Re: add checkpoint stats of snapshot and mapping files of pg_logical dir

From: Bharath Rupireddy <bharath(dot)rupireddyforpostgres(at)gmail(dot)com>
To: Nathan Bossart <nathandbossart(at)gmail(dot)com>
Cc: Cary Huang <cary(dot)huang(at)highgo(dot)ca>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: add checkpoint stats of snapshot and mapping files of pg_logical dir
Date: 2022-03-13 08:24:10
Message-ID: CALj2ACV_Zn7Y8hyLCv2_-9ikGeMXUPtsa3aSgkrGTGsZfLR1hg@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Sat, Mar 12, 2022 at 1:35 AM Nathan Bossart <nathandbossart(at)gmail(dot)com> wrote:
>
> + CheckpointStats.repl_map_cutoff_lsn = cutoff;
>
> Could we set repl_map_cutoff_lsn closer to where it is calculated? Right
> now, it's set at the bottom of the function just before the directory is
> freed. Is there a strong reason to do it there?
>
> + "logical rewrite mapping file(s) removed/synced=" UINT64_FORMAT ", size=%zu bytes, time=%ld.%03d s, cutoff LSN=%X/%X",
>
> Can we split out the "removed" versus "synced" files? Otherwise, I think
> this is basically just telling us how many files are in the mappings
> directory.
>
> + CheckpointStats.repl_snap_cutoff_lsn = cutoff;
>
> I have the same note for this one as I have for repl_map_cutoff_lsn.

Thanks for reviewing this. I agree with all of the above suggestions
and incorporated them in the v2 patch.

Another thing I added in v2 is to not emit snapshot and mapping files
stats in case of restartpoint as logical decoding isn't supported on
standbys, so it doesn't make sense to emit the stats there and cause
server log to grow unnecessarily. Having said that, I added a note
there to change it whenever logical decoding on standbys is supported.

Please review v2.

Regards,
Bharath Rupireddy.

Attachment Content-Type Size
v2-0001-add-checkpoint-stats-of-snapshot-and-mapping-file.patch application/octet-stream 7.1 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Etsuro Fujita 2022-03-13 09:28:23 Re: postgres_fdw: commit remote (sub)transactions in parallel during pre-commit
Previous Message Peter Eisentraut 2022-03-13 06:45:44 Re: logical decoding and replication of sequences