Re: Add checkpoint and redo LSN to LogCheckpointEnd log message

From: Fujii Masao <masao(dot)fujii(at)oss(dot)nttdata(dot)com>
To: Bharath Rupireddy <bharath(dot)rupireddyforpostgres(at)gmail(dot)com>
Cc: Nathan Bossart <nathandbossart(at)gmail(dot)com>, Stephen Frost <sfrost(at)snowman(dot)net>, Kyotaro Horiguchi <horikyota(dot)ntt(at)gmail(dot)com>, "Bossart, Nathan" <bossartn(at)amazon(dot)com>, Julien Rouhaud <rjuju123(at)gmail(dot)com>, Michael Paquier <michael(at)paquier(dot)xyz>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Add checkpoint and redo LSN to LogCheckpointEnd log message
Date: 2022-02-01 04:19:43
Message-ID: a1de2fd3-4a88-0325-c698-580c044f38b2@oss.nttdata.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 2022/02/01 13:01, Bharath Rupireddy wrote:
> On Tue, Feb 1, 2022 at 9:10 AM Fujii Masao <masao(dot)fujii(at)oss(dot)nttdata(dot)com> wrote:
>> The order of arguments for LSN seems wrong. LSN_FORMAT_ARGS(ControlFile->checkPoint) should be specified ahead of LSN_FORMAT_ARGS(ControlFile->checkPointCopy.redo)?
>
> Thanks. Corrected.

Thanks!

>> Could you tell me why the information for LSN is reported earlierly in the log message? Since ordinally users would be more interested in the information about I/O by checkpoint, the information for LSN should be placed later? Sorry if this was already discussed.
>
> It is useful (for debugging purposes) if the checkpoint end message
> has the checkpoint LSN(end) and REDO LSN(start). It gives more context
> while analyzing checkpoint-related issues. The pg_controldata gives
> the last checkpoint LSN and REDO LSN, but having this info alongside
> the log message helps analyze issues that happened previously, connect
> the dots and identify the root cause.

My previous comment was confusing... Probably I understand why you tried to put this information in checkpoint log message. But I was suggesting to put that information at the end of log message instead of the beginning of it. Because ordinary users would be less interested in this LSN information than other ones like the number of buffers written.

Regards,

--
Fujii Masao
Advanced Computing Technology Center
Research and Development Headquarters
NTT DATA CORPORATION

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Bharath Rupireddy 2022-02-01 04:22:26 Add DBState to pg_control_system function
Previous Message Bharath Rupireddy 2022-02-01 04:01:09 Re: Add checkpoint and redo LSN to LogCheckpointEnd log message