[pgpool-general: 1545] Re: Mysterious Load Spikes

[Lonni J Friedman]

Agreed, my pgpool server isn't the same server as the postgres servers.

On Thu, Mar 28, 2013 at 8:59 AM, Quentin Hartman
<qhartman at direwolfdigital.com> wrote:
> Yeah, I just finished reading through those threads and they are
> interesting, but their conclusions seem to be that the root issue is related
> to IO problems, particular with regard to checkpoints. Since I'm not running
> postgres on the pgpool2 boxes at all, I really don't think it's directly
> related to the problem I'm seeing. My pgpool2 boxes have virtually no IO
> load on them at all.
>
> QH
>
> On Thu, Mar 28, 2013 at 9:47 AM, Lonni J Friedman <netllama at gmail.com>
> wrote:
>>
>> Ah, that thread looks interesting, although the original poster seems
>> to have fixed his issues with a kernel upgrade to some 3.4.x kernel.
>> I'm using 3.6.6, so I'm not sure whether its the same thing at all.
>>
>> On Thu, Mar 28, 2013 at 8:40 AM, Tatsuo Ishii <ishii at postgresql.org>
>> wrote:
>> > Maybe this one?
>> >
>> > http://www.postgresql.org/message-id/512D4548.2000502@agliodbs.com
>> > --
>> > Tatsuo Ishii
>> > SRA OSS, Inc. Japan
>> > English: http://www.sraoss.co.jp/index_en.php
>> > Japanese: http://www.sraoss.co.jp
>> >
>> >> Did you have a particular thread in mind?  I scanned the list
>> >> archives, and honestly didn't see anything relevant.  The closest
>> >> thing was this very recent off-topic post:
>> >> http://www.postgresql.org/message-id/51519CE9.1070409@opinioni.net
>> >>
>> >> On Thu, Mar 28, 2013 at 2:27 AM, Will Ferguson
>> >> <wferguson at northplains.com> wrote:
>> >>>
>> >>> Have a look in the postgres performance mailing lists. There's been a
>> >>> lot of chatter recently about performance problems regarding 3.x kernels.
>> >>>
>> >>> Will
>> >>>
>> >>>
>> >>> On 27/03/13 15:24, Quentin Hartman wrote:
>> >>>
>> >>> On Wed, Mar 27, 2013 at 9:15 AM, Lonni J Friedman <netllama at gmail.com>
>> >>> wrote:
>> >>>>
>> >>>> FWIW, I also see roughly the same behavior, although my environment
>> >>>> is
>> >>>> somewhat different than yours.  Previously my pgpool server was
>> >>>> running on RHEL6, however it was migrated to Fedora17 (without
>> >>>> changing the pgpool version).  Before the migration the load on the
>> >>>> server was typically less than 1.00.  Since migrating the load
>> >>>> average
>> >>>> is 30+ (with spikes up to 60 quite often).  Yet there's nothing
>> >>>> obvious causing the load.  I'm certain something in pgpool is causing
>> >>>> this behavior, because the load drops to 0 during the rare times when
>> >>>> we need to stop pgpool for maintenance.
>> >>>
>> >>>
>> >>> Thanks for the quick reply Lonni. I'm glad to hear that it's not a
>> >>> problem limited to Ubuntu. Like you I suspect it has something to do with
>> >>> the 3.x kernel series. I'm going to be doing some experimentation soon to
>> >>> try to isolate and eliminate this problem. I'll update this thread with my
>> >>> findings.
>> >>>
>> >>> If anyone else is interested in coordinating efforts, I'd be glad for
>> >>> the help.
>> >>>
>> >>> QH
>> >>>