Re: BUG #15437: Segfault during insert into declarative partitioned table with a trigger creating partition

Amit Langote <langote_amit_f8@lab.ntt.co.jp>

From: Amit Langote <Langote_Amit_f8@lab.ntt.co.jp>
To: Michael Paquier <michael@paquier.xyz>, Dmitry Shalashov <skaurus@gmail.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, pgsql-bugs@lists.postgresql.org
Date: 2018-10-29T05:15:06Z
Lists: pgsql-bugs
On 2018/10/29 11:44, Michael Paquier wrote:
> On Fri, Oct 19, 2018 at 12:38:09PM +0300, Dmitry Shalashov wrote:
>> That may be off-topic here, but I feel that *some* way of auto-creating
>> partitions would be useful though.  Maybe all partitioning setup could
>> boil down to one table declaration, with auto-created partitions of
>> the same structure, including indices and stuff.
> 
> Some patches have been proposed for this purpose during v11 development
> timeline if I recall correctly, but they did not cross the finish line.

Hmm, I don't recall any, but maybe I missed some thread.

> You could also look at something like pg_partman:
> https://github.com/pgpartman/pg_partman

I haven't looked at how pg_partman does it, but I suspect it's not doing
it inside a trigger or we would've heard of this crash from pg_partman's
developers. :)  I don't hear you saying it does, but just thought I'd
mention it.

Thanks,
Amit



Commits

  1. Block creation of partitions with open references to its parent