Re: Row security violation error is misleading

On 8 April 2015 at 19:52, Dean Rasheed <dean(dot)a(dot)rasheed(at)gmail(dot)com> wrote:

> 2). In prepend_row_security_policies(), I think it is better to have
> any table RLS policies applied before any hook policies, so that a
> hook cannot be used to bypass built-in RLS.
>

A hook really has to be able to ensure that built-in RLS cannot bypass the
hook's policies, too, i.e. the hook policy *must* return true for the row
to be visible.

This is necessary for mandatory access control hooks, which need to be able
to say "permit if and only if..."

I'll take a closer look at this.

> 3). The infinite recursion detection in fireRIRrules() didn't properly
> manage the activeRIRs list in the case of WCOs, so it would
> incorrectly report infinite recusion if the same relation with RLS
> appeared more than once in the rtable, for example "UPDATE t ... FROM
> t ...".
>

I'm impressed you found that one.

--
Craig Ringer http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services