From: | Alvaro Herrera <alvherre(at)alvh(dot)no-ip(dot)org> |
---|---|
To: | Sven Klemm <sven(at)timescale(dot)com> |
Cc: | pgsql-hackers(at)lists(dot)postgresql(dot)org |
Subject: | Re: Fix dropped object handling in pg_event_trigger_ddl_commands |
Date: | 2021-04-25 19:40:40 |
Message-ID: | 20210425194040.GA11244@alvherre.pgsql |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On 2021-Apr-25, Sven Klemm wrote:
> On Sun, Apr 18, 2021 at 2:12 PM Sven Klemm <sven(at)timescale(dot)com> wrote:
> > when creating an event trigger for ddl_command_end that calls
> > pg_event_trigger_ddl_commands certain statements will cause the
> > trigger to fail with a cache lookup error. The error happens on
> > master, 13 and 12 I didnt test any previous versions.
> >
> > trg=# ALTER TABLE t ALTER COLUMN f1 SET DATA TYPE bigint, ALTER COLUMN
> > f1 DROP IDENTITY;
> > ERROR: XX000: cache lookup failed for relation 13476892
> > CONTEXT: PL/pgSQL function ddl_end() line 5 at FOR over SELECT rows
> > LOCATION: getRelationTypeDescription, objectaddress.c:4178
>
> Any opinions on the patch? Is this not worth the effort to fix or is
> there a better way to fix this?
Hello, I haven't looked at this but judging from the general shape of
function and error message, it seems clearly a bug that needs to be
fixed somehow. I'll try to make time to look at it sometime soon, but I
have other bugs to investigate and fix, so it may be some time.
I fear your proposal of ignoring the object may be the best we can do,
but I don't like it much.
--
Álvaro Herrera 39°49'30"S 73°17'W
"La verdad no siempre es bonita, pero el hambre de ella sí"
From | Date | Subject | |
---|---|---|---|
Next Message | Andres Freund | 2021-04-25 22:20:53 | GISTSTATE is too large |
Previous Message | Tom Lane | 2021-04-25 19:13:58 | Re: Allowing to create LEAKPROOF functions to non-superuser |