Обсуждение: Problem with accessing TOAST data in stored procedures

Поиск
Список
Период
Сортировка

Problem with accessing TOAST data in stored procedures

От
Yura Sokolov
Дата:
Good day.

There is reproducible bug posted 10 months ago by Konstantin Knizhnik:
https://www.postgresql.org/message-id/flat/5d335911-fb25-60cd-4aa7-a5bd0954aea0%40postgrespro.ru

I'll copy that message here.

> Stored procedure allows to commit/rollback transaction inside its body.
> Unfortunately it is not always correctly handled.
> This fragment of code reports ERROR: no known snapshots

> CREATE TABLE toasted(id serial primary key, data text);
> INSERT INTO toasted(data) VALUES((SELECT string_agg(random()::text,':') 
> FROM generate_series(1, 1000)));
> INSERT INTO toasted(data) VALUES((SELECT string_agg(random()::text,':') 
> FROM generate_series(1, 1000)));
> DO $$ DECLARE v_r record; BEGIN FOR v_r in SELECT data FROM toasted 
> LOOP INSERT INTO toasted(data) VALUES(v_r.data);COMMIT;END LOOP;END;$$;

> I found out that code responsible for persisting portal correctly 
> extracts TOAST data.
> But pl_pgsql is using prefetch and so takes records form SPI_tuptable, 
> not from stored tuplestore.
> I didn't not find better solution rather than disabling prefetch when 
> loop body contains COMMIT or ROLLBACK statements.
> Unfortunately there is no existed walker for plpgsql statements tree, 
> so I have to add such walker.
> I hope that it will be useful not only for this case.
> But may be there are some other ways to fix this problem...
> Please notice the following bug report which may be also related:
>
https://www.postgresql.org/message-id/flat/20190904105618.j5l6fhyesmprmstf%40alap3.anarazel.de#d93b550f2f0c1eae785f666251432929

Original message contains patch as well.

Andres, I've mentioned you've participated in GIST case in bottom link, 
and it looks like related,
because COMMIT is called there as well.

Regards,
Yura Sokolov



Re: Problem with accessing TOAST data in stored procedures

От
Yura Sokolov
Дата:
Yura Sokolov писал 2021-04-13 17:01:
> Good day.
> 
> There is reproducible bug posted 10 months ago by Konstantin Knizhnik:
> https://www.postgresql.org/message-id/flat/5d335911-fb25-60cd-4aa7-a5bd0954aea0%40postgrespro.ru
> 
> I'll copy that message here.
> 
>> Stored procedure allows to commit/rollback transaction inside its 
>> body.
>> Unfortunately it is not always correctly handled.
>> This fragment of code reports ERROR: no known snapshots
> 
>> CREATE TABLE toasted(id serial primary key, data text);
>> INSERT INTO toasted(data) VALUES((SELECT 
>> string_agg(random()::text,':') FROM generate_series(1, 1000)));
>> INSERT INTO toasted(data) VALUES((SELECT 
>> string_agg(random()::text,':') FROM generate_series(1, 1000)));
>> DO $$ DECLARE v_r record; BEGIN FOR v_r in SELECT data FROM toasted 
>> LOOP INSERT INTO toasted(data) VALUES(v_r.data);COMMIT;END 
>> LOOP;END;$$;
> 
>> I found out that code responsible for persisting portal correctly 
>> extracts TOAST data.
>> But pl_pgsql is using prefetch and so takes records form SPI_tuptable, 
>> not from stored tuplestore.
>> I didn't not find better solution rather than disabling prefetch when 
>> loop body contains COMMIT or ROLLBACK statements.
>> Unfortunately there is no existed walker for plpgsql statements tree, 
>> so I have to add such walker.
>> I hope that it will be useful not only for this case.
>> But may be there are some other ways to fix this problem...
>> Please notice the following bug report which may be also related:
>>
https://www.postgresql.org/message-id/flat/20190904105618.j5l6fhyesmprmstf%40alap3.anarazel.de#d93b550f2f0c1eae785f666251432929
> 
> Original message contains patch as well.
> 
> Andres, I've mentioned you've participated in GIST case in bottom
> link, and it looks like related,
> because COMMIT is called there as well.

Remark: I've checked both with 13.2 and master branch.

Regards,
Yura Sokolov



Re: Problem with accessing TOAST data in stored procedures

От
Michael Paquier
Дата:
On Tue, Apr 13, 2021 at 05:01:56PM +0300, Yura Sokolov wrote:
> Andres, I've mentioned you've participated in GIST case in bottom link, and
> it looks like related,
> because COMMIT is called there as well.

The address you have attempted to use for Andres here is incorrect.
Fixed that just now.
--
Michael

Вложения