Re: missing locking in at least INSERT INTO view WITH CHECK

On 2015-08-27 18:37:10 +0100, Dean Rasheed wrote:
> On 27 August 2015 at 12:18, Andres Freund <andres(at)anarazel(dot)de> wrote:
> >>
> >> /*
> >> + * If the view query contains any sublink subqueries, we should also
> >> + * acquire locks on any relations they refer to. We know that there won't
> >> + * be any subqueries in the range table or CTEs, so we can skip those, as
> >> + * in AcquireRewriteLocks.
> >> + */
> >> + if (viewquery->hasSubLinks)
> >> + query_tree_walker(viewquery, acquireLocksOnSubLinks, NULL,
> >> + QTW_IGNORE_RC_SUBQUERIES);
> >> +
> >> + /*
> >> * Create a new target RTE describing the base relation, and add it to the
> >> * outer query's rangetable. (What's happening in the next few steps is
> >> * very much like what the planner would do to "pull up" the view into the
> >
> > These days this seems to require a context parameter being passed
> > down. Other than that this patch still fixes the problem.
> >
>
> Yes, I concur. It now needs an acquireLocksOnSubLinks_context with
> for_execute = true, but otherwise it should work.

Ok, I'll push that then, unless somebody else wants to do the honors.

> I have a feeling that RLS might suffer from the same issue, but I
> haven't looked yet.

There's a similar issue there, yes:
http://archives.postgresql.org/message-id/20150827124931.GD15922%40awork2.anarazel.de

Are you thinking of that angle, or yet another one?

Regards,

Andres