Обсуждение: pg_stat_statements: rows not updated for CREATE TABLE AS SELECTstatements

Hello,
it seems that column "rows" is not updated after CREATE TABLE AS SELECT
statements.

pg13devel (snapshot 2020-03-14)
postgres=# select name,setting from pg_settings where name like 'pg_stat%';
               name               | setting
----------------------------------+---------
 pg_stat_statements.max           | 5000
 pg_stat_statements.save          | on
 pg_stat_statements.track         | all
 pg_stat_statements.track_utility | on
(4 rows)

postgres=# select pg_stat_statements_reset();
 pg_stat_statements_reset
--------------------------

(1 row)


postgres=# create table ctas as select * from pg_class;
SELECT 386
postgres=# select query,calls,rows from pg_stat_statements where query like
'create table ctas%';
                    query                    | calls | rows
---------------------------------------------+-------+------
 create table ctas as select * from pg_class |     1 |    0
(1 row)

after modifying the following line in pg_stat_statements.c

rows = (qc && qc->commandTag == CMDTAG_COPY) ? qc->nprocessed : 0;
into
rows = (qc && (qc->commandTag == CMDTAG_COPY 
                      || qc->commandTag == CMDTAG_SELECT)
           ) ? qc->nprocessed : 0;

column rows seems properly updated.

What do you think about that fix ?
Thanks in advance
Regards
PAscal




--
Sent from: https://www.postgresql-archive.org/PostgreSQL-bugs-f2117394.html

Same remark for syntax

CREATE MATERIALIZED VIEW

as well.

Regards
PAscal



--
Sent from: https://www.postgresql-archive.org/PostgreSQL-bugs-f2117394.html

On 2020/03/16 2:35, legrand legrand wrote:
> Hello,
> it seems that column "rows" is not updated after CREATE TABLE AS SELECT
> statements.
> 
> pg13devel (snapshot 2020-03-14)
> postgres=# select name,setting from pg_settings where name like 'pg_stat%';
>                 name               | setting
> ----------------------------------+---------
>   pg_stat_statements.max           | 5000
>   pg_stat_statements.save          | on
>   pg_stat_statements.track         | all
>   pg_stat_statements.track_utility | on
> (4 rows)
> 
> postgres=# select pg_stat_statements_reset();
>   pg_stat_statements_reset
> --------------------------
> 
> (1 row)
> 
> 
> postgres=# create table ctas as select * from pg_class;
> SELECT 386
> postgres=# select query,calls,rows from pg_stat_statements where query like
> 'create table ctas%';
>                      query                    | calls | rows
> ---------------------------------------------+-------+------
>   create table ctas as select * from pg_class |     1 |    0
> (1 row)

Thanks for the report! Yeah, it seems worth improving this.

> after modifying the following line in pg_stat_statements.c
> 
> rows = (qc && qc->commandTag == CMDTAG_COPY) ? qc->nprocessed : 0;
> into
> rows = (qc && (qc->commandTag == CMDTAG_COPY
>                        || qc->commandTag == CMDTAG_SELECT)
>             ) ? qc->nprocessed : 0;
> 
> column rows seems properly updated.
> 
> What do you think about that fix ?

The utility commands that return CMDTAG_SELECT are
only CREATE TABLE AS SELECT and CREATE MATERIALIZED VIEW?
I'd just like to confirm that there is no case where "rows" must not
be counted when CMDTAG_SELECT is returned.

BTW, "rows" should be updated when FETCH or MOVE is executed
because each command returns or affects the rows?

Regards,

-- 
Fujii Masao
NTT DATA CORPORATION
Advanced Platform Technology Group
Research and Development Headquarters

Thank you for those answers !

> The utility commands that return CMDTAG_SELECT are
> only CREATE TABLE AS SELECT and CREATE MATERIALIZED VIEW?
> I'd just like to confirm that there is no case where "rows" must not
> be counted when CMDTAG_SELECT is returned.

I don't have any in mind ...

> BTW, "rows" should be updated when FETCH or MOVE is executed
> because each command returns or affects the rows?

Yes they should, but they aren't yet (event with CMDTAG_SELECT added)

Note that implicit cursors behave the same way ;o(

postgres=# do $$ declare i integer; begin for i in (select 1 ) loop null;
end loop;end; $$;
DO
postgres=# select calls,query,rows from pg_stat_statements;
 calls |                                      query                                     
| rows
-------+---------------------------------------------------------------------------------+------
     1 | select pg_stat_statements_reset()                                              
|    1
     1 | (select $1 )                                                                   
|    0
     1 | do $$ declare i integer; begin for i in (select 1 ) loop null; end
loop;end; $$ |    0
(3 rows)

Regards
PAscal



--
Sent from: https://www.postgresql-archive.org/PostgreSQL-bugs-f2117394.html

On 2020/03/28 2:43, legrand legrand wrote:
> Thank you for those answers !
> 
>> The utility commands that return CMDTAG_SELECT are
>> only CREATE TABLE AS SELECT and CREATE MATERIALIZED VIEW?
>> I'd just like to confirm that there is no case where "rows" must not
>> be counted when CMDTAG_SELECT is returned.
> 
> I don't have any in mind ...

I found that SELECT INTO also returns CMDTAG_SELECT.

> 
>> BTW, "rows" should be updated when FETCH or MOVE is executed
>> because each command returns or affects the rows?
> 
> Yes they should, but they aren't yet (event with CMDTAG_SELECT added)
> 
> Note that implicit cursors behave the same way ;o(

Thanks for confirming this!

Attached is the patch that makes pgss track the total number of rows
retrieved or affected by CREATE TABLE AS, SELECT INTO,
CREATE MATERIALIZED VIEW and FETCH. I think this is new feature
rather than bug fix, so am planning to add this patch into next CommitFest
for v14. Thought?

Regards,

-- 
Fujii Masao
Advanced Computing Technology Center
Research and Development Headquarters
NTT DATA CORPORATION

Fujii Masao-4 wrote
> Attached is the patch that makes pgss track the total number of rows
> retrieved or affected by CREATE TABLE AS, SELECT INTO,
> CREATE MATERIALIZED VIEW and FETCH. I think this is new feature
> rather than bug fix, so am planning to add this patch into next CommitFest
> for v14. Thought?

Thanks !
maybe v13, if this was considered as a bug that don't need backport ?

Regards
PAscal



--
Sent from: https://www.postgresql-archive.org/PostgreSQL-bugs-f2117394.html

On 2020/04/21 20:37, legrand legrand wrote:
> Fujii Masao-4 wrote
>> Attached is the patch that makes pgss track the total number of rows
>> retrieved or affected by CREATE TABLE AS, SELECT INTO,
>> CREATE MATERIALIZED VIEW and FETCH. I think this is new feature
>> rather than bug fix, so am planning to add this patch into next CommitFest
>> for v14. Thought?
> 
> Thanks !
> maybe v13, if this was considered as a bug that don't need backport ?

Yeah, if many people think this is a bug, we can get it into v13.
But at least for me it looks like an improvement of the capability
of pgss rather than a bug fix. If the document of pgss clearly
explained "pgss tracks the total number of rows affect by even
utility command ...", I think that we can treat this as a bug. But...

Regards,

-- 
Fujii Masao
Advanced Computing Technology Center
Research and Development Headquarters
NTT DATA CORPORATION