[pgpool-general: 453] Re: Tuning for 1000 concurrent connections (load balancing)

[Alessandro De Noia]

On 15/05/12 13:36, Tatsuo Ishii wrote:
> Sounds like lack of resource, especially memory?
>
>
> I suggest to run vmstat 1 on the host where pgpool is running on and
> to see the swap activities.
> --
> Tatsuo Ishii
> SRA OSS, Inc. Japan
> English: http://www.sraoss.co.jp/index_en.php
> Japanese: http://www.sraoss.co.jp
>

vmstat 1 output

procs -----------memory---------- ---swap-- -----io---- -system-- 
----cpu----
r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id wa
0  0   1476 8205516 316828 13803428    0    0     0     0  522  705  0  
0 100  0
0  0   1476 8205392 316828 13803428    0    0     0     0  576  708  0  
0 100  0
0  0   1476 8205392 316828 13803428    0    0     0     0  621  708  0  
0 100  0
0  0   1476 8205392 316828 13803428    0    0     0     0  576  706  0  
0 100  0
1  0   1476 8205020 316828 13803428    0    0     0   168  654  759  0  
0 100  0
0  0   1476 8205020 316828 13803428    0    0     0     0  559  701  0  
0 100  0

I don't know if it can be useful, I'm using Jmeter to invoke the wsgi 
script 1000 times.


-- 
Alessandro De Noia
GIS Developer

e: alessandro at piemapping.com | w: www.piemapping.com
Caledonia House, 223 - 231 Pentonville Road, London, N1 9NG

Please don't print this e-mail unless you really need to

PIE trading name of MPIE Limited, is a company registered in England and Wales with company number 6991293. Vat number: 983340996