From: | Michael Paquier <michael(dot)paquier(at)gmail(dot)com> |
---|---|
To: | "Shinoda, Noriyoshi" <noriyoshi(dot)shinoda(at)hpe(dot)com> |
Cc: | "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org>, "snaga(at)uptime(dot)jp" <snaga(at)uptime(dot)jp> |
Subject: | Re: pg_sequences bug ? |
Date: | 2017-02-04 12:30:13 |
Message-ID: | CAB7nPqSeg2rU6AkaYZUMN_tz00q6Bx+EQ7X7kbWZoc1kx5VXpw@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Sat, Feb 4, 2017 at 2:50 PM, Shinoda, Noriyoshi
<noriyoshi(dot)shinoda(at)hpe(dot)com> wrote:
> I tried a committed pg_sequences for PostgreSQL 10dev (https://commitfest.postgresql.org/12/771/)
> I found that when multiple users create SEQUENCE, I cannot see the pg_sequences catalog. I think that should work just like pg_tables.
>
> $ psql -U user1
> postgres=> CREATE SEQUENCE seq1 ;
> CREATE SEQUENCE
>
> $ psql -U user2
> postgres=> CREATE SEQUENCE seq2 ;
> CREATE SEQUENCE
> postgres=> SELECT * FROM pg_sequences ;
> ERROR: permission denied for sequence seq1
>
> Apparently it seems that the pg_sequence_last_value function included in the pg_sequences view definition cannot be executed.
> Is this behavior supposed?
That seems user-unfriendly to me.
We could perhaps just use has_sequence_privilege() and return NULL if
the caller of pg_sequences does not have select and usage access to a
given sequence? Please see the patch attached.
--
Michael
Attachment | Content-Type | Size |
---|---|---|
pgsequences-read.patch | text/x-patch | 1.5 KB |
From | Date | Subject | |
---|---|---|---|
Next Message | amul sul | 2017-02-04 13:06:25 | Re: Constraint exclusion failed to prune partition in case of partition expression involves function call |
Previous Message | Ashutosh Sharma | 2017-02-04 12:06:23 | Re: pageinspect: Hash index support |