Re: proposal - log_full_scan - Mailing list pgsql-hackers

From Julien Rouhaud
Subject Re: proposal - log_full_scan
Date
Msg-id 20210417165450.wtblqyuxr7sbfetr@nol
Whole thread Raw
In response to Re: proposal - log_full_scan  (Pavel Stehule <pavel.stehule@gmail.com>)
Responses Re: proposal - log_full_scan  (Pavel Stehule <pavel.stehule@gmail.com>)
List pgsql-hackers
On Sat, Apr 17, 2021 at 05:22:59PM +0200, Pavel Stehule wrote:
> 
> The fullscan of this table needs about 30ms and has 200K rows. So
> decreasing log_min_duration to this value is very risky.
> 
> [...]
> 
> I use  pg_stat_all_tables.seq_scan and I see seq scans there. But I need to
> know the related queries.

Maybe you could use pg_qualstats ([1]) for that?  It will give you the list of
quals (with the underlying queryid) with a tag to specify if they were executed
as an index scan or a sequential scan.  It wouldn't detect queries doing
sequential scan that don't have any qual for the underlying relations, but
those shouldn't be a concern in your use case.

If you setup some sampling, the overhead should be minimal.

[1]: https://github.com/powa-team/pg_qualstats/



pgsql-hackers by date:

Previous
From: Andrew Dunstan
Date:
Subject: Re: multi-install PostgresNode fails with older postgres versions
Next
From: Pavel Stehule
Date:
Subject: Re: proposal - log_full_scan