RE: Parallel INSERT (INTO ... SELECT ...) - Mailing list pgsql-hackers

From Hou, Zhijie
Subject RE: Parallel INSERT (INTO ... SELECT ...)
Date
Msg-id eaeb347d297b4be6b2a660b89cdffb64@G08CNEXMBPEKD05.g08.fujitsu.local
Whole thread Raw
In response to Re: Parallel INSERT (INTO ... SELECT ...)  (Greg Nancarrow <gregn4422@gmail.com>)
Responses Re: Parallel INSERT (INTO ... SELECT ...)  (Greg Nancarrow <gregn4422@gmail.com>)
List pgsql-hackers
Hi,

I took a look into the hasModifyingCTE bugfix recently, 
and found a possible bug case without the parallel insert patch.

---------------------------------
drop table if exists test_data1;
create table test_data1(a int, b int) ;
insert into test_data1 select generate_series(1,1000), generate_series(1,1000);
set force_parallel_mode=on;

CREATE TEMP TABLE bug6051 AS
  select i from generate_series(1,3) as i;

SELECT * FROM bug6051;
CREATE RULE bug6051_ins AS ON INSERT TO bug6051 DO INSTEAD select a as i from test_data1;

WITH t1 AS ( DELETE FROM bug6051 RETURNING * ) INSERT INTO bug6051 SELECT * FROM t1;

*******
***ERROR:  cannot assign XIDs during a parallel operation
*******
---------------------------------

I debugged it and it did have modifycte in the parsetree after rewrite.
I think if we can properly set the hasModifyingCTE, we can avoid this error by not consider parallel for this.

Thoughts ?

Best regards,
houzj






pgsql-hackers by date:

Previous
From: Amit Langote
Date:
Subject: Re: making update/delete of inheritance trees scale better
Next
From: "tsunakawa.takay@fujitsu.com"
Date:
Subject: RE: POC: postgres_fdw insert batching