Re: autovacuum daemon stops doing work after about an hour

From: "Matthew T(dot) O'Connor" <matthew(at)zeut(dot)net>
To: Christopher Browne <cbbrowne(at)acm(dot)org>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: autovacuum daemon stops doing work after about an hour
Date: 2003-12-08 05:27:29
Message-ID: 3FD40BC1.7050906@zeut.net
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers pgsql-performance

Christopher Browne wrote:

>The world rejoiced as mendola(at)bigfoot(dot)com (Gaetano Mendola) wrote:
>
>
>>I think is a good Idea put a fflush after:
>>
>>fprintf(LOGOUTPUT, "[%s] %s\n", timebuffer, logentry);
>>
>>
>
>I thought I had put fflush()es at all the interesting locations...
>
>

I just looked through the code, I think there are fflush()es at all but
one interesting locations. The last log_entry call before sleeping
doesn't have an fflush call after it. I'll submit a patch that adds it.

>Apparently it was an error to not go to the effort of making sure it
>worked well on FreeBSD. (It was on my list, but I never got the Round
>Tuits...) There's an AMD-64 box coming in soon, targeted at FreeBSD,
>so that should change...
>
>
Yeah, FreeBSD testing would have been nice, but I don't have access to
any FreeBSD boxes so.....

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Tatsuo Ishii 2003-12-08 05:36:09 Re: aggregate + view + alias crash on 7.4 stable
Previous Message Tom Lane 2003-12-08 04:31:41 Re: Call for pg_dump testing

Browse pgsql-performance by date

  From Date Subject
Next Message Hartmut Raschick 2003-12-08 14:03:34 TRUNCATE veeeery slow compared to DELETE in 7.4
Previous Message Christopher Browne 2003-12-08 01:17:33 Re: autovacuum daemon stops doing work after about an hour