Re: Why doesn't autovacuum/analyze run in due time after calling pg_stat_reset?

From: Rihad <grihad(at)gmail(dot)com>
To: Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com>, pgsql-general General <pgsql-general(at)postgresql(dot)org>
Subject: Re: Why doesn't autovacuum/analyze run in due time after calling pg_stat_reset?
Date: 2023-08-21 16:31:08
Message-ID: 329abf6c-2165-f410-717f-c20b3dd0a3d7@gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On 8/21/23 20:17, Adrian Klaver wrote:
> On 8/21/23 09:09, Rihad wrote:
>> On 8/21/23 20:00, Adrian Klaver wrote:
>
>
>> Thanks for the detailed reply, no tables have custom settings.
>>
>> I need to make it clear once again that all autovac/analyze work as
>> expected when n_live_tup matches reality, i.e. when analyze has been
>> run on them since last reset.
>>
>> A way to fix this is to simply analyze the whole database. Before
>> doing that, while n_live_tup starts from basically 0 and grows based
>> on DB activity, these usual calculations of 10-20% table size for
>> vacuum/analyze don't work. They don't trigger autovac for most
>> tables, or do it much much later.
>>
>
> You still have not said or shown whether the other autovacuum settings
> are the default values or not. Assuming they are, then the only other
> explanation I can come up with is that there is a process or processes
> that are creating long running open transactions that prevent
> autovacuum from running on the affected tables.
>

Sorry, they are all as per default, commented out in the config.

There are no long running queries, otherwise they wouldn't be
vacuumed/analyzed in due time after running first manual analyze, which
updates n_live_tup to match reltuples.

In response to

Responses

Browse pgsql-general by date

  From Date Subject
Next Message Tom Lane 2023-08-21 16:45:48 Re: The same prepared query yield "-1" the first six times and then "-1.0"
Previous Message Adrian Klaver 2023-08-21 16:23:06 Re: Will PostgreSQL 16 supports native transparent data encryption ?