Re: BUG #14110: information_schema.constraint_column_usage not working as expected - Mailing list pgsql-bugs

From David G. Johnston
Subject Re: BUG #14110: information_schema.constraint_column_usage not working as expected
Date
Msg-id CAKFQuwZ04MmhTfj7Vyr99p4QyMrFO6udLbFL=+6gw6-d=uc5Lw@mail.gmail.com
Whole thread Raw
In response to BUG #14110: information_schema.constraint_column_usage not working as expected  (srkunze@mail.de)
List pgsql-bugs
On Monday, April 25, 2016, <srkunze@mail.de> wrote:

> The following bug has been logged on the website:
>
> Bug reference:      14110
> Logged by:          Sven R. Kunze
> Email address:      srkunze@mail.de <javascript:;>
> PostgreSQL version: 9.3.5
> Operating system:   OpenSUSE
> Description:
>
> Hi,
>
> I hope it qualifies as a proper bug. Since the issue is a bit longer, I
> will
> cut it into several pieces for better digestion.
>
> Interestingly, not a single database on our ci machines showed this
> behavior.
>
>
> Even more noticeable is the fact that user 'postgres' **can** read
> information_schema.constraint_column_usage of the "broken" database
> properly.
>
>
> However, '\du' reveals no differences between database users who read
> constraint_column_usage properly and those who don't.
>
>
This is unlikely to be a bug.  That view is filtered by current user
permissions.  Your discrepancies likely can be explained by variations in
object ownership between the various database being queried.

If you want further help I'd suggest picking two databases, one working and
one not, and asking on -general how to go about comparing the two setups
for differences that would prove this to your satisfaction.

Even if this was a bug you have not come close to prividing sufficient info
for us to confirm it.  A self-contained psql script exhibiting the
behavior would be necessary.

It's likely \du is insufficient to explore this, you have to look at the
tables, not the users.  You should see differences is ownership or grants.

http://www.postgresql.org/docs/8.0/static/infoschema-constraint-column-usage.html

David J.

pgsql-bugs by date:

Previous
From: "David G. Johnston"
Date:
Subject: Re: BUG #14113: LIKE INCLUDING STORAGE bug?
Next
From: Thomas Munro
Date:
Subject: Re: BUG #14112: sorting v and w is broken with et_EE locate