Re: [PATCH] Tracking statements entry timestamp in pg_stat_statements

Andrei Zubkov <zubkov@moonset.ru>

From: Andrei Zubkov <zubkov@moonset.ru>

To: Tomas Vondra <tomas.vondra@enterprisedb.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>

Cc: Greg Stark <stark@mit.edu>, Andres Freund <andres@anarazel.de>, Julien Rouhaud <rjuju123@gmail.com>

Date: 2023-01-25T15:46:40Z

Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

Improve quoting in some error messages
- 07e5e6674233 14.0 cited

Attachments

v15-0001-pg_stat_statements-Track-statement-entry-timestamp.patch (text/x-patch) patch v15-0001

Hi,

I've updated this patch for the current master. Also I have some
additional explanations..

On Wed, 2023-01-18 at 17:29 +0100, Tomas Vondra wrote:
> 1) I'm not sure why the patch is adding tests of permissions on the
> pg_stat_statements_reset function?

I've fixed that

> 
> 2) If we want the second timestamp, shouldn't it also cover resets of
> the mean values, not just min/max?

I think that mean values are not a targets for auxiliary resets because
any sampling solution can easily calculate the mean values between
samples without a reset.

> 
> 3) I don't understand why the patch is adding "IS NOT NULL AS t" to
> various places in the regression tests.

This change is necessary in the current version because the
pg_stat_statements_reset() function will return a timestamp of a reset,
needed for sampling solutions to detect resets, perfermed by someone
else.

Regards
-- 
Andrei Zubkov