Re: [bug fix] pg_ctl always uses the same event source

From: Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Robert Haas <robertmhaas(at)gmail(dot)com>, MauMau <maumau307(at)gmail(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: [bug fix] pg_ctl always uses the same event source
Date: 2014-01-28 03:36:19
Message-ID: CAA4eK1Jbz1Ze_fej=pA=gZb6z_aALJ7AEgJTmzw2UR7VxfDFZA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Mon, Jan 27, 2014 at 9:01 PM, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> writes:
>> On Fri, Jan 24, 2014 at 9:10 AM, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com> wrote:
>
>> To proceed with the review of this patch, I need to know about
>> whether appending version number or any other constant togli
>
>> Default Event Source name is acceptable or not, else for now
>> we can remove this part of code from patch and handle non-default
>> case where the change will be that pg_ctl will enquire non-default
>> event_source value from server.
>
>> Could you please let me know your views about same?
>
> Unless I'm missing something, this entire thread is a tempest in a teapot,
> because the default event_source value does not matter, because *by
> default we don't log to eventlog*. The presumption is that if the user
> turns on logging to eventlog, it's his responsibility to first make sure
> that event_source is set to something appropriate. And who's to say that
> plain "PostgreSQL" isn't what he wanted, anyway? Even if he's got
> multiple servers on one machine, maybe directing all their logs to the
> same place is okay by him.

I think it's matter of user preference, how exactly he wants the setup
and as currently we don't have any strong reason to change default, so
lets keep it intact.

> Also, those who don't run multiple servers are probably not going to
> thank us for moving their logs around unnecessarily.
>
> In short, I think we should just reject this idea as introducing more
> problems than it solves, and not fully solving even the problem it
> purports to solve.
>
> Possibly there's room for a documentation patch reminding users to
> make sure that event_source is set appropriately before they turn
> on eventlog.

Okay, but in that case also right now pg_ctl doesn't know the value
of event source, so I think thats a clear bug and we should go ahead
and fix it.

As you said, I think we can improve documentation in this regard so
that user will be able to setup event log without any such problems.

As part of this patch we can fix the issue (make pg_ctl aware for event
source name) and improve documentation.

With Regards,
Amit Kapila.
EnterpriseDB: http://www.enterprisedb.com

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Robert Haas 2014-01-28 03:39:25 Re: Performance Improvement by reducing WAL for Update Operation
Previous Message Craig Ringer 2014-01-28 03:34:56 Re: WIP patch (v2) for updatable security barrier views