Re: Additional startup logging

From: Michael Glaesemann <grzm(at)seespotcode(dot)net>
To: Takahiro Itagaki <itagaki(dot)takahiro(at)oss(dot)ntt(dot)co(dot)jp>
Cc: "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov>, <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Additional startup logging
Date: 2010-07-01 03:39:42
Message-ID: 476C109A-ADF9-4549-800A-50BF139D159A@seespotcode.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers


On Jun 30, 2010, at 22:43 , Takahiro Itagaki wrote:

>
> "Kevin Grittner" <Kevin(dot)Grittner(at)wicourts(dot)gov> wrote:
>
>> It seems potentially useful to LOG the version() string in the log
>> file during startup. It might also help to LOG any settings which
>> might result in the loss of committed transactions or in database
>> corruption during startup. (After a crash, the postgresql.conf file
>> might not show the values which were in effect during startup, and
>> it is too late to "show" the values.)
>
> I think such logs depends on purposes, so they should be customizable.
>
> You could write a module, that is registered in 'shared_preload_libraries'
> and logs internal information you want from _PG_init() or shmem_startup_hook.

For long-running systems, you may not have the beginning of the log file. Perhaps a method of dumping the version and/or setting information on demand (or perhaps at the beginning of each log file?): Shouldn't be too hard to put together a function which prints out such information via RAISE even now using PL/pgSQL.

Michael Glaesemann
grzm seespotcode net

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Fujii Masao 2010-07-01 04:09:37 Re: Streaming Replication: Checkpoint_segment and wal_keep_segments on standby
Previous Message Mark Wong 2010-07-01 03:24:18 Re: parallelizing subplan execution (was: explain and PARAM_EXEC)