Re: WIP patch - INSERT-able log statements

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: "FAST PostgreSQL" <fastpgs(at)fast(dot)fujitsu(dot)com(dot)au>
Cc: "Guillaume Smet" <guillaume(dot)smet(at)gmail(dot)com>, "Greg Smith" <gsmith(at)gregsmith(dot)com>, pgsql-patches(at)postgresql(dot)org
Subject: Re: WIP patch - INSERT-able log statements
Date: 2007-02-21 01:08:18
Message-ID: 1851.1172020098@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers pgsql-patches

"FAST PostgreSQL" <fastpgs(at)fast(dot)fujitsu(dot)com(dot)au> writes:
> - The log output will be in COPY format and will include the following
> information, irrespective of the log_line_prefix setting.
> ( timestamp_with_milliseconds, timestamp, username, databasename,
> sessionid, host_and_port, host, proc_id, command_tag, session_start,
> transaction_id, error_severity, SQL_State_Code, statement/error_message);

How exactly are you fitting the message structure
(primary/detail/context lines) into this? It looks like your proposal
loses that structure ...

A smaller problem is that this forces people to incur a gettimeofday
call for every message logged; depending on your hardware that can be a
pretty nasty overhead. Some people might find some of the other columns
not worth their weight, either. Is it worth providing a knob to
determine the set of columns emitted?

regards, tom lane

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Joshua D. Drake 2007-02-21 03:24:52 Re: tsearch in core patch, for inclusion
Previous Message Bruce Momjian 2007-02-21 01:07:11 Re: [HACKERS] timestamp subtraction (was Re: formatting intervals with to_char)

Browse pgsql-patches by date

  From Date Subject
Next Message Bruce Momjian 2007-02-21 01:45:15 Re: [pgsql-patches] Patch to avoid gprofprofilingoverwrites
Previous Message Nikolay Samokhvalov 2007-02-20 23:59:03 --enable-xml instead of --with-libxml?