From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Itagaki Takahiro <itagaki(dot)takahiro(at)oss(dot)ntt(dot)co(dot)jp> |
Cc: | pgsql-hackers(at)postgresql(dot)org |
Subject: | Re: logging hook for database audit |
Date: | 2009-09-10 14:44:22 |
Message-ID: | 23596.1252593862@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Itagaki Takahiro <itagaki(dot)takahiro(at)oss(dot)ntt(dot)co(dot)jp> writes:
> For the purpose, I'll submit to add "logging hook" in core, rather than
> fixed features like log_min_messages-per-destination.
The more functionality you put inside elog.c, the higher the probability
that it will fail when the going gets tough (out of memory, for instance),
leaving you with no idea of what went wrong. I also dislike the idea
of having such critical behavior dependent on a loadable module, as that
introduces a whole set of failure modes of its own, particularly with
regard to early-startup behavior.
I think the proposal to put a hook here is an end-run to try to avoid
getting a detailed feature proposal past the criticism of pgsql-hackers,
and I don't like that either.
I can see the potential value of per-destination minimum message levels,
so maybe you should just propose that. But this is a bad place for a
hook.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2009-09-10 15:01:32 | Re: pgsql: On Windows, when a file is deleted and another process still has |
Previous Message | Emmanuel Cecchet | 2009-09-10 14:33:48 | COPY enhancements |