Re: Table not cleaning up drom dead tuples

From: Glyn Astill <glynastill(at)yahoo(dot)co(dot)uk>
To: Антон Тарабрин <tarabanton(at)gmail(dot)com>, "pgsql-general(at)postgresql(dot)org" <pgsql-general(at)postgresql(dot)org>
Subject: Re: Table not cleaning up drom dead tuples
Date: 2017-03-14 14:34:33
Message-ID: 1007889347.3497957.1489502073448@mail.yahoo.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

> ________________________________

> From: Антон Тарабрин <tarabanton(at)gmail(dot)com>
> To: "pgsql-general(at)postgresql(dot)org" <pgsql-general(at)postgresql(dot)org>
> Sent: Tuesday, 14 March 2017, 14:05
> Subject: Re: [GENERAL] Table not cleaning up drom dead tuples
>
>
> Yep. VACUUM FULL not helping us on OLD table, that are not getting updated and not used in any requests. Bloat is still there
> This is production system, so now we are investigating why it's happening.
>
>
> > Information about problematic tables:
> > https://gist.github.com/tarabanton/edf7f8dc26eb7ec37a9cfa3424493871
> At the link above is some results of VACUUM (ANALYZE, VERBOSE) from source (OLD) table.
>

So what's the output of vacuum full? Or are you saying you can't sustain the exclusive lock vacuum full would require?

Plain vacuum can only reclaim free space at the end of the table, fragmented dead rows can only be marked available for reuse.

Perhaps give us some idea of activity on your database/tables:

select * from pg_stat_user_tables where relname in ('__orders_y2017_m3_to_drop', '__orders_y2017_m2_to_drop');
select * from pg_stat_database;

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Adrian Klaver 2017-03-14 14:57:50 Re: UPDATE ... ON CONFLICT DO NOTHING
Previous Message Adrian Klaver 2017-03-14 14:31:00 Re: Maximum of connections in PG