Re: Minimal logical decoding on standbys - Mailing list pgsql-hackers

From Andres Freund
Subject Re: Minimal logical decoding on standbys
Date
Msg-id 20230408062626.pklygrtdfgzpujhx@awork3.anarazel.de
Whole thread Raw
In response to Re: Minimal logical decoding on standbys  (Andres Freund <andres@anarazel.de>)
Responses Re: Minimal logical decoding on standbys  (Bertrand Drouvot <bertranddrouvot.pg@gmail.com>)
List pgsql-hackers
Hi,

On 2023-04-07 11:12:26 -0700, Andres Freund wrote:
> +     <row>
> +      <entry role="catalog_table_entry"><para role="column_definition">
> +       <structfield>confl_active_logicalslot</structfield> <type>bigint</type>
> +      </para>
> +      <para>
> +       Number of active logical slots in this database that have been
> +       invalidated because they conflict with recovery (note that inactive ones
> +       are also invalidated but do not increment this counter)
> +      </para></entry>
> +     </row>
>      </tbody>
>     </tgroup>
>    </table>

This seems wrong to me. The counter is not for invalidated slots, it's for
recovery conflict interrupts. If phrased that way, the parenthetical would be
unnecessary.

I think something like
       Number of uses of logical slots in this database that have been
       canceled due to old snapshots or a too low <xref linkend="guc-wal-level"/>
       on the primary

would work and fit with the documentation of the other fields? Reads a bit
stilted, but so do several of the other fields...

Greetings,

Andres Freund



pgsql-hackers by date:

Previous
From: Andres Freund
Date:
Subject: Re: Direct I/O
Next
From: Bertrand Drouvot
Date:
Subject: Re: Minimal logical decoding on standbys