[pgpool-general: 1680] Re: Mysterious Load Spikes

Tatsuo Ishii ishii at postgresql.org
Mon May 6 10:31:37 JST 2013


Done.

http://www.pgpool.net/mediawiki/index.php/FAQ#Every_few_minites_load_of_the_system_which_pgpool-II_running_on_gets_high_as_much_as_5-10._Why.3F
--
Tatsuo Ishii
SRA OSS, Inc. Japan
English: http://www.sraoss.co.jp/index_en.php
Japanese: http://www.sraoss.co.jp

> Seems like a good idea. Had it been there it would have saved me a couple
> days of fruitless firedrills.
> 
> The problem isn't completely solved, I'm still seeing brief spikes of one
> minute load to as high as 7 or 8 (on a dual core machine) on kernel 3.2.
> It's not the constant, chronic load it was on the 3.0 kernel though, so
> that's something. For reference, under the same user load running a 2.6
> series kernel, these boxes never registered a load average higher than 0.5.
> 
> 
> On Sat, Apr 20, 2013 at 3:21 AM, Tatsuo Ishii <ishii at sraoss.co.jp> wrote:
> 
>> > So I was able to upgrade the kernel on my PGPool2 boxes yesterday from
>> 3.0
>> > to 3.2, the current shipping kernel for ubuntu 12.04.2 . Interestingly
>> the
>> > measured cpu utilization has increased by about 100% (still only a 3-5%
>> > though in absolute terms) and the measured load average numbers have
>> > dropped by about half. It's early yet to say for certain that this is
>> > resolved, but it's certainly looking much better. Based on this I would
>> > recommend that anyone deploying PGPool2 avoid 3.0.x revisions of the
>> kernel.
>>
>> Thanks for the information. Maybe we should add this to FAQ?
>>
>> http://www.pgpool.net/mediawiki/index.php/FAQ
>> --
>> Tatsuo Ishii
>> SRA OSS, Inc. Japan
>> English: http://www.sraoss.co.jp/index_en.php
>> Japanese: http://www.sraoss.co.jp
>>


More information about the pgpool-general mailing list