and used role
membership to implement them, I think they ought to behave both
self-consistently as a group, and like other real roles.
That is what this patch does unless I am missing something.
I guess an alternative is to discuss a "proper fix", but it seems to me
that would be a version 16 thing given how late we are in this
development cycle and how invasive it is likely to be. And doing nothing
for pg15 is not a very satisfying proposition :-/
Joe
--
Crunchy Data - http://crunchydata.com
PostgreSQL Support for Secure Enterprises
Consulting, Training, & Open Source Development
